ImageMagick Logo ImageMagick Sprite
Mac OS X

Command-line Options

Below is list of command-line options recognized by the ImageMagick command-linetools. If you want a description of a particular option, click on theoption name in the navigation bar above and you will go right to it. Unlessotherwise noted, each option is recognized by the commands convert, mogrify.

-adaptive-blur radius[xsigma]

Adaptively blur pixels, with decreasing effect near edges.

A Gaussian operator of the given radius and standard deviation (sigma) is used. If sigma is not given itdefaults to 1.

-adaptive-resize geometry

Resize the image using data-dependent triangulation.

See Image Geometry for complete details about the geometry argument. The -adaptive-resizeoption defaults to data-dependent triangulation. Use the -filter to choose a different resampling algorithm.Offsets, if present in the geometry string, are ignored, and the -gravity option has no effect.

-adaptive-sharpen radius[xsigma]

Adaptively sharpen pixels, with increasing effect near edges.

A Gaussian operator of the given radius and standard deviation (sigma) is used. If sigma is not given itdefaults to 1.


Join images into a single multi-image file.

This option is enabled by default. An attempt is made to save all images ofan image sequence into the given output file. However, some formats, such asJPEG and PNG, do not support more than one image per file, and in that caseImageMagick is forced to write each image as a separate file. As such, ifmore than one image needs to be written, the filename given is modified byadding a -scene number before the suffix, in order tomake distinct names for each image.

Use +adjoin to force each image to be written toseparate files, whether or not the file format allows multiple images per file(for example, GIF, MIFF, and TIFF).

Including a C-style integer format string in the output filename willautomagically enable +adjoin and are used to specifywhere the -scene number is placed in the filenames. Thesestrings, such as '%d' or '%03d', are familiar to thosewho have used the standard printf()' C-library function. As anexample, the command

$ convert logo: rose: -morph 15 my%02dmorph.jpg

will create a sequence of 17 images (the two given plus 15 more created by-morph), named: my00morph.jpg, my01morph.jpg,my02morph.jpg, ..., my16morph.jpg.

In summary, ImageMagick tries to write all images to one file, but willsave to multiple files, if any of the following conditions exist...

  1. the output image's file format does not allow multi-image files,
  2. the +adjoin option is given, or
  3. a printf() integer format string (eg: "%d") is present in the output filename.

-affine sx,rx,ry,sy[,tx,ty]

Set the drawing transformation matrix for combined rotating and scaling.

This option sets a transformation matrix, for use by subsequent -draw or -transform options.

The matrix entries are entered as comma-separated numeric values either inquotes or without spaces.

Internally, the transformation matrix has 3x3 elements, but three of themare omitted from the input because they are constant. The new (transformed)coordinates (x', y') of a pixel atposition (x, y) in the originalimage are calculated using the following matrix equation.

affine transformation

The size of the resulting image is that of the smallest rectangle thatcontains the transformed source image. The parameters tx and tysubsequently shift the image pixels so that those that are moved out of theimage area are cut off.

The transformation matrix complies with the left-handed pixel coordinatesystem: positive x and y directionsare rightward and downward, resp.; positive rotation is clockwise.

If the translation coefficients tx and ty are omitted they default to 0,0. Therefore,four parameters suffice for rotation and scaling without translation.

Scaling by the factors sx and sy in the x and y directions,respectively, is accomplished with the following.

See -transform, and the -distort method 'Affineprojection for moreinformation

-affine sx,0,0,sy

Translation by a displacement (tx, ty) is accomplished like so:

-affine 1,0,0,1,tx,ty

Rotate clockwise about the origin (the upper left-hand corner) by an anglea by letting c = cos(a), s= sin(a), and using the following.

-affine c,s,-s,c

The cumulative effect of a sequence of -affinetransformations can be accomplished by instead by a single -affine operation using the matrix equal to the product of the matricesof the individual transformations.

An attempt is made to detect near-singular transformation matrices. If thematrix determinant has a sufficiently small absolute value it is rejected.

-alpha type

Gives control of the alpha/matte channel of an image.

Used to set a flag on an image indicating whether or not to use existing alphachannel data, to create an alpha channel, or to perform other operations on the alpha channel. Choose the argument type from the list below.

type Description
Activate or On Enable the image's transparency channel. Note normally Set should be used instead of this, unless you specifically need to preserve existing (but specifically turned Off) transparency channel.
Associate associate the alpha channel with the image.
Deactivate or Off Disables the image's transparency channel. Does not delete or change the existing data, just turns off the use of that data.
Disassociate disassociate the alpha channel from the image.
Set Activates the alpha/matte channel. If it was previously turned off then it also resets the channel to opaque. If the image already had the alpha channel turned on, it will have no effect.
Opaque Enables the alpha/matte channel and forces it to be fully opaque.
Transparent Activates the alpha/matte channel and forces it to be fully transparent. This effectively creates a fully transparent image the same size as the original and with all its original RGB data still intact, but fully transparent.
Extract Copies the alpha channel values into all the color channels and turns 'Off' the the image's transparency, so as to generate a gray-scale mask of the image's shape. The alpha channel data is left intact just deactivated. This is the inverse of 'Copy'.
Copy Turns 'On' the alpha/matte channel, then copies the gray-scale intensity of the image, into the alpha channel, converting a gray-scale mask into a transparent shaped mask ready to be colored appropriately. The color channels are not modified.
Shape As per 'Copy' but also colors the resulting shape mask with the current background color. That is the RGB color channels is replaced, with appropriate alpha shape.
Remove Composite the image over the background color.
Background Set any fully-transparent pixel to the background color, while leaving it fully-transparent. This can make some image file formats, such as PNG, smaller as the RGB values of transparent pixels are more uniform, and thus can compress better.

Note that while the obsolete +matte operation was thesame as "-alpha Off", the >-matte operation was the same as "-alphaSet" and not "-alpha On".

-annotate degrees text
-annotate XdegreesxYdegrees text
-annotate XdegreesxYdegrees {+-}tx{+-}ty text
-annotate {+-}tx{+-}ty text

Annotate an image with text.

This is a convenience for annotating an image with text. For more precisecontrol over text annotations, use -draw.

The values Xdegrees and Ydegreescontrol the shears applied to the text, while tx and ty are offsets that give the location of the text relative any -gravity setting and defaults to the upper left corner of the image.

Using -annotate degreesor -annotate degreesxdegrees produces an unsheared rotation of the text. Thedirection of the rotation is positive, which means a clockwise rotation if degrees is positive. (This conforms to the usual mathematicalconvention once it is realized that the positive y–direction isconventionally considered to be downward for images.)

The new (transformed) coordinates (x', y') of a pixel at position (x, y) in the image are calculated using the following matrixequation.

annotate transformation

If tx and ty are omitted, they default to 0. This makes thebottom-left of the text becomes the upper-left corner of the image, which isprobably undesirable. Adding a -gravity option in thiscase leads to nice results.

Text is any UTF-8 encoded character sequence. If textis of the form '@mytext.txt', the text is read from the filemytext.txt. Text in a file is taken literally; no embeddedformatting characters are recognized.


Enable/Disable of the rendering of anti-aliasing pixels whendrawing fonts and lines.

By default, objects (e.g. text, lines, polygons, etc.) are antialiased whendrawn. Use +antialias to disable the addition ofantialiasing edge pixels. This will then reduce the number of colors added toan image to just the colors being directly drawn. That is, no mixed colorsare added when drawing such objects.


Join current images vertically or horizontally.

This option creates a single longer image, by joining all the currentimages in sequence top-to-bottom. Use +append tostack images left-to-right.

If they are not of the same width, narrower images are padded with thecurrent -background color setting, and theirposition relative to each other can be controlled by the current -gravity setting.

-attenuate value

Lessen (or intensify) when adding noise to an image.

If unset the value is equivalent to 1.0, or a maximum noise addition

-authenticate password

Decrypt a PDF with a password.

Use this option to supply a password for decryptinga PDF that has been encrypted using Microsoft Crypto API (MSC API). Theencrypting using the MSC API is not supported.

For a different encryption method, see -encipherand -decipher.


Automagically adjust gamma level of image.

This calculates the mean values of an image, then applies a calculated -gamma adjustment so that is the mean color exists in theimage it will get a have a value of 50%.

This means that any solid 'gray' image becomes 50% gray.

This works well for real-life images with little or no extreme dark andlight areas, but tend to fail for images with large amounts of bright sky ordark shadows. It also does not work well for diagrams or cartoon like images.

It uses the -channel setting, (including the'sync' flag for channel synchronization), to determine which colorvalues is used and modified. As the default -channel setting is 'RGB,sync', channels are modifiedtogether by the same gamma value, preserving colors.


Automagically adjust color levels of image.

This is a 'perfect' image normalization operator. It finds the exactminimum and maximum color values in the image and then applies a -level operator to stretch the values to the full range ofvalues.

The operator is not typically used for real-life images, image scans, orJPEG format images, as a single 'out-rider' pixel can set a bad min/max valuesfor the -level operation. On the other hand it is theright operator to use for color stretching gradient images being used togenerate Color lookup tables, distortion maps, or other 'mathematically'defined images.

The operator is very similar to the -normalize, -contrast-stretch, and -linear-stretch operators, but without 'histogram binning' or 'clipping'problems that these operators may have. That is -auto-level is the perfect or ideal version these operators.

It uses the -channel setting, (including thespecial 'sync' flag for channel synchronization), to determinewhich color values are used and modified. As the default +channel setting is 'RGB,sync', the'sync' ensures that the color channels will are modifiedtogether by the same gamma value, preserving colors, and ignoringtransparency.


adjusts an image so that its orientation is suitable for viewing (i.e. top-left orientation).

This operator reads and resets the EXIF image profile setting 'Orientation'and then performs the appropriate 90 degree rotation on the image to orientthe image, for correct viewing.

This EXIF profile setting is usually set using a gravity sensor in digitalcamera, however photos taken directly downward or upward may not have anappropriate value. Also images that have been orientation 'corrected' withoutreseting this setting, may be 'corrected' again resulting in a incorrectresult. If the EXIF profile was previously stripped, the -auto-orient operator will do nothing.


Average a set of images.

An error results if the images are not identically sized.


Display the image centered on a backdrop.[animate, display]

This backdrop covers the entire workstation screen and is useful for hidingother X window activity while viewing the image. The color of the backdrop isspecified as the background color. The color is specified using the formatdescribed under the -fill option.

-background color

Set the background color.

The color is specified using the format described under the -fill option. The default background color (if none isspecified or found in the image) is white.

-bench iterations

Measure performance.

Repeat the entire command for the given number of iterations and report the user-time and elapsed time. For instance,consider the following command and its output. Modify the benchmark with the-duration to run the benchmark for a fixed number of seconds and -concurrentto run the benchmark in parallel (requires the OpenMP feature).

$ convert logo: -resize 1000% -bench 5 logo.pngPerformance[4]: 5i 0.875657ips 6.880u 0:05.710

In this example, 5 iterations were completed at 0.875657 iterations persecond, using 4 threads and 6.88 seconds of the user's allotted time, fora total elapsed time of 5.71 seconds.

-bias value{%}

Add bias when convolving an image.

This option shifts the output of ‑convolve so thatpositive and negative results are relative to the specified bias value.

This is important for non-HDRI compilations of ImageMagick when dealingwith convolutions that contain negative as well as positive values. This isespecially the case with convolutions involving high pass filters or edgedetection. Without an output bias, the negative values are clipped atzero.

When using an ImageMagick with the HDRI compile-time setting, ‑bias is not needed, as ImageMagick is able to store/handle anynegative results without clipping to the color value range(0..QuantumRange).

See the discussion on HDRI implementations of ImageMagick on the page High Dynamic-Range Images. For moreabout HDRI go the ImageMagick Usage pages or thisWikipediaentry.


Use black point compensation.

-black-threshold value{%}

Force to black all pixels below the threshold while leaving all pixels at or above the threshold unchanged.

The threshold value can be given as a percentage or as an absolute integervalue within [0, QuantumRange] corresponding to thedesired ‑channel value. See ‑thresholdfor more details on thresholds and resulting values.

-blend geometry

blend an image into another by the given absolute value or percent.[composite]

Blend will average the images together ('plus') according to thepercentages given and each pixels transparency. If only a single percentagevalue is given it sets the weight of the composite or 'source' image, whilethe background image is weighted by the exact opposite amount. That is a-blend 30% merges 30% of the 'source' image with 70% of the'destination' image. Thus it is equivalent to -blend 30x70%.

-blue-primary x,y

Set the blue chromaticity primary point.

-blue-shift factor

simulate a scene at nighttime in the moonlight. Start with a factor of 1.5

-blur radius
-blur radiusxsigma

Reduce image noise and reduce detail levels.

Convolve the image with a Gaussian or normal distribution using the givenSigma value. The formula is:

gaussian distribution

The Sigma value is the important argument, anddetermines the actual amount of blurring that will take place.

The Radius is only used to determine the size of thearray which will hold the calculated Gaussian distribution. It should be aninteger. If not given, or set to zero, IM will calculate the largest possibleradius that will provide meaningful results for the Gaussian distribution.

The larger the Radius the slower theoperation is. However too small a Radius, and severaliasing effects may result. As a guideline, Radiusshould be at least twice the Sigma value, though threetimes will produce a more accurate result.

This option differs from -gaussian-blur simplyby taking advantage of the separability properties of the distribution. Herewe apply a single-dimensional Gaussian matrix in the horizontal direction,then repeat the process in the vertical direction.

The -virtual-pixel setting will determine howpixels which are outside the image proper are blurred into the final result.

-blur Width[xHeight[+Angle]]

Variably blur an image according to the overlay mapping.[composite]

Each pixel in the overlaid region is replaced with an Elliptical WeightedAverage (EWA) of the source image, scaled according to the grayscalemapping.

The ellipse is weighted with sigma set to the given Width and Height. The Heightdefaults to the Width for a normal circular Gaussianweighting. The Angle will rotate the ellipse fromhorizontal clock-wise.

The -virtual-pixel setting will determine howpixels which are outside the image proper are blurred into the final result.

-border geometry

Surround the image with a border of color.

Set the width and height using the size portion of thegeometry argument. See Image Geometry for complete details about the geometry argument. Offsets areignored.

As of IM 6.7.8-8, the geometry arguments behave as follows:

size General description
value value is added to both left/right and top/bottom
value-xx value-x is added only to left/right and top/bottom are unchanged
xvalue-y value-y is added only to top/bottom and left/right are unchanged
value-xxvalue-y value-x is added to left/right and value-y added to top/bottom
value-xx0 value-x is added only to left/right and top/bottom are unchanged
0xvalue-y value-y is added only to top/bottom and left/right are unchanged
value% value % of width is added to left/right and value % of height is added to top/bottom
value-xx% value-x % of width is added to left/right and to top/bottom
xvalue-y% value-y % of height is added to top/bottom and to left/right
value-x%xvalue-y% value-x % of width is added to left/right and value-y % of height is added to top/bottom
value-x%x0% value-x % of width is added to left/right and top/bottom are unchanged
0%xvalue-y% value-y % of height is added to top/bottom and left/right are unchanged

Set the border color by preceding with the -bordercolor setting.

The -border operation is affected by the current -compose setting and assumes that this is using the default'Over' composition method. It generates an image of the appropriatesize colors by the current -bordercolor beforeoverlaying the original image in the center of this net image. This means thatwith the default compose method of 'Over' any transparent parts maybe replaced by the current -bordercolor setting.

See also the -frame option, which has morefunctionality.

-bordercolor color

Set the border color.

The color is specified using the format described under the -fill option.

The default border color is #DFDFDF, this shade of gray.

-borderwidth geometry

Set the border width.[animate, display]

-brightness-contrast brightness
-brightness-contrast brightness{xcontrast}{%}}

Adjust the brightness and/or contrast of the image.

Brightness and Contrast values apply changes to the input image. They arenot absolute settings. A brightness or contrast value of zero means no change.The range of values is -100 to +100 on each. Positive values increase thebrightness or contrast and negative values decrease the brightness or contrast.To control only contrast, set the brightness=0. To control only brightness,set contrast=0 or just leave it off.

You may also use -channel to control which channels toapply the brightness and/or contrast change. The default is to apply the sametransformation to all channels.

Brightness and Contrast arguments are converted to offset and slope of alinear transform and appliedusing -function polynomial "slope,offset".

The slope varies from 0 at contrast=-100 to almost vertical atcontrast=+100. For brightness=0 and contrast=-100, the result are totallymidgray. For brightness=0 and contrast=+100, the result will approach butnot quite reach a threshold at midgray; that is the linear transformationis a very steep vertical line at mid gray.

Negative slopes, i.e. negating the image, are not possible with thisfunction. All achievable slopes are zero or positive.

The offset varies from -0.5 at brightness=-100 to 0 at brightness=0 to +0.5at brightness=+100. Thus, when contrast=0 and brightness=100, the result istotally white. Similarly, when contrast=0 and brightness=-100, the result istotally black.

As the range of values for the arguments are -100 to +100, adding the '%'symbol is no different than leaving it off.

-cache threshold

(This option has been replaced by the -limit option.)

-canny radius
-canny radiusxsigma{+lower-percent}{+upper-percent}

Canny edge detector uses a multi-stage algorithm to detect a wide range of edges in the image.

The thresholds range from 0 to 100% (e.g. -canny 0x1+10%+30%) with {+lower-percent} < {+upper-percent}. If {+upper-percent} is increased but {+lower-percent} remains the same, lesser edge components will be detected, but their lengths will be the same. If {+lower-percent} is increased but {+upper-percent} is the same, the same number of edge components will be detected but their lengths will be shorter. The default thresholds are shown. The radiusxsigma controls a gaussian blur applied to the input image to reduce noise and smooth the edges.

-caption string

Assign a caption to an image.

This option sets the caption meta-data of an image read in after thisoption has been given. To modify a caption of images already in memory use"-set caption".

The caption can contain special format characters listed in the Format andPrint Image Properties. These attributes are expanded when the captionis finally assigned to the individual images.

If the first character of string is @, the image caption is read from a file titled by theremaining characters in the string. Comments read in from a file are literal;no embedded formatting characters are recognized.

Caption meta-data is not visible on the image itself. To do that use the-annotate or -draw optionsinstead.

For example,

-caption "%m:%f %wx%h" bird.miff

produces an image caption of MIFF:bird.miff 512x480 (assumingthat the image bird.miff has a width of 512 and a height of480.

-cdl filename

color correct with a color decision list.

Here is an example color correction collection:

<?xml version="1.0" encoding="UTF-8"?><ColorCorrectionCollection xmlns="urn:ASC:CDL:v1.2">  <ColorCorrection id="cc06668">    <SOPNode>      <Slope> 0.9 1.2 0.5 </Slope>      <Offset> 0.4 -0.5 0.6 </Offset>      <Power> 1.0 0.8 1.5 </Power>    </SOPNode>    <SATNode>      <Saturation> 0.85 </Saturation>    </SATNode>  </ColorCorrection></ColorCorrectionCollection>

-channel type

Specify those image color channels to which subsequent operators are limited.

Choose from: Red, Green, Blue,Alpha, Gray, Cyan, Magenta,Yellow, Black, Opacity,Index, RGB, RGBA, CMYK, orCMYKA.

The channels above can also be specified as a comma-separated list or can beabbreviated as a concatenation of the letters 'R', 'G','B', 'A', 'O', 'C','M', 'Y', 'K'.For example, to only select the Red and Blue channelsyou can either use

-channel Red,Blue

or you can use the short hand form

-channel RB

All the channels that are present in an image can be specified using thespecial channel type All. Not all operators are 'channel capable',but generally any operators that are generally 'grey-scale' image operators,will understand this setting. See individual operator documentation.

On top of the normal channel selection an extra flag can be specified,'Sync'. This is turned on by default and if set means thatoperators that understand this flag should perform: cross-channelsynchronization of the channels. If not specified, then most grey-scaleoperators will apply their image processing operations to each individualchannel (as specified by the rest of the -channelsetting) completely independently from each other.

For example for operators such as -auto-level and-auto-gamma the color channels are modifiedtogether in exactly the same way so that colors will remain in-sync. Withoutit being set, then each channel is modified separately andindependently, which may produce color distortion.

The -morphology 'Convolve' methodand the -compose mathematical methods, also understandsthe 'Sync' flag to modify the behaviour of pixel colors accordingto the alpha channel (if present). That is to say it will modify the imageprocessing with the understanding that fully-transparent colors should notcontribute to the final result.

Basically, by default, operators work with color channels in synchronous, andtreats transparency as special, unless the -channelsetting is modified so as to remove the effect of the 'Sync' flag.How each operator does this depends on that operators current implementation.Not all operators understands this flag at this time, but that is changing.

To print a complete list of channel types, use -listchannel.

By default, ImageMagick sets -channel to the value'RGBK,sync', which specifies that operators act on all colorchannels except the transparency channel, and that all the color channels areto be modified in exactly the same way, with an understanding of transparency(depending on the operation being applied). The 'plus' form +channel will reset the value back to this default.

Options that are affected by the -channel settinginclude the following.-auto-gamma,-auto-level,-black-threshold,-blur,-clamp,-clut,-combine,-composite (Mathematical compose methods only),-convolve,-contrast-stretch,-evaluate,-function,-fx,-gaussian-blur,-hald-clut,-motion-blur,-morphology,-negate,-normalize,-ordered-dither,-radial-blur,-random-threshold,-separate,-threshold, and-white-threshold.

Warning, some operators behave differently when the +channel default setting is in effect, verses ANY user defined -channel setting (including the equivalent of thedefault). These operators have yet to be made to understand the newer 'Sync'flag.

For example -threshold will by default gray-scalethe image before thresholding, if no -channel settinghas been defined. This is not 'Sync flag controlled, yet.

Also some operators such as -blur, -gaussian-blur, will modify their handling of thecolor channels if the 'alpha' channel is also enabled by -channel. Generally this done to ensure thatfully-transparent colors are treated as being fully-transparent, and thus anyunderlying 'hidden' color has no effect on the final results. Typicallyresulting in 'halo' effects. The newer -morphologyconvolution equivalents however does have a understanding of the 'Sync' flagand will thus handle transparency correctly by default.

As a alpha channel is optional within images, some operators will read thecolor channels of an image as a greyscale alpha mask, when the image has noalpha channel present, and the -channel setting tellsthe operator to apply the operation using alpha channels. The -clut operator is a good example of this.

-charcoal factor

Simulate a charcoal drawing.

-chop geometry

Remove pixels from the interior of an image.

See Image Geometry for complete details about the geometry argument. The widthand height given in the of the sizeportion of the geometry argument give the number ofcolumns and rows to remove. The offset portion ofthe geometry argument is influenced bya -gravity setting, if present.

The -chop option removes entire rows and columns,and moves the remaining corner blocks leftward and upward to close the gaps.

While it can remove internal rows and columns of pixels, it is moretypically used with as -gravity setting and zerooffsets so as to remove a single edge from an image. Compare this to -shave which removes equal numbers of pixels from oppisitesides of the image.

Using -chop effectively undoes the results of a -splice that was given the same geometry and -gravity settings.


set each pixel whose value is below zero to zero and any the pixel whose value is above the quantum range to the quantum range (e.g. 65535) otherwise the pixel value remains unchanged.


Apply the clipping path if one is present.

If a clipping path is present, it is applied to subsequent operations.

For example, in the command

$ convert cockatoo.tif -clip -negate negated.tif

only the pixels within the clipping path are negated.

The -clip feature requires SVG support. If the SVGdelegate library is not present, the option is ignored.

Use +clip to disable clipping for subsequent operations.


Clip the image as defined by this mask.

Use the alpha channel of the current image as a mask. Any areas that iswhite is not modified by any of the 'image processing operators' that follow,until the mask is removed. Pixels in the black areas of the clip mask aremodified per the requirements of the operator.

In some ways this is similar to (though not the same) as defininga rectangular -region, or using the negative of themask (third) image in a three image -composite,operation.

Use +clip-mask to disable clipping for subsequent operations.

-clip-path id

Clip along a named path from the 8BIM profile.

This is identical to -clip except choose a specific clip path in the event the image has more than one path available.

Use +clip-path to disable clipping for subsequent operations.

-clone index(s)

make a clone of an image (or images).

Inside parenthesis (where the operator is normally used) it will make aclone of the images from the last 'pushed' image sequence, and adds them tothe end of the current image sequence. Outside parenthesis(not recommended) it clones the images from the current image sequence.

Specify the image by its index in the sequence. The first image is index0. Negative indexes are relative to the end of the sequence; forexample, −1represents the last image of the sequence. Specify a range of images with adash (e.g. 0−4). Separate multiple indexes with commas but nospaces (e.g. 0,2,5). A value of '0−−1 willeffectively clone all the images.

The +clone will simply make a copy of the last imagein the image sequence, and is thus equivalent to using a argument of'−1'.


Replace the channel values in the first image using eachcorresponding channel in the second image as a colorlookup table.

The second (LUT) image is ordinarily a gradient image containing thehistogram mapping of how each channel should be modified. Typically it is aeither a single row or column image of replacement color values. If largerthan a single row or column, values are taken from a diagonal line fromtop-left to bottom-right corners.

The lookup is further controlled by the -interpolate setting, which is especially handy for anLUT which is not the full length needed by the ImageMagick installed Quality(Q) level. Good settings for this are the 'bilinear' and'bicubic' interpolation settings, which give smooth colorgradients, and the 'integer' setting for a direct, unsmoothedlookup of color values.

This operator is especially suited to replacing a grayscale image with aspecific color gradient from the CLUT image.

Only the channel values defined by the -channelsetting will have their values replaced. In particular, since the default -channel setting is RGB, this means thattransparency (alpha/matte channel) is not affected, unless the -channel setting is modified. When the alpha channel isset, it is treated by the -clut operator in the same wayas the other channels, implying that alpha/matte values are replaced using thealpha/matte values of the original image.

If either the image being modified, or the lookup image, contains notransparency (i.e. -alpha is turned 'off') but the -channel setting includes alpha replacement, then it isassumed that image represents a gray-scale gradient which is used for thereplacement alpha values. That is you can use a gray-scale CLUT image toadjust a existing images alpha channel, or you can color a gray-scale imageusing colors form CLUT containing the desired colors, including transparency.

See also -hald-clut which replaces colorsaccording to the lookup of the full color RGB value from a 2D representationof a 3D color cube.


Fully define the look of each frame of an GIF animation sequence, to form a 'film strip' animation.

Overlay each image in an image sequence according toits -dispose meta-data, to reproduce the look ofan animation at each point in the animation sequence. All images should bethe same size, and are assigned appropriate GIF disposal settings for theanimation to continue working as expected as a GIF animation. Such framesare more easily viewed and processed than the highly optimized GIF overlayimages.

The animation can be re-optimized after processing usingthe -layers method 'optimize', althoughthere is no guarantee that the restored GIF animation optimization isbetter than the original.

-colorize value

Colorize the image by an amount specified by value using the color specified by the most recent -fill setting.

Specify the amount of colorization as a percentage. Separate colorizationvalues can be applied to the red, green, and blue channels of the image witha comma-delimited list of colorizationvalues (e.g., -colorize 0,0,50).

-colormap type

Define the colormap type.[animate, display]

The type can be shared or private.

This option only applies when the default X server visualis PseudoColor or GrayScale. Referto -visual for more details. By default,a shared colormap is allocated. The image shares colors withother X clients. Some image colors could be approximated,therefore your image may look very different than intended.If private is chosen, the image colors appear exactlyas they are defined. However, other clients may go technicolorwhen the image colormap is installed.

-colors value

Set the preferred number of colors in the image.

The actual number of colors in the image may be less than your request,but never more. Note that this a color reduction option. Images with fewerunique colors than specified by value will have anyduplicate or unused colors removed. The ordering of an existing colorpalette may be altered. When converting an image from color to grayscale,it is more efficient to convert the image to the gray colorspace beforereducing the number of colors. Refer tothe color reduction algorithm for more details.

-color-matrix matrix

apply color correction to the image.

This option permits saturation changes, hue rotation, luminance to alpha,and various other effects. Although variable-sized transformation matricescan be used, typically one uses a 5x5 matrix for an RGBA image and a 6x6for CMYKA (or RGBA with offsets). The matrix is similar to those used byAdobe Flash except offsets are in column 6 rather than 5 (in support ofCMYKA images) and offsets are normalized (divide Flash offset by 255).

As an example, to add contrast to an image with offsets, try this command:

   convert kittens.jpg -color-matrix \     " 1.5 0.0 0.0 0.0, 0.0, -0.157 \       0.0 1.5 0.0 0.0, 0.0, -0.157 \       0.0 0.0 1.5 0.0, 0.0, -0.157 \       0.0 0.0 0.0 1.0, 0.0,  0.0 \       0.0 0.0 0.0 0.0, 1.0,  0.0 \       0.0 0.0 0.0 0.0, 0.0,  1.0" kittens.png

-colorspace value

Set the image colorspace.

Choices are:

   CMY          CMYK         Gray         HCL   HCLp         HSB          HSI          HSL   HSV          HWB          Lab          LCHab   LCHuv        LMS          Log          Luv   OHTA         Rec601YCbCr  Rec709YCbCr  RGB   scRGB        sRGB         Transparent  xyY   XYZ          YCbCr        YCC          YDbDr   YIQ          YPbPr        YUV

To print a complete list of colorspaces, use -list colorspace.

For a more accurate color conversion to or from the linear RGB, CMYK, or grayscale colorspaces, use the -profile option. Note, ImageMagick assumes the sRGB colorspace if the image format does not indicate otherwise. For colorspace conversion, the gamma function is first removed to produce linear RGB.

Conversion of RGB to Other Color Spaces
CMYK — starts with CMY from above
Gray = 0.298839*R+0.586811*G+0.114350*B
HSB — Hue, Saturation, Brightness; like a cone peak downward
H=angle around perimeter (0 to 360 deg); H=0 is red; increasing angles toward green
S=distance from axis outward
B=distance along axis from bottom upward; B=max(R,G,B); intensity-like
HSL — Hue, Saturation, Lightness; like a double cone end-to-end with peaks at very top and bottom
H=angle around perimeter (0 to 360 deg); H=0 is red; increasing angles toward green
S=distance from axis outward
L=distance along axis from bottom upward; L=0.5*max(R,G,B) + 0.5*min(R,G,B); intensity-like
HWB — Hue, Whiteness, Blackness
Hue (complicated equation)
Whiteness (complicated equation)
Blackness (complicated equation)
L (complicated equation relating X,Y,Z)
A (complicated equation relating X,Y,Z)
B (complicated equation relating X,Y,Z)
I1 (complicated equation involving logarithm of R)
I2 (complicated equation involving logarithm of G)
I3 (complicated equation involving logarithm of B)
OHTA — approximates principal components transformation
I1=0.33333*R+0.33334*G+0.33333*B; intensity-like
Gray = 0.298839*R+0.586811*G+0.114350*B
Y=0.2988390*R+0.5868110*G+0.1143500*B; intensity-like
Y=0.212656*R+0.715158*G+0.072186*B; intensity-like
if R ≤ .0.0031308 then Rs=R/12.92 else Rs=1.055 R ^ (1.0 / 2.4) ? 0.055
if G ≤ .0.0031308 then Gs=B/12.92 else Gs=1.055 R ^ (1.0 / 2.4) ? 0.055
if B ≤ .0.0031308 then Bs=B/12.92 else Bs=1.055 R ^ (1.0 / 2.4) ? 0.055
Y=(0.298839*R+0.586811*G+0.114350*B) (with complicated scaling); intensity-like
C1=(−0.298839*R−0.586811*G+0.88600*B) (with complicated scaling)
C2=(0.70100*R−0.586811*G−0.114350*B) (with complicated scaling)
Y=0.2988390*R+0.5868110*G+0.1143500*B; intensity-like
Y=0.298839*R+0.586811*G+0.114350*B; intensity-like
Y=0.2988390*R+0.5868110*G+0.1143500*B; intensity-like
Y=0.298839*R+0.586811*G+0.114350*B; intensity-like

Note the scRGB colorspace requires HDRI support otherwise it behaves just like linear RGB.


Combine one or more images into a single image.

The channels (previously set by -channel) of thecombined image are taken from the grayscale values of each image in thesequence, in order. For the default -channel setting of RGB, thismeans the first image is assigned to the Red channel, the secondto the Green channel, the third to the Blue.

This option can be thought of as the inverse to -separate, so long as the channel settings are the same.Thus, in the following example, the final image should be a copy of theoriginal.

$ convert original.png -channel RGB -separate sepimage.png $ convert sepimage-0.png sepimage-1.png sepimage-2.png -channel RGB \
-combine imagecopy.png

-comment string

Embed a comment in an image.

This option sets the comment meta-data of an image read in after thisoption has been given. To modify a comment of images already in memory use"-set comment".

The comment can contain special format characters listed in the Format andPrint Image Properties. These attributes are expanded when the commentis finally assigned to the individual images.

If the first character of string is @, the image comment is read from a file titled by theremaining characters in the string. Comments read in from a file are literal;no embedded formatting characters are recognized.

Comment meta-data are not visible on the image itself. To do that use the-annotate or -draw optionsinstead.

For example,

-comment "%m:%f %wx%h" bird.miff

produces an image comment of MIFF:bird.miff 512x480 (assumingthat the image bird.miff has a width of 512 and a height of480.


mathematically and visually annotate the difference between an image and its reconstruction

This is a convert version of "compare" for two same sized images. The syntax is as follows, but other metrics are allowed.

$ convert image.png reference.png -metric RMSE -compare \

To get the metric value use the string format "%[distortion]".

$ convert image.png reference.png -metric RMSE -compare -format \
"%[distortion]" info:

-complex operator

perform complex mathematics on an image sequence
Choose from these operators:
  add  conjugate  divide  magnuitude-phase  multiply  real-imaginary  subtract

Optionally specify the divide operator SNR with -define complex:snr=float.

-compose operator

Set the type of image composition.

See Alpha Compositing fora detailed discussion of alpha compositing.

This setting effects image processing operators that merge two (or more)images together in some way. This includes the operators,-compare,-composite,-layers composite,-flatten,-mosaic,-layers merge,-border,-frame,and -extent.

It is also one of the primary options for the "composite"command.


Perform alpha composition on two images and an optional mask

Take the first image 'destination' and overlay the second 'source' imageaccording to the current -compose setting. The locationof the 'source' or 'overlay' image is controlled according to -geometry, and -geometrysettings.

If a third image is given this is treated as a gray-scale blending 'mask' imagerelative to the first 'destination' image. This mask is blended with thesource image. However for the 'displace' compose method, themask is used to provide a separate Y-displacement image instead.

If a -compose method requires extra numericalarguments or flags these can be provided by setting the -set 'option:compose:args'appropriately for the compose method.

Some -compose methods can modify the 'destination'image outside the overlay area. You can disable this by setting the special -set 'option:compose:outside-overlay'to 'false'.

-compress type

Use pixel compression specified by type when writing the image.

Choices are: None, BZip, Fax, Group4, JPEG, JPEG2000, Lossless, LZW, RLE or Zip.

To print a complete list of compression types, use -listcompress.

Specify +compress to store the binary image in anuncompressed format. The default is the compression type of the specifiedimage file.

If LZW compression is specified but LZW compression has not beenenabled, the image data is written in an uncompressed LZW format that can beread by LZW decoders. This may result in larger-than-expected GIF files.

Lossless refers to lossless JPEG, which is only available if theJPEG library has been patched to support it. Use of lossless JPEG is generallynot recommended.

When writing an ICO file, you may request that the images be encoded inPNG format, by specifying Zip compression.

When writing a JNG file, specify Zip compression to request thatthe alpha channel be encoded in PNG "IDAT" format, or JPEGto request that it be encoded in JPG "JDAA" format.

Use the -quality option to set the compression levelto be used by JPEG, PNG, MIFF, and MPEG encoders.Use the -sampling-factor option to set thesampling factor to be used by JPEG, MPEG, and YUV encoders for down-samplingthe chroma channels.

-connected-components connectivity

connected-components uniquely labeled, choose from 4 or 8 way connectivity.

Use -define connected-components:verbose=true to output statistics associated with each unique label.


Enhance or reduce the image contrast.

This option enhances the intensity differences between the lighter anddarker elements of the image. Use -contrast to enhancethe image or +contrast to reduce the imagecontrast.

For a more pronounced effect you can repeat the option:

$ convert rose: -contrast -contrast rose_c2.png

-contrast-stretch black-point
-contrast-stretch black-point{xwhite-point}{%}}

Increase the contrast in an image by stretching the range of intensity values.

While performing the stretch, black-out at most black-point pixels and white-out at most white-point pixels. Or, if percent is used, black-out at mostblack-point % pixels and white-out at most white-point % pixels.

Prior to ImageMagick 6.4.7-0, -contrast-stretch will black-out at most black-point pixels and white-out at most total pixelsminus white-point pixels. Or, if percent is used, black-out at most black-point % pixels and white-out at most 100% minus white-point % pixels.

Note that -contrast-stretch 0 will modify the image such thatthe image's min and max values are stretched to 0 and QuantumRange, respectively, without any loss of data due to burn-out orclipping at either end. This is not the same as -normalize, which is equivalent to -contrast-stretch 0.15x0.05% (orprior to ImageMagick 6.4.7-0, -contrast-stretch 2%x99%).

Internally operator works by creating a histogram bin, and then uses thatbin to modify the image. As such some colors may be merged together when theyoriginally fell into the same 'bin'.

All the channels are normalized in concert by the came amount so as topreserve color integrity, when the default +channelsetting is in use. Specifying any other -channelsetting will normalize the RGB channels independently.

See also -auto-level for a 'perfect'normalization of mathematical images.

This operator is under review for re-development.

-convolve kernel

Convolve an image with a user-supplied convolution kernel.

The kernel is a matrix specified asa comma-separated list of integers (with no spaces), ordered left-to right,starting with the top row. Presently, only odd-dimensioned kernels aresupported, and therefore the number of entries in the specified kernel must be 32=9, 52=25,72=49, etc.

Note that the ‑convolve operator supports the ‑bias setting. This option shifts the convolution so thatpositive and negative results are relative to a user-specified bias value.This is important for non-HDRI compilations of ImageMagick when dealing withconvolutions that contain negative as well as positive values. This isespecially the case with convolutions involving high pass filters or edgedetection. Without an output bias, the negative values is clipped at zero.

When using an ImageMagick with the HDRI compile-time setting, ‑bias is not needed, as ImageMagick is able to store/handle anynegative results without clipping to the color value range (0..QuantumRange).See the discussion on HDRI implementations of ImageMagick on the page HighDynamic-Range Images. For more about HDRI go the ImageMagick Usage pages or thisWikipediaentry.

-crop geometry{@}{!}

Cut out one or more rectangular regions of the image.

See Image Geometry for complete details about the geometry argument.

The width and height of the geometry argument give the size of the image that remainsafter cropping, and x and y in theoffset (if present) gives the location of the top leftcorner of the cropped image with respect to the original image. To specify theamount to be removed, use -shave instead.

If the x and y offsets arepresent, a single image is generated, consisting of the pixels from thecropping region. The offsets specify the location of the upper left corner ofthe cropping region measured downward and rightward with respect to the upperleft corner of the image. If the -gravity option ispresent with NorthEast, East, or SouthEastgravity, it gives the distance leftward from the right edge of the image tothe right edge of the cropping region. Similarly, if the -gravity option is present with SouthWest,South, or SouthEast gravity, the distance is measuredupward between the bottom edges.

If the x and y offsets areomitted, a set of tiles of the specified geometry, covering the entire inputimage, is generated. The rightmost tiles and the bottom tiles are smaller ifthe specified geometry extends beyond the dimensions of the input image.

You can add the @ to the geometry argument to equally divide the image into the number of tiles generated.

By adding a exclamation character flag to the geometry argument, thecropped images virtual canvas page size and offset is set as if thegeometry argument was a viewport or window. This means the canvas page sizeis set to exactly the same size you specified, the image offset setrelative top left corner of the region cropped.

If the cropped image 'missed' the actual image on its virtual canvas, aspecial single pixel transparent 'missed' image is returned, and a 'cropmissed' warning given.

It might be necessary to +repage the image prior tocropping the image to ensure the crop coordinate frame is relocated to theupper-left corner of the visible image.Similarly you may want to use +repage after cropping toremove the page offset that will be left behind. This is especially true whenyou are going to write to an image format such as PNG that supports an imageoffset.

-cycle amount

displace image colormap by amount.

Amount defines the number of positions eachcolormap entry is shifted.

-debug events

enable debug printout.

The events parameter specifies which events are to be logged. Itcan be either None, All, Trace, ora comma-separated list consisting of one or more of the following domains:Accelerate, Annotate, Blob, Cache,Coder, Configure, Deprecate,Exception, Locale, Render,Resource, Security, TemporaryFile,Transform, X11, or User.

For example, to log cache and blob events, use.

$ convert -debug "Cache,Blob" rose: rose.png

The User domain is normally empty, but developers can log userevents in their private copy of ImageMagick.

To print the complete list of debug methods, use -listdebug.

Use the -log option to specify the format for debuggingoutput.

Use +debug to turn off all logging.

Debugging may also be set using the MAGICK_DEBUG environment variable. The allowed values for the MAGICK_DEBUGenvironment variable are the same as for the -debugoption.

-decipher filename

Decipher and restore pixels that were previously transformed by-encipher.

Get the passphrase from the file specified by filename.

For more information, see the webpage, ImageMagick: Encipher orDecipher an Image.


find areas that has changed between images

Given a sequence of images all the same size, such as produced by -coalesce, replace the second and later images, witha smaller image of just the area that changed relative to the previous image.

The resulting sequence of images can be used to optimize an animationsequence, though will not work correctly for GIF animations when parts of theanimation can go from opaque to transparent.

This option is actually equivalent to the -layersmethod 'compare-any'.

-define key{=value}...

add specific global settings generally used to controlcoders and image processing operations.

This option creates one or more definitions for coders and decoders to usewhile reading and writing image data. Definitions are generally used tocontrol image file format coder modules, and image processing operations,beyond what is provided by normal means. Defined settings are listed in -verbose information ("info:" output format)as "Artifacts".

If value is missing for a definition, an empty-valueddefinition of a flag is created with that name. This used to control on/offoptions. Use +define key to remove definitionspreviously created. Use +define "*" to remove allexisting definitions.

The same 'artifact' settings can also be defined using the -set "option:key" "value" option, which also allows the use of Format and Print ImageProperties in the defined value.

The option and key are case-independent (they areconverted to lowercase for use within the decoders) while the valueis case-dependent.

Such settings are global in scope, and affect all images and operations.

The following definitions are just some of the artifacts that areavailable:

valid values are bmp2, bmp3, and bmp4. This option can be useful when the method of prepending "BMP2:" to the output filename is inconvenient or is not available, such as when using the mogrify utility.
Set the divide SNR constant-complex
Sets certain compose argument values when using convert ... -compose ... -composite. See Image Composition
convolve:scale={kernel_scale}[!^] [,{origin_addition}] [%]
Defines the kernel scaling. The special flag ! automatically scales to full dynamic range. The ! flag can be used in combination with a factor or percent. The factor or percent is then applied after the automatic scaling. An example is 50%!. This produces a result 50% darker than full dynamic range scaling. The ^ flag assures the kernel is 'zero-summing', for example when some values are positive and some are negative as in edge detection kernels. The origin addition adds that value to the center pixel of the kernel. This produces and effect that is like adding the image that many times to the result of the filtered image. The typical value is 1 so that the original image is added to the result of the convolution. The default is 0.
Outputs (to 'standard error') all the information about a specified -morphology convolve kernel.
Sets the display range to the minimum and maximum pixel values for the DCM image format.
Enables the dds cluster-fit.
Sets the dds compression.
Sets the dds number of mipmaps.
Enables the dds alpha weighting.
Specifies direct conversion from Postscript to PDF.
Sets the output scaling factor for use with -distort
Sets the viewport for use with -distort
Specifies the layout engine for the DOT image format (e.g. neato).
Set a filter option for use with -resize. See -filter for details.
Sets the location for the FFT/IFT normalization as use by +-fft and +-ift. The default is forward.
Automatically stores multiple sizes when writing an ico image (requires a 256x256 input image).
Sets the maximum number of quality layers to decode. Same for JPT, JC2, and J2K
Sets the number of resolutions to encode.Same for JPT, JC2, and J2K
choose from LRCP, RLCP, RPCL, PCRL or CPRL. Same for JPT, JC2, and J2K
Sets the quality layer PSNR, given in dB. The order is from left to right in ascending order. The default is a single lossless quality layer. Same for JPT, JC2, and J2K
Specify the compression factor to use while writing JPEG-2000 files. The compression factor is the reciprocal of the compression ratio. The valid range is 0.0 to 1.0, with 1.0 indicating lossless compression. If defined, this value overrides the -quality setting. A quality setting of 75 results in a rate value of 0.06641. Same for JPT, JC2, and J2K
Sets the number of highest resolution levels to be discarded.Same for JPT, JC2, and J2K
Set the desired number of colors and let the JPEG encoder do the quantizing.
Choose from default, fastest, float, ifast, and islow.
Restrict the maximum JPEG file size, for example -define jpeg:extent=400KB. The JPEG encoder will search for the highest compression quality level that results in an output file that does not exceed the value. The -quality option is ignored if it is also present.
Set the size hint of a JPEG image, for example, -define jpeg:size=128x128. It is most useful for increasing performance and reducing the memory requirements when reducing the size of a large JPEG image.
includes features in verbose information
includes image moments in verbose information
turn playback caching off for streaming MNG.
Specifies how to merge results generated by multiple-morphology kernel. The default is none. One typical value is 'lighten' as used, for example, with the sobel edge kernels.
Outputs (to 'standard error') all the information about a generated -morphology kernel.
geometry specifies the scaling dimensions for resizing when the PDF is being read. The geometry is either WxH{%} or page size. No offsets are allowed. (introduced in IM 6.8.8-8)
desired bit-depth and color-type for PNG output. You can force the PNG encoder to use a different bit-depth and color-type than it would have normally selected, but only if this does not cause any loss of image quality. Any attempt to reduce image quality is treated as an error and no PNG file is written. E.g., if you have a 1-bit black-and-white image, you can use these "defines" to cause it to be written as an 8-bit grayscale, indexed, or even a 64-bit RGBA. But if you have a 16-million color image, you cannot force it to be written as a grayscale or indexed PNG. If you wish to do this, you must use the appropriate -depth, -colors, or -type directives to reduce the image quality prior to using the PNG encoder. Note that in indexed PNG files, "bit-depth" refers to the number of bits per index, which can be 1, 2, 4, or 8. In such files, the color samples always have 8-bit depth.
valid values are 0 through 9. 0-4 are the corresponding PNG filters, 5 means adaptive filtering except for images with a colormap, 6 means adaptive filtering for all images, 7 means MNG "loco" compression, 8 means Z_RLE strategy with adaptive filtering, and 9 means Z_RLE strategy with no filtering.
valid values are 0 through 9, with 0 providing the least but fastest compression and 9 usually providing the best and always the slowest.
valid values are 0 through 4, meaning default, filtered, huffman_only, rle, and fixed ZLIB compression strategy. If you are using an old zlib that does not support Z_RLE (before 1.2.0) or Z_FIXED (before, values 3 and 4, respectively, will use the zlib default strategy instead.
valid values are png8, png24, png32, png48, png64, and png00. This property can be useful for specifying the specific PNG format to be used, when the usual method of prepending the format name to the output filename is inconvenient, such as when writing a PNG-encoded ICO file or when using mogrify. Value = png8 reduces the number of colors to 256, only one of which may be fully transparent, if necessary. The other values do not force any reduction of quality; it is an error to request a format that cannot represent the image data without loss (except that it is allowed to reduce the bit-depth from 16 to 8 for all formats). Value = png24 and png48 allow transparency, only if a single color is fully transparent and that color does not also appear in an opaque pixel; such transparency is written in a PNG tRNS chunk. Value = png00 causes the image to inherit its color-type and bit-depth from the input image, if the input was also a PNG.
ancillary chunks to be excluded from or included in PNG output.

The value can be the name of a PNG chunk-type such as bKGD, a comma-separated list of chunk-names (which can include the word date, the word all, or the word none). Although PNG chunk-names are case-dependent, you can use all lowercase names if you prefer.

The "include-chunk" and "exclude-chunk" lists only affect the behavior of the PNG encoder and have no effect on the PNG decoder.

As a special case, if the sRGB chunk is excluded and the gAMA chunk is included, the gAMA chunk will only be written if gamma is not 1/2.2, since most decoders assume sRGB and gamma=1/2.2 when no colorspace information is included in the PNG file. Because the list is processed from left to right, you can achieve this with a single define:

   -define png:include-chunk=none,gAMA

As a special case, if the sRGB chunk is not excluded and the PNG encoder recognizes that the image contains the sRGB ICC profile, the PNG encoder will write the sRGB chunk instead of the entire ICC profile. To force the PNG encoder to write the sRGB profile as an iCCP chunk in the output PNG instead of the sRGB chunk, exclude the sRGB chunk.

The critical PNG chunks IHDR, PLTE, IDAT, and IEND cannot be excluded. Any such entries appearing in the list will be ignored.

If the ancillary PNG tRNS chunk is excluded and the image has transparency, the PNG colortype is forced to be 4 or 6 (GRAY_ALPHA or RGBA). If the image is not transparent, then the tRNS chunk isn't written anyhow, and there is no effect on the PNG colortype of the output image.

The -strip option does the equivalent of the following for PNG output:

   -define png:exclude-chunk=EXIF,iCCP,iTXt,sRGB,tEXt,zCCP,zTXt,date

The default behavior is to include all known PNG ancillary chunks plus ImageMagick's private vpAg ("virtual page") chunk, and to exclude all PNG chunks that are unknown to ImageMagick, regardless of their PNG "copy-safe" status as described in the PNG specification.

Any chunk names that are not known to ImageMagick are ignored if they appear in either the "include-chunk" or "exclude-chunk" list. The ancillary chunks currently known to ImageMagick are bKGD, cHRM, gAMA, iCCP, oFFs, pHYs, sRGB, tEXt, tRNS, vpAg, and zTXt.

You can also put date in the list to include or exclude the "Date:create" and "Date:modify" text chunks that ImageMagick normally inserts in the output PNG.

Use the existing image->colormap. Normally the PNG encoder will try to optimize the palette, eliminating unused entries and putting the transparent colors first. If this flag is set, that behavior is suppressed.
By default, the PNG decoder and encoder examine any ICC profile that is present, either from an iCCP chunk in the PNG input or supplied via an option, and if the profile is recognized to be the sRGB profile, converts it to the sRGB chunk. You can use -define png:preserve-iCCP to prevent this from happening; in such cases the iCCP chunk will be read or written and no sRGB chunk will be written. There are some ICC profiles that claim to be sRGB but have various errors that cause them to be rejected by libpng16; such profiles are recognized anyhow and converted to the sRGB chunk, but are rejected if the -define png:preserve-iCCP is present. Note that not all "sRGB" ICC profiles are recognized yet; we will add them to the list as we encounter them.
The PNG specification requires that any multi-byte integers be stored in network byte order (MSB-LSB endian). This option allows you to fix any invalid PNG files that have 16-bit samples stored incorrectly in little-endian order (LSB-MSB). The "-define png:swap-bytes" option must appear before the input filename on the commandline. The swapping is done during the libpng decoding operation.
Skip the named profile[s] when reading the image. Use skip="*" to skip all named profiles in the image. Many named profiles exist, including ICC, EXIF, APP1, IPTC, XMP, and others.
If the ps:imagemask flag is defined, the PS3 and EPS3 coders will create Postscript files that render bilevel images with the Postscript imagemask operator instead of the image operator.
Set the type to floating-point to specify a floating-point format for raw files (e.g. GRAY:) or for MIFF and TIFF images in HDRI mode to preserve negative values. If -depth 16 is included, the result is a single precision floating point format. If -depth 32 is included, the result is double precision floating point format.
Set the photometric-interpretation of an image (typically for TIFF image file format) to either min-is-black (default) or min-is-white.
Location of the sampling point within the sub-region being sampled, expressed as percentages (see -sample).
Outputs (to 'standard error') all the information about a generated -morphology kernel.
Specify the alpha extra samples as associated, unassociated or unspecified
Skips reading the EXIF properties.
Allows one or more tag ID values to be ignored.
Sets the number of rows per strip
Sets the tile size for pyramid tiffs. Requires the suffix PTIF: before the outputname

For example, to create a postscript file that will render only the blackpixels of a bilevel image, use:

$ convert bilevel.tif -define ps:imagemask

Set attributes of the image registry by prefixing the value withregistry:. For example, to set a temporary path to put work files,use:

-define registry:temporary-path=/data/tmp

-delay ticks
-delay ticksxticks-per-second {<} {>}

display the next image after pausing.

This option is useful for regulating the animation of image sequencesticks/ticks-per-second seconds must expire before the display of thenext image. The default is no delay between each showing of the imagesequence. The default ticks-per-second is 100.

Use > to change the image delay only if its currentvalue exceeds the given delay. < changes the image delayonly if current value is less than the given delay. For example, ifyou specify 30> and the image delay is 20, the image delay doesnot change. However, if the image delay is 40 or 50, the delay it is changedto 30. Enclose the given delay in quotation marks to prevent the< or > from being interpreted by your shell asa file redirection.

-delete indexes

delete the images specified by index, from the image sequence.

Specify the image by its index in the sequence. The first image is index0. Negative indexes are relative to the end of the sequence, for example, -1represents the last image of the sequence. Specify a range of images witha dash (e.g. 0-4). Separate indexes with a comma (e.g. 0,2). Use+delete to delete the last image in the current image sequence.

-density width
-density widthxheight

Set the horizontal and vertical resolution of an image forrendering to devices.

This option specifies the image resolution to store while encoding a rasterimage or the canvas resolution while rendering (reading) vector formats suchas Postscript, PDF, WMF, and SVG into a raster image. Image resolutionprovides the unit of measure to apply when rendering to an output device orraster image. The default unit of measure is in dots per inch (DPI). The -units option may be used to select dots per centimeterinstead.

The default resolution is 72 dots per inch, which is equivalent to onepoint per pixel (Macintosh and Postscript standard). Computer screens arenormally 72 or 96 dots per inch, while printers typically support 150, 300,600, or 1200 dots per inch. To determine the resolution of your display, usea ruler to measure the width of your screen in inches, and divide by thenumber of horizontal pixels (1024 on a 1024x768 display).

If the file format supports it, this option may be used to update thestored image resolution. Note that Photoshop stores and obtains imageresolution from a proprietary embedded profile. If this profile is notstripped from the image, then Photoshop will continue to treat the image usingits former resolution, ignoring the image resolution specified in the standardfile header.

The -density option sets an attribute anddoes not alter the underlying raster image. It may be used to adjust therendered size for desktop publishing purposes by adjusting the scale appliedto the pixels. To resize the image so that it is the same size at a differentresolution, use the -resample option.

-depth value

depth of the image.

This the number of bits in a color sample within a pixel. Use this optionto specify the depth of raw images whose depth is unknown such as GRAY, RGB,or CMYK, or to change the depth of any image after it has been read.


obtain image by descending window hierarchy.

-deskew threshold

straighten an image. A threshold of 40% works for most images.

Use -set option:deskew:auto-cropwidth to auto crop the image. The set argument is the pixelwidth of the image background (e.g 40).


reduce the speckles within an image.

-direction type

render text right-to-left or left-to-right.

-displace horizontal-scale
-displace horizontal-scalexvertical-scale

shift image pixels as defined by a displacement map.[composite]

With this option, the 'overlay' image, and optionally the 'mask' image,is used as a displacement map, which is used to displace the lookup ofwhat part of the 'background' image is seen at each point of the overlaidarea. Much like the displacement map is a 'lens' that redirects light shiningthrough it so as to present a distorted view the original 'background' imagebehind it.

Any perfect grey areas of the displacement map produce a zerodisplacement of the image. Black areas produce the given maximum negativedisplacement of the lookup point, while white produce a maximum positivedisplacement of the lookup.

Note that it is the lookup of the 'background' that is displaced, not adisplacement of the image itself. As such an area of the displacement mapcontaining 'white' will have the lookup point 'shifted' by a positive amount,and thus generating a copy of the destination image to the right/downward fromthe correct position. That is the image will look like it may have been'shifted' in a negative left/upward direction. Understanding this is a veryimportant in understanding how displacement maps work.

The given arguments define the maximum amount of displacement in pixelsthat a particular map can produce. If the displacement scale is large enoughit is also possible to lookup parts of the 'background' image that lie welloutside the bounds of the displacement map itself. That is you could veryeasily copy a section of the original image from outside the overlay areainto the overlay area.

The '%' flag makes the displacement scale relative to the size of theoverlay image (100% = half width/height of image). Using '!' switchespercentage arguments to refer to the destination image size instead.these flags were added as of IM v6.5.3-5.

Normally a single grayscale displacement map is provided, which with thegiven scaling values will determine a single direction (vector) in whichdisplacements can occur (positively or negatively). However, if you alsospecify a third image which is normally used as a mask,the composite image is used for horizontal Xdisplacement, while the mask image is used for vertical Ydisplacement. This allows you to define completely different displacementvalues for the X and Y directions, and allowing you to lookup any point withinthe scale bounds. In other words each pixel can lookupany other nearby pixel, producing complex 2 dimensional displacements, ratherthan a simple 1 dimensional vector displacements.

Alternatively rather than supplying two separate images, as of IM v6.4.4-0,you can use the 'red' channel of the overlay image to specify the horizontalor X displacement, and the 'green' channel for the vertical or Y displacement.

As of IM v6.5.3-5 any alpha channel in the overlay image is used as amask the transparency of the destination image. However areas outside theoverlaid areas will not be effected.

-display host:display[.screen]

Specifies the X server to contact.[animate, display]

This option is used with convert for obtaining image or font from thisX server. See X(1).

-dispose method

define the GIF disposal image setting for images that are being created or read in.

The layer disposal method defines the way each the displayed image is to bemodified after the current 'frame' of an animation has finished beingdisplayed (after its 'delay' period), but before the next frame on ananimation is to be overlaid onto the display.

Here are the valid methods:

   Undefined   0  No disposal specified (equivalent to 'none').   None        1  Do not dispose, just overlay next frame image.   Background  2  Clear the frame area with the background color.   Previous    3  Clear to the image prior to this frames overlay.

You can also use the numbers given above, which is what the GIF formatuses internally to represent the above settings.

To print a complete list of dispose methods, use -list dispose.

Use +dispose, turn off the setting and preventresetting the layer disposal methods of images being read in.

Use -set 'dispose' method to set the imagedisposal method for images already in memory.

-dissimilarity-threshold value

maximum RMSE for subimage match (default 0.2).[compare]

-dissolve src_percent[xdst_percent]

dissolve an image into another by the given percent.[composite]

The opacity of the composite image is multiplied by the given percent, thenit is composited 'over' the main image. If src_percentis greater than 100, start dissolving the main image so it becomestransparent at a value of '200'. If both percentagesare given, each image are dissolved to the percentages given.

Note that dissolve percentages do not add, two opaque images dissolved'50,50', produce a 75% transparency. For a 50% + 50% blending of the twoimages, you would need to use dissolve values of '50,100'.

-distort method arguments

distort an image, using the given method and its required arguments.

The arguments is a single string containing a listof floating point numbers separated by commas or spaces. The number ofand meaning of the floating point values depends on the distortion method being used.

Choose from these distortion types:

Method Description
or SRT
Distort image by first scaling and rotating about a given 'center', before translating that 'center' to the new location, in that order. It is an alternative method of specifying a 'Affine' type of distortion, but without shearing effects. It also provides a good way of rotating and displacing a smaller image for tiling onto a larger background (IE 2-dimensional animations).
The number of arguments determine the specific meaning of each argument for the scales, rotation, and translation operations.
# arguments meaning
2:Scale Angle
3:X,Y Angle
4:X,Y Scale Angle
5: X,Y ScaleX,ScaleY Angle
6: X,Y Scale Angle NewX,NewY
7: X,Y ScaleX,ScaleY Angle NewX,NewY
This is actually an alternative way of specifying a 2 dimensional linear 'Affine' or 'AffineProjection' distortion.
Affine Distort the image linearly by moving a list of at least 3 or more sets of control points (as defined below). Ideally 3 sets or 12 floating point values are given allowing the image to be linearly scaled, rotated, sheared, and translated, according to those three points. See also the related 'AffineProjection' and 'SRT' distortions.
More than 3 sets given control point pairs (12 numbers) is least squares fitted to best match a linear affine distortion. If only 2 control point pairs (8 numbers) are given a two point image translation rotation and scaling is performed, without any possible shearing, flipping or changes in aspect ratio to the resulting image. If only one control point pair is provides the image is only translated, (which may be a floating point non-integer translation).
This distortion does not include any form of perspective distortion.
AffineProjection Linearly distort an image using the given Affine Matrix of 6 pre-calculated coefficients forming a set of Affine Equations to map the source image to the destination image.
sx, rx, ry, sy, tx, ty
See -affine setting for more detail, and meanings of these coefficients.
The distortions 'Affine' and 'SRT' provide alternative methods of defining this distortion, with ImageMagick doing the calculations needed to generate the required coefficients. You can see the internally generated coefficients, by using a -verbose setting with those other variants.
Bilinear Distortion, given a minimum of 4 sets of coordinate pairs, or 16 values (see below). Not that lines may not appear straight after distortion, though the distance between coordinates will remain consistent.
The 'BilinearForward' is used to map rectangles to any quadrilateral, while the 'BilinearReverse' form maps any quadrilateral to a rectangle, while preserving the straight line edges in each case.
Note that 'BilinearForward' can generate invalid pixels which will be colored using the -mattecolor color setting. Also if the quadrilateral becomes 'flipped' the image may disappear.
There are future plans to produce a true Bilinear distortion that will attempt to map any quadrilateral to any other quadrilateral, while preserving edges (and edge distance ratios).
Perspective Perspective distort the images, using a list of 4 or more sets of control points (as defined below). More that 4 sets (16 numbers) of control points provide least squares fitting for more accurate distortions (for the purposes of image registration and panorama effects). Less than 4 sets will fall back to a 'Affine' linear distortion.
Perspective Distorted images ensures that straight lines remain straight, but the scale of the distorted image will vary. The horizon is anti-aliased, and the 'sky' color may be set using the -mattecolor setting.
PerspectiveProjection Do a 'Perspective' distortion biased on a set of 8 pre-calculated coefficients. You can get these coefficients by looking at the -verbose output of a 'Perspective' distortion, or by calculating them yourself. If the last two perspective scaling coefficients are zero, the remaining 6 represents a transposed 'Affine Matrix'.
Arc Arc the image (variation of polar mapping) over the angle given around a circle.
Argument Meaning
arc_angle The angle over which to arc the image side-to-side
rotate_angle Angle to rotate resulting image from vertical center
top_radius Set top edge of source image at this radius
bottom_radius Set bottom edge to this radius (radial scaling)
The resulting image is always resized to best fit the resulting image, (as if using +distort) while attempting to preserve scale and aspect ratio of the original image as much as possible with the arguments given by the user. All four arguments will be needed to change the overall aspect ratio of an 'Arc'ed image.
This a variation of a polar distortion designed to try to preserve the aspect ratio of the image rather than direct Cartesian to Polar conversion.
Polar Like 'Arc' but do a complete Cartesian to Polar mapping of the image. that is the height of the input image is mapped to the radius limits, while the width is wrapped around between the angle limits.
Arguments: Rmax,Rmin CenterX,CenterY, start,end_angle
All arguments are optional. With Rmin defaulting to zero, the center to the center of the image, and the angles going from -180 (top) to +180 (top). If Rmax is given the special value of '0', the the distance from the center to the nearest edge is used for the radius of the output image, which will ensure the whole image is visible (though scaled smaller). However a special value of '-1' will use the distance from the center to the furthest corner, This may 'clip' the corners from the input rectangular image, but will generate the exact reverse of a 'DePolar' with the same arguments.
If the plus form of distort (+distort) is used output image center will default to 0,0 of the virtual canvas, and the image size adjusted to ensure the whole input image is made visible in the output image on the virtual canvas.
DePolar Uses the same arguments and meanings as a 'Polar' distortion but generates the reverse Polar to Cartesian distortion.
The special Rmax setting of '0' may however clip the corners of the input image. However using the special Rmax setting of '-1' (maximum center to corner distance) will ensure the whole distorted image is preserved in the generated result, so that the same argument to 'Polar' will reverse the distortion re-producing the original. Note that as this distortion requires the area resampling of a circular arc, which can not be handled by the builtin EWA resampling function. As such the normal EWA filters are turned off. It is recommended some form of 'super-sampling' image processing technique be used to produce a high quality result.
Barrel Given the four coefficients (A,B,C,D) as defined by Helmut Dersch, perform a barrel or pin-cushion distortion appropriate to correct radial lens distortions. That is in photographs, make straight lines straight again.
Arguments: A B C [ D [ X , Y ] ]
or Ax Bx Cx Dx Ay By Cy Dy [ X , Y ]
So that it forms the function
Rsrc = r * ( A*r3 + B*r2 + C*r + D )
Where X,Y is the optional center of the distortion (defaulting to the center of the image).
The second form is typically used to distort images, rather than correct lens distortions.
BarrelInverse This is very similar to 'Barrel' with the same set of arguments, and argument handling. However it uses the inverse of the radial polynomial, so that it forms the function
Rsrc = r / ( A*r3 + B*r2 + C*r + D )
Note that this is not the reverse of the 'Barrel' distortion, just a different barrel-like radial distortion method.
Shepards Distort the given list control points (any number) using an Inverse Squared Distance Interpolation Method (Shepards Method). The control points in effect do 'localized' displacement of the image around the given control point (preserving the look and the rotation of the area near the control points. For best results extra control points should be added to 'lock' the positions of the corners, edges and other unchanging parts of the image, to prevent their movement.
The distortion has been likened to 'taffy pulling' using nails, or pins' stuck in a block of 'jelly' which is then moved to the new position, distorting the surface of the jelly.
Internally it is equivalent to generating a displacement map (see -displace) for source image color look-up using the -sparse-color method of the same name.

To print a complete list of distortion methods, use -listdistort.

Many of the above distortion methods such as 'Affine','Perspective', and 'Shepards' use a list control pointsdefining how these points in the given image should be distorted in thedestination image. Each set of four floating point values represent a sourceimage coordinate, followed immediately by the destination image coordinate.This produces a list of values such as...

U1,V1 X1,Y1 U2,V2 X2,Y2 U3,V3 X3,Y3 ... Un,Vn Xn,Yn

where U,V on the source image is mapped to X,Y on thedestination image.

For example, to warp an image using 'perspective' distortion,needs a list of at least 4 sets of coordinates, or 16 numbers. Here is theperspective distortion of the built-in "rose:" image. Note how spaces wereused to group the 4 sets of coordinate pairs, to make it easier to read andunderstand.

$ convert rose: -virtual-pixel black \
-distort Perspective '0,0,0,0 0,45,0,45 69,0,60,10 69,45,60,35' \

If more that the required number of coordinate pairs are given fora distortion, the distortion method is 'least squares' fitted to produce thebest result for all the coordinate pairs given. If less than the ideal numberof points are given, the distort will generally fall back to a simpler form ofdistortion that can handles the smaller number of coordinates (usually a linear'Affine' distortion).

By using more coordinates you can make use of image registration tool tofind matching coordinate pairs in overlapping images, so as to improve the'fit' of the distortion. Of course a bad coordinate pair can also make the'fit' worse. Caution is always advised.

Colors are acquired from the source image according to a cylindricalresampling -filter, using a special technique known asEWA resampling. This produces very high quality results, especially whenimages become smaller (minified) in the output, which is very common whenusing 'perspective' distortion. For example here we viewa infinitely tiled 'plane' all the way to the horizon.

$ convert -size 90x90 pattern:checkerboard -normalize -virtual-pixel tile \
-distort perspective '0,0,5,45 89,0,45,46 0,89,0,89 89,89,89,89' \

Note that a infinitely tiled perspective images involving the horizon canbe very slow, because of the number of pixels that are compressed to generateeach individual pixel close to the 'horizon'. You can turn off EWAresampling, by specifing the special -filter setting of'point' (recommended if you plan to use super-sampling instead).

If an image generates invalid pixels, such as the 'sky' in the lastexample, -distort will use the current -mattecolor setting for these pixels. If you do notwhat these pixels to be visible, set the color to match the rest of theground.

The output image size will by default be the same as the input image. Thismeans that if the part of the distorted image falls outside the viewed area ofthe 'distorted space', those parts is clipped and lost. However if you usethe plus form of the operator (+distort) the operatorwill attempt (if possible) to show the whole of the distorted image, whileretaining a correct 'virtual canvas' offset, for image layering. This offsetmay need to be removed using +repage, to remove if itis unwanted.

Setting -verbose setting, will cause -distort to attempt to output the internal coefficients,and the -fx equivalent to the distortion, for expert study,and debugging purposes. This many not be available for all distorts.

You can alternatively specify a special "-define distort:viewport={geometry_string}" setting which willspecify the size and the offset of the generated 'viewport' image of thedistorted image space.

Setting a "-definedistort:scale={scale_factor}" will scale the output image (viewport orotherwise) by that factor without changing the viewed contents of thedistorted image. This can be used either for 'super-sampling' the image fora higher quality result, or for panning and zooming around the image (withappropriate viewport changes, or post-distort cropping and resizing).

Setting "-define resample:verbose=1"will output the cylindrical filter lookup table created by the EWA (EllipticalWeighted Average) resampling algorithm. Note this table uses a squared radiuslookup value. This is typically only used for debugging EWA resampling.

-distribute-cache port

launch a distributed pixel cache server.

-dither method

Apply a Riemersma or Floyd-Steinberg error diffusion dither toimages when general color reduction is applied via an option, or automagicallywhen saving to specific formats. This enabled by default.

Dithering places two or more colors in neighboring pixels so that to theeye a closer approximation of the images original color is reproduced. Thisreduces the number of colors needed to reproduce the image but at the cost ofa lower level pattern of colors. Error diffusion dithers can use any set ofcolors (generated or user defined) to an image.

Dithering is turned on by default, to turn it off use the plus form of thesetting, +dither. This will also also render PostScriptwithout text or graphic aliasing. Disabling dithering often (but not always)leads to faster process, a smaller number of colors, but more cartoon likeimage coloring. Generally resulting in 'color banding' effects in areas withcolor gradients.

The color reduction operators -colors, -monochrome, -remap, and -posterize, apply dithering to images using the reducedcolor set they created. These operators are also used as part of automaticcolor reduction when saving images to formats with limited color support, suchas GIF:, XBM:, and others, so dithering may also be usedin these cases.

Alternatively you can use -random-thresholdto generate purely random dither. Or use -ordered-dither to apply threshold mapped ditherpatterns, using uniform color maps, rather than specific color maps.

-draw string

Annotate an image with one or more graphic primitives.

Use this option to annotate or decorate an image with one or more graphicprimitives. The primitives include shapes, text, transformations, and pixeloperations.

The shape primitives:

   point           x,y   line            x0,y0 x1,y1   rectangle       x0,y0 x1,y1   roundRectangle  x0,y0 x1,y1 wc,hc   arc             x0,y0 x1,y1 a0,a1   ellipse         x0,y0 rx,ry a0,a1   circle          x0,y0 x1,y1   polyline        x0,y0  ...  xn,yn   polygon         x0,y0  ...  xn,yn   bezier          x0,y0  ...  xn,yn   path            path specification   image           operator x0,y0 w,h filename

The text primitive:

   text            x0,y0 string

The text gravity primitive:

   gravity         NorthWest, North, NorthEast, West, Center,                   East, SouthWest, South, or SouthEast

The text gravity primitive only affects the placement of text and does notinteract with the other primitives. It is equivalent to using the -gravity command-line option, except that it is limited inscope to the -draw option in which it appears.

The transformation primitives:

   rotate     degrees   translate  dx,dy   scale      sx,sy   skewX      degrees   skewY      degrees

The pixel operation primitives:

   color  x0,y0 method   matte  x0,y0 method

The shape primitives are drawn in the color specified by the preceding -fill setting. For unfilled shapes, use -fill none. You can optionally control the stroke (the"outline" of a shape) with the -stroke and -strokewidth settings.

A point primitive is specified by a single point in thepixel plane, that is, by an ordered pair of integer coordinates,x,y. (As it involves only a single pixel, a pointprimitive is not affected by -stroke or -strokewidth.)

A line primitive requires a start point and end point.

A rectangle primitive is specified by the pair of points at theupper left and lower right corners.

A roundRectangle primitive takes the same corner points asa rectangle followed by the width and height of the rounded cornersto be removed.

The circle primitive makes a disk (filled) or circle (unfilled).Give the center and any point on the perimeter (boundary).

The arc primitive is used to inscribe an elliptical segment into a given rectangle. An arc requires the two corners used forrectangle (see above) followed by the start and end angles of thearc of the segment segment (e.g. 130,30 200,100 45,90). The start and endpoints produced are then joined with a line segment and the resulting segmentof an ellipse is filled.

Use ellipse to draw a partial (or whole) ellipse. Give thecenter point, the horizontal and vertical "radii" (the semi-axes ofthe ellipse) and start and end angles in degrees (e.g. 100,100 100,1500,360).

The polyline and polygon primitives require three ormore points to define their perimeters. A polyline is simplya polygon in which the final point is not stroked to the startpoint. When unfilled, this is a polygonal line. If the -stroke setting is none (the default), thena polyline is identical to a polygon.

A coordinate is a pair of integers separated by a space oroptional comma.

As an example, to define a circle centered at 100,100 that extends to150,150 use:

-draw 'circle 100,100 150,150'

The Bezier primitive creates a spline curve and requires threeor points to define its shape. The first and last points are theknots and these points are attained by the curve, while anyintermediate coordinates are control points. If two control pointsare specified, the line between each end knot and its sequentially respectivecontrol point determines the tangent direction of the curve at that end. Ifone control point is specified, the lines from the end knots to the onecontrol point determines the tangent directions of the curve at each end. Ifmore than two control points are specified, then the additional control pointsact in combination to determine the intermediate shape of the curve. In orderto draw complex curves, it is highly recommended either to use thepath primitive or to draw multiple four-point bezier segments withthe start and end knots of each successive segment repeated. For example:

-draw 'bezier 20,50 45,100 45,0 70,50'

-draw 'bezier 70,50 95,100 95,0 120,50'

A path represents an outline of an object, defined in terms ofmoveto (set a new current point), lineto (draw a straight line), curveto (drawa Bezier curve), arc (elliptical or circular arc) and closepath (close thecurrent shape by drawing a line to the last moveto) elements. Compound paths(i.e., a path with subpaths, each consisting of a single moveto followed byone or more line or curve operations) are possible to allow effects such asdonut holes in objects. (See Paths.)

Use image to composite an image with another image. Follow theimage keyword with the composite operator, image location, image size, andfilename:

-draw 'image SrcOver 100,100 225,225 image.jpg'

You can use 0,0 for the image size, which means to use the actualdimensions found in the image header. Otherwise, it is scaled to the givendimensions. See Alpha Compositing fora detailed discussion of alpha composition methods that are available.

The "special augmented compose operators" such as "dissolve" that requirearguments cannot be used at present with the -draw image option.

Use text to annotate an image with text. Follow the textcoordinates with a string. If the string has embedded spaces, enclose it insingle or double quotes.

For example, the following annotates the image with Works likemagick! for an image titled bird.miff.

-draw "text 100,100 'Works like magick!' "

See the -annotate option for another convenient wayto annotate an image with text.

The rotate primitive rotates subsequent shape primitives andtext primitives about the origin of the main image. If the -region option precedes the -drawoption, the origin for transformations is the upper left corner of theregion.

The translate primitive translates subsequent shape and textprimitives.

The scale primitive scales them.

The skewX and skewY primitives skew them with respectto the origin of the main image or the region.

The transformations modify the current affine matrix, which is initializedfrom the initial affine matrix defined by the -affineoption. Transformations are cumulative within the -drawoption. The initial affine matrix is not affected; that matrix is only changedby the appearance of another -affine option. If another-draw option appears, the current affine matrix isreinitialized from the initial affine matrix.

Use the color primitive to change the color of a pixel to thefill color (see -fill). Follow the pixel coordinate witha method:

   point   replace   floodfill   filltoborder   reset

Consider the target pixel as that specified by your coordinate. Thepoint method recolors the target pixel. The replacemethod recolors any pixel that matches the color of the target pixel.Floodfill recolors any pixel that matches the color of the targetpixel and is a neighbor, whereas filltoborder recolors any neighborpixel that is not the border color. Finally, reset recolors allpixels.

Use matte to the change the pixel matte value to transparent.Follow the pixel coordinate with a method (see the color primitivefor a description of methods). The point method changes the mattevalue of the target pixel. The replace method changes the mattevalue of any pixel that matches the color of the target pixel.Floodfill changes the matte value of any pixel that matches thecolor of the target pixel and is a neighbor, whereas filltoborderchanges the matte value of any neighbor pixel that is not the border color (-bordercolor). Finally reset changes thematte value of all pixels.

You can set the primitive color, font, and font bounding box color with -fill, -font, and -boxrespectively. Options are processed in command line order so be sure to usethese options before the -draw option.

Strings that begin with a number must be quoted (e.g. use '1.png' ratherthan 1.png).

Drawing primitives conform to the MagickVector Graphics format.

-duplicate count,indexes

duplicate an image one or more times.

Specify the count and the image to duplicate by its index in the sequence.The first image is index 0. Negative indexes are relative to the end of thesequence, for example, -1 represents the last image of the sequence. Specifya range of images with a dash (e.g. 0-4). Separate indexes with a comma (e.g.0,2). Use +duplicate to duplicate the last image in the currentimage sequence.

-edge radius

detect edges within an image.

-emboss radius

emboss an image.

-encipher filename

Encipher pixels for later deciphering by -decipher.

Get the passphrase from the file specified by filename.

For more information, see the webpage, ImageMagick: Encipher orDecipher an Image.

-encoding type

specify the text encoding.

Choose from AdobeCustom, AdobeExpert,AdobeStandard, AppleRoman, BIG5,GB2312, Latin 2, None, SJIScode,Symbol, Unicode, Wansung.

-endian type

Specify endianness (MSB or LSB) of the image.

To print a complete list of endian types, use the -list endian option.

Use +endian to revert to unspecified endianness.


Apply a digital filter to enhance a noisy image.


perform histogram equalization on the image channel-by-channel.

To perform histogram equalization on all channels in concert, transform theimage into some other color space, such as HSL, OHTA, YIQ or YUV, thenequalize the appropriate intensity-like channel, then convert back to RGB.

For example using HSL, we have: ... -colorspace HSL -channel lightness-equalize -colorspace RGB ...

For YIQ, YUV and OHTA use the red channel. For example, OHTA is a principalcomponents transformation that puts most of the information in the firstchannel. Here we have ... -colorspace OHTA -channel red -equalize-colorspace RGB ...

-evaluate operator value

Alter channel pixels by evaluating an arithmetic, relational, or logical expression.

(See the -function operator for somemulti-parameter functions. See the -fx operator if moreelaborate calculations are needed.)

The behaviors of each operator are summarized in thefollowing list. For brevity, the numerical value of a "pixel" referred tobelow is the value of the corresponding channel of that pixel, whilea "normalized pixel" is that number divided by the maximum(installation-dependent) value QuantumRange. (Ifnormalized pixels are used, they are restored, following the othercalculations, to the full range by multiplying by QuantumRange.)

operator Summary (see further below for details)
Abs Add value to pixels and return absolute value.
Add Add value to pixels.
AddModulus Add value to pixels modulo QuantumRange.
And Binary AND of pixels with value.
Cos, Cosine Apply cosine to pixels with frequency value with 50% bias added.
Divide Divide pixels by value.
Exp base-e exponential function
Exponential base-e exponential function
LeftShift Shift the pixel values left by value bits (i.e., multiply pixels by 2value).
Log Apply scaled logarithm to normalized pixels.
Max Clip pixels at lower bound value.
Mean Add the value and divide by 2.
Median Choose the median value from an image sequence.
Min Clip pixels at upper bound value.
Multiply Multiply pixels by value.
Or Binary OR of pixels with value.
Pow Raise normalized pixels to the power value.
RightShift Shift the pixel values right by value bits (i.e., divide pixels by 2value).
RMS Square the pixel and add the value.
RootMeanSquare Square the pixel and add the value.
Set Set pixel equal to value.
Sin, Sine Apply sine to pixels with frequency value with 50% bias added.
Subtract Subtract value from pixels.
Xor Binary XOR of pixels with value.
Multiplicative-noise (These are equivalent to the corresponding -noise operators.)
Threshold Threshold pixels larger than value.
ThresholdBlack Threshold pixels to zero values equal to or below value.
ThresholdWhite Threshold pixels to maximum values above value.

The specified functions are applied only to each previously set -channel in the image. If necessary, the results of thecalculations are truncated (clipped) to fit in the interval [0, QuantumRange]. The transparency channel of the image isrepresented as a 'alpha' values (0 = fully transparent), so, for example, aDivide by 2 of the alpha channel will make the imagesemi-transparent. Append the percent symbol '%' to specify a valueas a percentage of the QuantumRange.

To print a complete list of -evaluate operators, use-list evaluate.

The results of the Add, Subtract andMultiply methods can also be achieved using either the -level or the +level operator, withappropriate argument, to linearly modify the overall range of color values.Please note, however, that -level treats transparency as'matte' values (0 = opaque), while -evaluate works with'alpha' values.

AddModulus has been added as of ImageMagick 6.4.8-4 and providesaddition modulo the QuantumRange. It is thereforeequivalent to Add unless the resulting pixel value is outside theinterval [0, QuantumRange].

Exp or Exponential has been added as of ImageMagick 6.6.5-1 andworks on normalized pixel values. The value used withExp should be negative so as to produce a decaying exponentialfunction. Non-negative values will always produce results larger unity andthus outside the interval [0, QuantumRange]. Theformula is expressed below.

exp(value × u)

If the input image is squared, for example, using -function polynomial "2 0 0", then a decaying Gaussian function will bethe result.

Log has been added as of ImageMagick 6.4.2-1 and works onnormalized pixel values. This a scaled log function. The value used with Log provides a scalingfactor that adjusts the curvature in the graph of the log function. Theformula applied to a normalized value u is below.

log(value × u + 1) / log(value + 1)

Pow has been added as of ImageMagick 6.4.1-9, and works onnormalized pixel values. Note that Pow is related to the -gamma operator. For example, -gamma 2 is equivalentto -evaluate pow 0.5, i.e., a 'square root' function. The value usedwith -gamma is simply the reciprocal of the value usedwith Pow.

Cosine and Sine was added as of IM v6.4.8-8 andconverts the image values into a value according to a (co)sine wave function.The synonyms Cos and Sin may also be used. The outputis biased 50% and normalized by 50% so as to fit in the respective color valuerange. The value scaling of the period of thefunction (its frequency), and thus determines the number of 'waves' that willbe generated over the input color range. For example, if the value is 1, the effective period is simply the QuantumRange; but if the value is 2,then the effective period is the half the QuantumRange.

0.5 + 0.5 × cos(2 π u × value).

See also the -function operator, which is amulti-value version of evaluate.

-evaluate-sequence operator

Alter channel pixels by evaluating an arithmetic, relational, orlogical expression over a sequence of images.

To print a complete list of -evaluate-sequence operators, use -list evaluate.

-extent geometry

Set the image size and offset.

If the image is enlarged, unfilled areas are set to the background color.To position the image, use offsets in the geometryspecification or precede with a -gravity setting. Tospecify how to compose the image with the background, use -compose.

This command reduces or expands a JPEG image to fit on an 800x600display. If the aspect ratio of the input image isn't exactly 4:3, then theimage is centered on an 800x600 black canvas:

$ convert input.jpg -resize 800x600 -background black -compose Copy \
-gravity center -extent 800x600 -quality 92 output.jpg

See Image Geometry for complete details about the geometry argument.

-extract geometry

Extract the specified area from image.

This option is most useful for extracting a subregion of a very large rawimage. Note that these two commands are equivalent:

$ convert -size 16000x16000 -depth 8 -extract 640x480+1280+960 \
image.rgb image.png
$ convert -size 16000x16000 -depth 8 'image.rgb[640x480+1280+960]' \
image.rgb image.png

If you omit the offsets, as in

$ convert -size 16000x16000 -depth 8 -extract 640x480 \
image.rgb image.png

the image is resized to the specified dimensions instead,equivalent to:

$ convert -size 16000x16000 -depth 8 -resize 640x480 image.rgb image.png

See Image Geometry for complete details about the geometry argument.

-family fontFamily

Set a font family for text.

This setting suggests a font family that ImageMagick should try to use forrendering text. If the family can be found it is used; if not, a default font(e.g., "Arial") or a family known to be similar is substituted (e.g.,"Courier" might be used if "System" is requested but not found).

For other settings that affect fonts, see the options -font, -stretch, -style, and -weight.

-features distance

display (cooccurrence matrix) texture measure features for each channel in the image in each of four directions (horizontal, vertical, left and right diagonals) for the specified distance.[identify]
      Angular Second Moment      Contrast      Correlation      Sum of Squares Variance      Inverse Difference Moment      Sum Average      Sum Variance      Sum Entropy      Entropy      Difference Variance      Difference Entropy      Information Measure of Correlation 1      Information Measure of Correlation 2      Maximum Correlation Coefficient


implements the forward discrete Fourier transform (DFT).

This option is new as of ImageMagick 6.5.4-3 (and now working for Windowsusers in ImageMagick 6.6.0-9). It transforms an image from the normal(spatial) domain to the frequency domain. In the frequency domain, an image isrepresented as a superposition of complex sinusoidal waves of varyingamplitudes. The image x and y coordinates are the possible frequencies alongthe x and y directions, respectively, and the pixel intensity values arecomplex numbers that correspond to the sinusoidal wave amplitudes. See forexample, FourierTransform, Discrete FourierTransform and Fast FourierTransform.

A single image name is provided as output for this option. However, theoutput result will have two components. It is either a two-frame image or twoseparate images, depending upon whether the image format specified supportsmulti-frame images. The reason that we get a dual output result is because thefrequency domain represents an image using complex numbers, which cannot bevisualized directly. Therefore, the complex values are automagically separatedinto a two-component image representation. The first component is themagnitude of the complex number and the second is the phase of the complexnumber. See for example, Complex Numbers.

The magnitude and phase component images must be specified using imageformats that do not limit the color or compress the image. Thus, MIFF, TIF,PFM, EXR and PNG are the recommended image formats to use. All of theseformats, except PNG support multi-frame images. So for example,

$ convert image.png -fft fft_image.miff

generates a magnitude image as fft_image.miff[0] and a phaseimage as fft_image.miff[1]. Similarly,

$ convert image.png -fft fft_image.png

generates a magnitude image as fft_image-0.png and a phase imageas fft_image-1.png. If you prefer this representation, then you canforce any of the other formats to produce two output images by including +adjoin following -fft in the command line.

The input image can be any size, but if not square and even-dimensioned, itis padded automagically to the larger of the width or height of the inputimage and to an even number of pixels. The padding will occur at the bottomand/or right sides of the input image. The resulting output magnitude andphase images is square at this size. The kind of padding relies on the -virtual-pixel setting.

Both output components will have dynamic ranges that fit within[0, QuantumRange], so that HDRI need not be enabled.Phase values nominally range from 0 to 2*π, but is scaled to span the fulldynamic range. (The first few releases had non-HDRI scaled but HDRI notscaled). The magnitude image is not scaled and thus generally will containvery small values. As such, the image normally will appear totally black. Inorder to view any detail, the magnitude image typically is enhanced with a logfunction into what is usually called the spectrum. A log function is used toenhance the darker values more in comparison to the lighter values. This canbe done, for example, as follows:

$ convert fft_image.miff[0] -contrast-stretch 0 \
-evaluate log 1000 fft_image_spectrum.png

where either -contrast-stretch 0 or -auto-level is used to scale the image to full dynamicrange, first. The argument to the -evaluate logtypically is specified between 100 and 10,000, depending upon the amount ofdetail that one wants to bring out in the spectrum. Larger values produce morevisible detail. Too much detail, however, may hide the important features.

The FFTW delegate library is required touse -fft.

Use +fft to produce two output images that are the realand imaginary components of the complex valued Fourier transform.

However, as the real and imaginary components can contain negative values,this requires that IM be configured with HDRI enabled. In this case, you mustuse either MIFF, TIF, PFM or MPC formats for the real and imaginary componentresults, since they are formats that preserve both negative and fractionalvalues without clipping them or truncating the fractional part. With eitherMIFF or TIF, one should add -define quantum:format=32, to allow those imagetypes to work properly in HDRI mode without clipping.

The real and imaginary component images resulting from +fft are also square, even dimensioned images due to the samepadding that was discussed above for the magnitude and phase componentimages.

See the discussion on HDRI implementations of ImageMagick on the page High Dynamic-Range Images. For moreabout HDRI go the ImageMagick Usage pages, Fred's Fourier Processing With ImageMagick page or this Wikipedia entry.

By default the FFT is normalized (and the IFT is not). Use "-define fourier:normalize=forward to explicityly normalize the FFT and unnormalize the IFT.

-fill color

color to use when filling a graphic primitive.

This option accepts a color name, a hex color, or a numerical RGB, RGBA,HSL, HSLA, CMYK, or CMYKA specification. See Color Names fora description of how to properly specify the color argument.

Enclose the color specification in quotation marks to prevent the "#" orthe parentheses from being interpreted by your shell.

For example,

-fill blue

-fill "#ddddff"

-fill "rgb(255,255,255)"

See -draw for further details.

To print a complete list of color names, use the -list color option.

-filter type

Use this type of filter when resizing ordistorting an image.

Use this option to affect the resizing operation of an image duringoperations such as -resize and -distort. For example you can use a simple resize filter such as:

   Point       Hermite       Cubic   Box         Gaussian      Catrom   Triangle    Quadratic     Mitchell

The Bessel and Sinc filter is also provided (as wellas a faster SincFast equivalent form). However these filters aregenerally useless on their own as they are infinite filters that are beingclipped to the filters support size. Their direct use is not recommendedexcept via expert settings (see below).

Instead these special filter functions are typically windowed by a windowingfunction that the -filter setting defines. That isusing these functions will define a 'Windowed' filter, appropriate to theoperator involved. Windowed filters include:

   Lanczos       Hamming       Parzen   Blackman      Kaiser        Welsh   Hanning       Bartlett      Bohman

Also one special self-windowing filter is also providedLagrange, which will automagically re-adjust its function dependingon the current 'support' or 'lobes' expert settings (see below).

If you do not select a filter with this option, the filter defaults toMitchell for a colormapped image, an image with a matte channel, orif the image is enlarged. Otherwise the filter default toLanczos.

To print a complete list of resize filters, use the -listfilter option.

You can modify how the filter behaves as it scales your image through theuse of these expert settings (see also -define and -set):-

-define filter:blur=factor
Scale the X axis of the filter (and its window). Use > 1.0 for blurry or < 1.0 for sharp. This should only be used with Gaussian and Gaussian-like filters simple filters, or you may not get the expected results.
-define filter:support=radius
Set the filter support radius. Defines how large the filter should be and thus directly defines how slow the filtered resampling process is. All filters have a default 'prefered' support size. Some filters like Lagrange and windowed filters adjust themselves depending on this value. With simple filters this value either does nothing (but slow the resampling), or will clip the filter function in a detrimental way.
-define filter:lobes=count
Set the number of lobes to use for the Sinc/Bessel filter. This an alternative way of specifying the 'support' range of the filter, that is designed to be more suited to windowed filters, especially when used for image distorts.
-define filter:sigma=value
The 'sigma' value used to define the Gaussian filter. Default sigma value is '0.5'. It only effects Gaussian but does not shrink (but may enlarge) the filter's 'support'. It can be used to generate very small blurs but without the filter 'missing' pixels due to using a small support setting. A larger value of '0.707' (a value of '1/sqrt(2)') is another common setting.
-define filter:b=b-spline_factor
-define filter:c=keys_alpha_factor
Redefine the values used for cubic filters such as Cubic, Catrom, Mitchel, and Hermite, as well as the Parzen cubic windowing function. If only one of the values are defined, the other is set so as to generate a 'Cubic-Keys' filter. The values meaning was defined by a research paper by Mitchell-Netravali.
-define filter:kaiser-beta=value
The 'alpha' value used to as part of the Kaiser Windowing function. Default value is '6.5'. It only effects Kaiser windowing function, and does not effect any other attributes.
Before IM v6.7.6-10, this option was known as "filter:alpha", (an inheritance from the very old "zoom" program). It was changed to bring the function in line with more modern academic research usage, and better assign it be more definitive.
-define filter:kaiser-alpha=value
This value when multiplied by 'PI' is equivalent to "kaiser-beta", and will override that setting. It only effects Kaiser windowing function, and does not effect any other attributes.
-define filter:filter=filter_function
Use this function directly as the weighting filter. This will allow you to directly use a windowing function such as Blackman, as a resampling filter, rather than as its normal usage as a windowing function.
If defined, no windowing function also defined, the window function is set to Box). Directly specifying Sinc or Jinc as a filter will also do this.
-define filter:window=filter_function
The IIR (infinite impulse response) filters Sinc and Jinc are windowed (brought down to zero over the defined support range) with the given filter. This allows you to specify a filter function to be used as a windowing function for these IIR filters.
Many of the defined filters are actually windowing functions for these IIR filters. A typical choices is Box, (which effectively turns off the windowing function).
-define filter:win-support=radius
Scale windowing function to this size instead. This causes the windowing (or self-windowing Lagrange filter) to act is if the support window is larger than what is actually supplied to the calling operator. The filter however is still clipped to the real support size given. If unset this will equal the normal filter support size.
-define filter:verbose=1
This causes IM to print information on the final internal filter selection to standard output. This includes a commented header on the filter settings being used, and data allowing the filter weights to be easily graphed.
Note however that some filters are internally defined in terms of other filters. The Lanczos filter for example is defined in terms of a SincFast windowed SincFast filter, while Mitchell is defined as a general Cubic family filter with specific 'B' and 'C' settings.

For example, to get a 8 lobe jinc windowed sinc filter (Genseng filter?):

$ convert image.png \
-filter sinc \
-set filter:window=jinc \
-set filter:lobes=8 \
-resize 150% image.jpg

Or a raw un-windowed Sinc filter with 4 lobes:

$ convert image.png \
-set filter:filter=sinc \
-set filter:lobes=4 \
-resize 150% image.jpg

To extract the data for a raw windowing function, combine it witha 'Box' filter. For example the 'Welch parabolicwindowing function.

$ convert null: -define filter:filter=Box \
-define filter:window=Welch \
-define filter:support=1.0 \
-define filter:verbose=1 \
-resize 2 null: > window_welch.dat
set grid
plot "window_welch.dat" with lines

Note that the use of expert options is provided for image processing expertswho have studied and understand how resize filters work. Without thisknowledge, and an understanding of the definition of the actual filtersinvolved, using expert settings are more likely to be detrimental to your imageresizing.


This is a simple alias for the -layers method "flatten".


create a mirror image.

reflect the scanlines in the vertical direction. The image will be mirroredupside-down.

-floodfill {+-}x{+-}y color

floodfill the image with color at the specified offset.

Flood fill starts from the given 'seed point' which is not gravity effected.Any color that matches within -fuzz color distance of thegiven color argument, connected to that 'seed point'will be replaced with the current -fill color.

Note that if the pixel at the 'seed point' does not itself match the givencolor (according to -fuzz), then noaction will be taken.

This operator works more like the -opaque option, thana more general flood fill that reads the matching color directly at the 'seedpoint'. For this form of flood fill, look at -draw andits 'color floodfill' drawing method.


create a mirror image.

Reflect the scanlines in the horizontal direction, just like the image ina vertical mirror.

-font name

set the font to use when annotating images with text, or creating labels.

To print a complete list of fonts, use the -list fontoption (for versions prior to 6.3.6, use 'type' instead of 'font').

In addition to the fonts specified by the above pre-defined list, you canalso specify a font from a specific source. For example Arial.ttfis a TrueType font file, ps:helvetica is PostScript font, andx:fixed is X11 font.

For other settings that affect fonts, see the options -family, -stretch, -style, and -weight.

-foreground color

Define the foreground color for menus.[display]

The color is specified using the format described under the -fill option.

The default foreground color is black.

-format type

the image format type.

When used with the mogrify utility, this option converts anyimage to the image format you specify.For a list of image format types supported by ImageMagick, use -list format.

By default the file is written to its original name. However, if thefilename extension matches a supported format, the extension is replaced withthe image format type specified with -format. Forexample, if you specify tiff as the format type and theinput image filename is image.gif, the output imagefilename becomes image.tiff.

-format string

output formatted image characteristics.[identify]

See Format and Print ImageProperties for an explanation on how to specify the argument to thisoption.

-frame geometry

Surround the image with a border or beveled frame.

The color of the border is specified with the -mattecolor command line option.

See Image Geometry for complete details about the geometry argument. The size portion of the geometry argument indicates the amount of extra width andheight that is added to the dimensions of the image. If no offsets are givenin the geometry argument, then the border added isa solid color. Offsets x and y, if present, specify thatthe width and height of the border is partitioned to form an outer bevel ofthickness x pixels and an inner bevel of thicknessy pixels. Negative offsets make no sense as frame arguments.

The -frame option is affected by the current -compose setting and assumes that this is using the default'Over' composition method. It generates an image of the appropriatesize with the current -bordercolor setting, and thendraws the frame of four distinct colors close to the current -mattecolor. The original image is then overlaid ontocenter of this image. This means that with the default compose method of'Over' any transparent parts may be replaced by the current -bordercolor setting.

The image composition is notaffected by the -gravity option.


include the X window frame in the imported image.[import]

-function function parameters

Apply a function to channel values.

This operator performs calculations based on the given arguments to modifyeach of the color values for each previously set -channel in the image. See -evaluate for details concerning how the results of thecalculations are handled.

This is can be considered a multi-argument version of the -evaluate operator. (Added inImageMagick 6.4.8−8.)

Here, parameters is a comma-separated list ofnumerical values. The number of values varies depending on which function is selected. Choose the function from:

   Polynomial   Sinusoid   Arcsin   Arctan

To print a complete list of -function operators,use -list function. Descriptions follow.


The Polynomial function takes an arbitrary number of parameters,these being the coefficients of a polynomial, in decreasing order of degree.That is, entering

-function Polynomial an,an-1,...a1,a0

will invoke a polynomial function given by

an un + an-1 un-1 + ··· a1 u + a0,

where u is pixel's original normalized channel value.

The Polynomial function can be used in place of Set(the constant polynomial) and Add, Divide,Multiply, and Subtract (some linearpolynomials) of the -evaluate operator. The -level operator also affects channels linearly. Somecorrespondences follow.

-evaluate Set value -function Polynomial value (Constant functions; set value×100% gray when channels are RGB.)
-evaluate Add value -function Polynomial 1,value
-evaluate Subtract value -function Polynomial 1,−value
-evaluate Multiply value -function Polynomial value,0
+level black% x white% -function Polynomial A,B (Reduce contrast. Here, A=(white-black)/100 and B=black/100.)

The Polynomial function gives great versatility, sincepolynomials can be used to fit any continuous curve to any degree of accuracydesired.


The Sinusoid function can be used to vary the channel valuessinusoidally by setting frequency, phase shift, amplitude, and a bias. Thesevalues are given as one to four parameters, as follows,

-function Sinusoid freq,[phase,[amp,[bias]]]

where phase is in degrees. (The domain [0,1] of the functioncorresponds to 0 through freq×360 degrees.)The result is that if a pixel's normalized channel value is originallyu, its resulting normalized value is given by

amp * sin(2*π* (freq * u + phase / 360)) + bias

For example, the following generates a curve that starts and ends at 0.9(when u=0 and 1, resp.), oscillating three times between.7−.2=.5 and .7+.2=.9.

-function Sinusoid 3,-90,.2,.7

The default values of amp and bias are both .5. The default for phaseis 0.

The Sinusoid function generalizes Sin andCos of the -evaluate operator by allowingvarying amplitude, phase and bias. The correspondence is as follows.

-evaluate Sin freq -function Sinusoid freq,0
-evaluate Cos freq -function Sinusoid freq,90

The ArcSin function generates the inverse curve of a Sinusoid,and can be used to generate cylindrical distortion and displacement maps.The curve can be adjusted relative to both the input values and output rangeof values.

-function ArcSin width,[center,[range,[bias]]]

with all values given in terms of normalized color values (0.0 for black,1.0 for white). Defaulting to values covering the full range from 0.0 to 1.0for bout input (width), and output (width) values. '1.0,0.5,1.0,0.5'

range/π * asin( 2/width * ( u - center ) ) + bias


The ArcTan function generates a curve that smooth crosses fromlimit values at infinities, though a center using the given slope value.All these values can be adjusted via the arguments.

-function ArcTan slope,[center,[range,[bias]]]

Defaulting to '1.0,0.5,1.0,0.5'.

range/π * atan( slope*π * ( u - center ) ) + bias

-fuzz distance{%}

Colors within this distance are considered equal.

A number of algorithms search for a target color. By default the color mustbe exact. Use this option to match colors that are close to the target colorin RGB space. For example, if you want to automagically trim the edges of animage with -trim but the image was scanned and the targetbackground color may differ by a small amount. This option can account forthese differences.

The distance can be in absolute intensity units or, byappending % as a percentage of the maximum possible intensity (255,65535, or 4294967295).

Use +fuzz to reset the fuzz value to 0.

-fx expression

apply a mathematical expression to an image or image channels.

If the first character of expression is @,the expression is read from a file titled by the remaining characters in thestring.

See FX,The Special Effects Image Operator for a detailed discussion of thisoption.

-gamma value

level of gamma correction.

The same color image displayed on two different workstations may lookdifferent due to differences in the display monitor. Use gamma correction toadjust for this color difference. Reasonable values extend from0.8 to 2.3. Gamma less than 1.0 darkens the image andgamma greater than 1.0 lightens it. Large adjustments to image gamma mayresult in the loss of some image information if the pixel quantum size is onlyeight bits (quantum range 0 to 255).

Gamma adjusts the image's channel values pixel-by-pixel according toa power law, namely, pow(pixel,1/gamma) or pixel^(1/gamma), where pixel is thenormalized or 0 to 1 color value. For example, using a value of gamma=2 is thesame as taking the square root of the image.

You can apply separate gamma values to the red, green, and blue channels ofthe image with a gamma value list delimited with commas (e.g.,1.7,2.3,1.2).

Use +gamma value to set theimage gamma level without actually adjusting the image pixels. This optionis useful if the image is of a known gamma but not set as an image attribute(e.g. PNG images). Write the "file gamma" which is the reciprocal of thedisplay gamma; e.g., if your image is sRGB and you want to write a PNG gAMAchunk, use

$ convert input.png +gamma .45455 output.png

(0.45455 is 1/2.2)

Note that gamma adjustments are also available via the -level operator.

-gaussian-blur radius
-gaussian-blur radiusxsigma

Blur the image with a Gaussian operator.

Convolve the image with a Gaussian or normal distribution using the givenSigma value. The formula is:

gaussian distribution

The Sigma value is the important argument, anddetermines the actual amount of blurring that will take place.

The Radius is only used to determine the size of thearray which will hold the calculated Gaussian distribution. It should be aninteger. If not given, or set to zero, IM will calculate the largest possibleradius that will provide meaningful results for the Gaussian distribution.

The larger the Radius the radius the slower theoperation is. However too small a Radius, and severaliasing effects may result. As a guideline, Radiusshould be at least twice the Sigma value, though threetimes will produce a more accurate result.

This differs from the faster -blur operator in that afull 2-dimensional convolution is used to generate the weighted average of theneighboring pixels.

The -virtual-pixel setting will determine howpixels which are outside the image proper are blurred into the final result.

-geometry geometry

Set the preferred size and location of the image.

See Image Geometry for complete details about the geometry argument.

-gravity type

Sets the current gravity suggestion for various other settings and options.

Choices include: NorthWest, North,NorthEast, West, Center, East,SouthWest, South, SouthEast. Use -list gravity to get a complete list of -gravity settings available in your ImageMagickinstallation.

The direction you choose specifies where to position text or subimages. Forexample, a gravity of Center forces the text to be centered withinthe image. By default, the image gravity is NorthWest. See -draw for more details about graphic primitives. Only thetext primitive of -draw is affected by the -gravity option.

The -gravity option is also used in concert with the-geometry setting and other settings or options thattake geometry as an argument, such as the -crop option.

If a -gravity setting occurs before another optionor setting having a geometry argument that specifies anoffset, the offset is usually applied to the point within the image suggestedby the -gravity argument. Thus, in the followingcommand, for example, suppose the file image.png has dimensions200x100. The offset specified by the argument to -regionis (−40,+20). The argument to -gravity isCenter, which suggests the midpoint of the image, at the point(100,50). The offset (−40,20) is applied to that point, giving(100−40,50+20)=(60,70), so the specified 10x10 region is located atthat point. (In addition, the -gravity affects theregion itself, which is centered at the pixelcoordinate (60,70). (See Image Geometry for complete details about the geometry argument.)

$ convert image.png -gravity Center -region 10x10-40+20 \
-negate output.png

When used as an option to composite, -gravity gives the direction that the image gravitateswithin the composite.

When used as an option to montage, -gravity gives the direction that an image gravitateswithin a tile. The default gravity is Center for this purpose.

-grayscale method

convert image to grayscale.

This will use one of the -intensity methods toconvert the given image into a linear-grayscale image.

For example, to convert an image to (linear) Rec709Luminance grayscale, type:

$ convert in.png -grayscale Rec709Luminance out.png

which is equivalent to:

$ convert in.png -colorspace gray out.png

Similarly, to convert an image to (non-linear) Rec709Luma grayscale, type:

$ convert in.png -grayscale Rec709Luma out.png

which is equivalent to:

$ convert in.png -set colorspace RGB -colorspace gray out.png

Note that a 'colorspace' intensity method will produce the same resultregardless of the current colorpsace of the image. But a 'mathematical'intensity method will depend on the current colorspace the image is currentlyusing.

While this operation uses an -intensity method, it does not use or set the -intensity setting, sowill not effect other operations that may use that setting.

-green-primary x,y

green chromaticity primary point.


apply a Hald color lookup table to the image.

A Hald color lookup table is a 3-dimensional color cube mapped to 2dimensions. Create it with the HALD: prefix (e.g. HALD:8). Youcan apply any color transformation to the Hald image and then use this optionto apply the transform to the image.

$ convert image.png hald.png -hald-clut transform.png

This option provides a convenient method for you to use Gimp or Photoshopto make color corrections to the Hald CLUT image and subsequently apply themto multiple images using an ImageMagick script.

Note that the representation is only of the normal RGB color space and thatthe whole color value triplet is used for the interpolated lookup of therepresented Hald color cube image. Because of this the operation is not -channel setting effected, nor can it adjust or modify animages transparency or alpha/matte channel.

See also -clut which provides color value replacementof the individual color channels, usually involving a simpler gray-scaleimage. E.g: gray-scale to color replacement, or modification by a histogrammapping.


print usage instructions.

-highlight-color color

when comparing images, emphasize pixel differences with this color.

-hough-lines widthxheight{+threshold}

identify straight lines in the image (e.g. -hough-lines 9x9+195).

Use the Hough line detector with any binary edge extracted image to locate and draw any straight lines that it finds.

The process accumulates counts for every white pixel in the binary edge image for every possible orientation (for angles from 0 to 179 in 1 deg increments) and distance from the center of the image to the corners (in 1 px increments). It stores the counts in an accumulator matrix of angle vs distance. The size of the accumulator will be 180x(diagonal/2). Next it searches the accumulator for peaks in counts and converts the locations of the peaks to slope and intercept in the normal x,y input image space. The algorithm uses slope/intercepts to find the endpoints clipped to the bounds of the image. The lines are drawn from the given endpoints. The counts are a measure of the length of the lines.The WxH arguments specify the filter size for locating the peaks in the Hough accumulator. The threshold excludes lines whose counts are less than the threshold value.

Use -background to specify the color of the background onto which the lines will be drawn. The default is black.

Use -fill to specify the color of the lines. The default is black.

Use -stroke and -strokewidth to specify the thickness of the lines. The default is black and no strokewidth.

A text file listing the endpoints and counts may be created by using the suffix, .mvg, for the output image.

Use -define hough-lines:accumulator=true to return the accumulator image in addition to the lines image.

-iconGeometry geometry

specify the icon geometry.

Offsets, if present in the geometry specification, are handled in the samemanner as the -geometry option, using X11 style tohandle negative offsets.

See Image Geometry for complete details about the geometry argument.


start in icon mode in X Windows[animate, display]


identify the format and characteristics of the image.

This information is printed: image scene number; image name; image size;the image class (DirectClass or PseudoClass); the total number of unique colors; and thenumber of seconds to read and transform the image. Refer to MIFF fora description of the image class.

If -colors is also specified, the total unique colorsin the image and color reduction error values are printed. Refer to colorreduction algorithm for a description of these values.

If -verbose precedes this option, copiousamounts of image properties are displayed including image statistics, profiles,image histogram, and others.


implements the inverse discrete Fourier transform (DFT).

This option is new as of ImageMagick 6.5.4-3 (and now working for Windowsusers in ImageMagick 6.6.0-9). It transforms a pair of magnitude and phaseimages from the frequency domain to a single image in the normal or spatialdomain. See for example, Fourier Transform,Discrete Fourier Transform andFast Fourier Transform.

For example, depending upon the image format used to store the result ofthe -fft, one would use either

$ convert fft_image.miff -ift fft_image_ift.png


$ convert fft_image-0.png fft_image-1.png -ift fft_image_ift.png

The resulting image may need to be cropped due to padding introduced whenthe original image, prior to the -fft or +fft, was not square or even dimensioned. Any padding is atthe right and/or bottom sides of the image.

The FFTW delegate library is required touse -ift.

Use +ift (with HDRI enabled) to transform a pair of realand imaginary images from the frequency domain to a single image in the normal(spatial) domain.

By default the IFT is not normalized (and the FFT is). Use "-define fourier:normalize=inverse to explictly normalize the IFT and unnormalize the FFT.


make image immutable.

-implode factor

implode image pixels about the center.

-insert index

insert the last image into the image sequence.

This option takes last image in the current image sequence and inserts itat the given index. If a negative index is used, the insert position iscalculated before the last image is removed from the sequence. As such-insert -1 will result in no change to the image sequence.

The +insert option is equivalent to -insert -1. Inother words, insert the last image, at the end of the current image sequence.Consequently this has no effect on the image sequence order.

-intensity method

method to generate intensity value from pixel.

ImageMagick provides a number of methods used in situations where anoperatory needs to determine a single grayscale value for some purpose, froman image with red, green, and blue pixel components. Typically the linearRec709Luminance formula is used, which is the same formula used when converting images to -colorspace gray.

The following formulas are currently provided, and will first convertthe pixel values to linear-RGB or non-linear sRGB colorspace beforebeing applied to calulate the final greyscale value.

   Rec601Luma        0.298839R' + 0.586811G'+ 0.114350B'   Rec601Luminance   0.298839R + 0.586811G + 0.114350B   Rec709Luma        0.212656R' + 0.715158G' + 0.072186B'   Rec709Luminance   0.212656R + 0.715158G + 0.072186B   Brightness        max(R', G', B')   Lightness         (min(R', G', B') + max(R', G', B')) / 2.0

Note that the above R,G,B values is the image's linear-RGB values, whileR',G',B' are sRGB non-linear values.

These intensity methods are mathematical in nature and will use thecurrent value in the images respective R,G,B channel regardless ofwhat that is, or what colorspace the image is currently using.

   Average           (R + G + B) / 3.0   MS                (R^2 + G^2 + B^2) / 3.0   RMS               sqrt( (R^2 + G^2 + B^2) / 3.0 )

These methods are often used for other purposes, such as generating agrayscale difference image between two color images (using -compose 'Difference' composition.

For example The 'MS' (Mean Squared) setting is good for minimizing colorerror comparisions. While... The method 'RMS' (Root Mean Squared) forexample is appropriate for calculating color vector distance, from a colordifference image. This is equivalent to the color only component of the -fuzz factor color compare setting.

See also -grayscale which applies one of the abovegrayscaling formula directly to an image without setting the -intensity setting.

The -colorspace gray image conversion also usesthe current intensity setting, but will always convert the image to theappropriate sRGB or linear-RGB colorspace before appling the abovefunction.

To print a complete list of posible pixel intensity setting methods, use -list intensity.

Operators affected by the -intensity setting include:

-adaptive-blur-adaptive-sharpen-black-threshold-clut (when mapping greyscale CLUT image to alpha channel if set by -channels)-colors for gray colorspace-compose {LightenIntensity, DarkenIntensity, CopyOpacity, CopyBlack}-contrast-stretch-distort {ErodeIntensity, DilateIntensity}-normalize-random-threshold-selective-blur-shade-threshold-tint-white-threshold

-intent type

use this type of rendering intent when managing the image color.

Use this option to affect the color management operation of an image (see-profile). Choose from these intents: Absolute,Perceptual, Relative, Saturation.

The default intent is Perceptual for the sRGB colorspace and undefined for the RGB and gray colorspaces.

To print a complete list of rendering intents, use -list intent.

-interlace type

the type of interlacing scheme.

Choose from:

   none   line   plane   partition   JPEG   GIF   PNG

This option is used to specify the type of interlacing scheme for raw imageformats such as RGB or YUV.

None means do not interlace (RGBRGBRGBRGBRGBRGB...),

Line uses scanline interlacing (RRR...GGG...BBB...RRR...GGG...BBB...), and.

Plane uses plane interlacing (RRRRRR...GGGGGG...BBBBBB...).

Partition is like plane except the different planes are saved toindividual files (e.g. image.R, image.G, and image.B).

Use Line or Plane to create an interlacedPNG or GIF or progressive JPEG image.

To print a complete list of interlacing schemes, use -listinterlace.

-interpolate type

Set the pixel color interpolation method to use when looking up a color based on a floating point or real value.

When looking up the color of a pixel using a non-integer floating pointvalue, you typically fall in between the pixel colors defined by the sourceimage. This setting determines how the color is determined from the colors ofthe pixels surrounding that point. That is how to determine the color of apoint that falls between two, or even four different colored pixels.

   integer           The color of the top-left pixel (floor function)   nearest-neighbor  The nearest pixel to the lookup point (rounded function)   average           The average color of the surrounding four pixels   bilinear          A double linear interpolation of pixels (the default)   mesh              Divide area into two flat triangular interpolations   bicubic           Fitted bicubic-spines of surrounding 16 pixels   spline            Direct spline curves (colors are blurred)   filter            Use resize -filter settings

This most important for distortion operators such as -distort, -implode, -transform and -fx.

To print a complete list of interpolation methods, use -list interpolate.

See also -virtual-pixel, for control of thelookup for positions outside the boundaries of the image.

-interline-spacing value

the space between two text lines.

-interword-spacing value

the space between two words.

-kerning value

the space between two letters.

-kuwahara radius
-kuwahara radiusxsigma

edge preserving noise reduction filter.

The radius is more important than the sigma. If sigma is left off, it will be computed automatically from the radius as sigma=radius-0.5. The sigma provides a bit of additional smoothing control.

-label name

assign a label to an image.

Use this option to assign a specific label to the image, as it is read inor created. You can use the -set operation to re-assigna the labels of images already read in. Image formats such as TIFF, PNG,MIFF, supports saving the label information with the image.

When saving an image to a PostScript file, any labelassigned to an image is used as a header string to print above the postscriptimage.

You can include the image filename, type, width, height, or other imageattribute by embedding special format character. See Format and Print ImageProperties for details of the percent escape codes.

For example,

-label "%m:%f %wx%h" bird.miff

assigns an image label of MIFF:bird.miff 512x480 to the"bird.miff" image and whose width is 512 and height is 480, as itis read in. If a +label option was used instead, anyexisting label present in the image would be used. You can remove all labelsfrom an image by assigning the empty string.

A label is not drawn on the image, but is embedded in the image datastreamvia Label tag or similar mechanism. If you want the label to bevisible on the image itself, use the -draw option, orduring the final processing in the creation of an image montage.

If the first character of string is @, the image label is read from a file titled by theremaining characters in the string. Labels in a file are literal, no embeddedformatting characters are recognized.

-lat width
-lat widthxheight{+-}offset{%}

perform local adaptive threshold.

Adaptively threshold each pixel based on the value of pixels in asurrounding window. If the current pixel is lighter than this average plusthe optional offset, then it is made white, otherwise it is madeblack. Small variations in pixel values such as found in scanned documentscan be ignored if offset is positive. A negative offset will make it moresensitive to those small variations.

This is commonly used to threshold images with an uneven background. It isbased on the assumption that average color of the small window is thethe local background color, from which to separate the foreground color.

-layers method

handle multiple images forming a set of image layers or animation frames.

Perform various image operation methods to a ordered sequence of imageswhich may represent either a set of overlaid 'image layers', a GIF disposalanimation, or a fully-'coalesced' animation sequence.

Method Description
compare-any Crop the second and later frames to the smallest rectangle that contains all the differences between the two images. No GIF -dispose methods are taken into account.
This exactly the same as the -deconstruct operator, and does not preserve animations normal working, especially when animation used layer disposal methods such as 'Previous' or 'Background'.
compare-clear As 'compare-any' but crop to the bounds of any opaque pixels which become transparent in the second frame. That is the smallest image needed to mask or erase pixels for the next frame.
compare-overlay As 'compare-any' but crop to pixels that add extra color to the next image, as a result of overlaying color pixels. That is the smallest single overlaid image to add or change colors.
This can be used with the -compose alpha composition method 'change-mask', to reduce the image to just the pixels that need to be overlaid.
coalesce Equivalent to a call to the -coalesce operator. Apply the layer disposal methods set in the current image sequence to form a fully defined animation sequence, as it should be displayed. Effectively converting a GIF animation into a 'film strip'-like animation.
composite Alpha Composition of two image lists, separated by a "null:" image, with the destination image list first, and the source images last. An image from each list are composited together until one list is finished. The separator image and source image lists are removed.
The -geometry offset is adjusted according to -gravity in accordance of the virtual canvas size of the first image in each list. Unlike a normal -composite operation, the canvas offset is also added to the final composite positioning of each image.
If one of the image lists only contains one image, that image is applied to all the images in the other image list, regardless of which list it is. In this case it is the image meta-data of the list which preserved.
dispose This like 'coalesce' but shows the look of the animation after the layer disposal method has been applied, before the next sub-frame image is overlaid. That is the 'dispose' image that results from the application of the GIF -dispose method. This allows you to check what is going wrong with a particular animation you may be developing.
flatten Create a canvas the size of the first images virtual canvas using the current -background color, and -compose each image in turn onto that canvas. Images falling outside that canvas is clipped. Final image will have a zero virtual canvas offset.
This usually used as one of the final 'image layering' operations overlaying all the prepared image layers into a final image.
For a single image this method can also be used to fillout a virtual canvas with real pixels, or to underlay an opaque color to remove transparency from an image.
merge As 'flatten' method but merging all the given image layers to create a new layer image just large enough to hold all the image without clipping or extra space. The new images virtual offset will preserve the position of the new layer, even if this offset is negative. The virtual canvas size of the first image is preserved.
Caution is advised when handling image layers with negative offsets as few image file formats handle them correctly. Following this operation method with +repage will remove the layer offset, and create an image in which all the overlaid image positions relative to each other is preserved, though not necessarily exactly where you specified them.
See also 'trim-bounds' below which is closely related but without doing the'flatten' to merge the images together.
mosaic As 'flatten' method but expanding the initial canvas size of the first image in a positive direction only so as to hold all the image layers. However as a virtual canvas is 'locked' to the origin, by its own definition, image layers with a negative offsets will still become clipped by the top and left edges. See 'merge' or 'trim-bounds' if this could be a problem.
This method is commonly used to layout individual image using various offset but without knowing the final canvas size. The resulting image will, like 'flatten' not have any virtual offset, so can be saved to any image file format.
optimize Optimize a coalesced animation, into GIF animation using a number of general techniques. This currently a short cut to apply both the 'optimize-frame', and 'optimize-transparency' methods but may be expanded to include other optimization methods as they are developed.
optimize-frame Optimize a coalesced animation, into GIF animation by reducing the number of pixels per frame as much as possible by attempting to pick the best layer disposal method to use, while ensuring the result will continue to animate properly.
There is no guarantee that the best optimization is found. But then no reasonably fast GIF optimization algorithm can do this. However this does seem to do better than most other GIF frame optimizers seen.
optimize-plus As 'optimize-frame' but attempt to improve the overall optimization by adding extra frames to the animation, without changing the final look or timing of the animation. The frames are added to attempt to separate the clearing of pixels from the overlaying of new additional pixels from one animation frame to the next. If this does not improve the optimization (for the next frame only), it will fall back to the results of the previous normal 'optimize-frame' technique.
There is the possibility that the change in the disposal style will result in a worsening in the optimization of later frames, though this is unlikely. In other words there no guarantee that it is better than the normal 'optimize-frame' technique. For some animations however you can get a vast improvement in the final animation size.
optimize-transparency Given a GIF animation, replace any pixel in the sub-frame overlay images with transparency, if it does not change the resulting animation by more than the current -fuzz factor.
This should allow a existing frame optimized GIF animation to compress into a smaller file size due to larger areas of one (transparent) color rather than a pattern of multiple colors repeating the current disposed image of the last frame.
remove-dups Remove (and merge time delays) of duplicate consecutive images, so as to simplify layer overlays of coalesced animations.
Usually this a result of using a constant time delay across the whole animation, or after a larger animation was split into smaller sub-animations. The duplicate frames could also have been used as part of some frame optimization methods.
remove-zero Remove any image with a zero time delay, unless ALL the images have a zero time delay (and is not a proper timed animation, a warning is then issued).
In a GIF animation, such images are usually frames which provide partial intermediary updates between the frames that are actually displayed to users. These frames are usually added for improved frame optimization in GIF animations.
trim-bounds Find the bounds of all the images in the current image sequence, then adjust the offsets so all images are contained on a minimal positive canvas. None of the image data is modified or merged, only the individual image virtual canvas size and offset. All the images is given the same canvas size, and and will have a positive offset, but will remain in the same position relative to each other. As a result of the minimal canvas size at least one image will touch every edge of that canvas. The image data touching those edges however may be transparent.
The result is much like if you used 'merge' followed by a +repage option, except that all the images have been kept separate. If 'flatten' is used after using 'trim-bounds' you will get the same result.

To print a complete list of layer types, use -list layers.

The operators -coalesce, -deconstruct, -flatten, and -mosaic are only aliases for the above methods and may be depreciated inthe future. Also see -page, -repage operators, the -compose setting, and theGIF -dispose and -delaysettings.

-level black_point{,white_point}{%}{,gamma}

adjust the level of image channels.

Given one, two or three values delimited with commas: black-point,white-point, gamma (for example: 10,250,1.0 or 2%,98%,0.5). The black andwhite points range from 0 to QuantumRange, or from 0 to100%; if the white point is omitted it is set to (QuantumRange - black_point), so as to center contrast changes.If a % sign is present anywhere in the string, both black and whitepoints are percentages of the full color range. Gamma will do a -gamma adjustment of the values. If it is omitted, thedefault of 1.0 (no gamma correction) is assumed.

In normal usage (-level) the image values are stretched so thatthe given 'black_point' value in the original image is set to zero(or black), while the given 'white_point' value is set to QuantumRange (or white). This provides you with directcontrast adjustments to the image. The 'gamma' of the resultingimage will then be adjusted.

From ImageMagick v6.4.1-9 using the plus form of the operator(+level) or adding the special '!' flag anywhere in the argumentlist, will cause the operator to do the reverse of the level adjustment. Thatis a zero, or QuantumRange value (black, and white, resp.)in the original image, is adjusted to the given level values, allowing you tode-contrast, or compress the channel values within the image. The'gamma' is adjusted before the level adjustment to de-contrast theimage is made.

Only the channels defined by the current -channelsetting are adjusted (defaults to RGB color channels only), allowing you tolimit the effect of this operator.

Please note that the transparency channel is treated as 'matte'values (0 is opaque) and not as 'alpha' values (0 is transparent).

-level-colors {black_color}{,}{white_color}

adjust the level of an image using the provided dash separated colors.

This function is exactly like -level, except that thevalue value for each color channel is determined by the'black_color' and 'white_color' colors given (asdescribed under the -fill option).

This effectually means the colors provided to -level-colorsis mapped to become 'black' and 'white' respectively, with all the othercolors linearly adjusted (or clipped) to match that change. Each channel isadjusted separately using the channel values of the colors specified.

On the other hand the plus form of the operator (+level-colors)will map the image color 'black' and 'white' to the given colorsrespectively, resulting in a gradient (de-contrasting) tint of the image tothose colors. This can also be used to convert a plain gray-scale image into aone using the gradient of colors specified.

By supplying a single color with a comma separator either before or afterthat color, will just replace the respective 'black' or 'white' pointrespectively. But if no comma separator is provided, the given color isused for both the black and white color points, making the operator eitherthreshold the images around that color (- form) or set all colors to thatcolor (+ form).

-limit type value

Set the pixel cache resource limit.

Choose from: area, disk, file,map, memory, thread, or time.

The value for file is in number of files. The other limits arein bytes. Define arguments for the memory, map, area, and disk resource limitswith SI prefixes (.e.g 100MB).

By default the limits are 768 files, 3GB of image area, 1.5GiB memory, 3GiBmemory map, and 18.45EB of disk. These limits are adjusted relative to theavailable resources on your computer if this information is available. Whenany limit is reached, ImageMagick fails in some fashion but attempts to takecompensating actions, if possible. For example, the following limitsmemory:

-limit memory 32MiB -limit map 64MiB

Use -list resource to list the current limits. For example, our system shows these limits:

$ identify -list resource

Resource limits:  Width: 100MP  Height: 100MP  Area: 25.181GB  Memory: 11.726GiB  Map: 23.452GiB  Disk: unlimited  File: 768  Thread: 12  Throttle: 0  Time: unlimited

Requests for pixel storage to keep intermediate images are satisfied by oneof three resource categories: in-memory pool, memory-mapped files pool, anddisk pool (in that order) depending on the ‑limit settingsand whether the system honors a resource request. If the total size ofallocated pixel storage in the given pool reaches the corresponding limit, therequest is passed to the next pool. Additionally, requests that exceed thearea limit automagically are allocated on disk.

To illustrate how ImageMagick utilizes resource limits, consider a typicalimage resource request. First, ImageMagick tries to allocate the pixels inmemory. The request might be denied if the resource request exceeds thememory limit or if the system does not honor the request. Ifa memory request is not honored, the pixels are allocated to disk and the fileis memory-mapped. However, if the allocation request exceeds themap limit, the resource allocation goes to disk. In all cases, ifthe resource request exceeds the area limit, the pixels areautomagically cached to disk. If the disk has a hard limit, the programfails.

In most cases you simply do not need to concern yourself with resourcelimits. ImageMagick chooses reasonable defaults and most images do not taxyour computer resources. Where limits do come in handy is when you processimages that are large or on shared systems where ImageMagick can consume allor most of the available memory. In this case, the ImageMagick workflow slowsother processes or, in extreme cases, brings the system to a halt. Underthese circumstances, setting limits give some assurances that the ImageMagickworkflow will not interfere with other concurrent uses of the computer. Forexample, assume you have a web interface that processes images uploaded fromthe Internet. To assure ImageMagick does not exceed 10MiB of memory you cansimply set the area limit to 10MiB:

-limit area 10MB

Now whenever a large image is processed, the pixels are automagicallycached to disk instead of memory. This of course implies that large imagestypically process very slowly, simply because pixel processing in memory canbe an order of magnitude faster than on disk. Because your web site usersmight inadvertently upload a huge image to process, you should set a disklimit as well:

-limit area 10MB -limit disk 500MB

Here ImageMagick stops processing if an image requires more than 500MB of disk storage.

In addition to command-line resource limit option, resources can be setwith environment variables. Set theenvironment variables MAGICK_AREA_LIMIT,MAGICK_DISK_LIMIT, MAGICK_FILE_LIMIT,MAGICK_MEMORY_LIMIT, MAGICK_MAP_LIMIT,MAGICK_THREAD_LIMIT, MAGICK_TIME_LIMIT for limits ofimage area, disk space, open files, heap memory, memory map, number of threadsof execution, and maximum elapsed time in seconds respectively.

Inquisitive users can try adding -debug cache totheir commands and then scouring the generated output for references to thepixel cache, in order to determine how the pixel cache was allocated and howresources were consumed. Advanced Unix/Linux users can pipe that outputthrough grep memory|open|destroy|disk for more readable sifting.

For more about ImageMagick's use of resources, see the section CacheStorage and Resource Requirements on the Architecture page.

-linear-stretch black-point
-linear-stretch black-point{xwhite-point}{%}}

Linear with saturation stretch.

This is very similar to -contrast-stretch,and uses a 'histogram bin' to determine the range of color values that needs tobe stretched. However it then stretches those colors using the -level operator.

As such while the initial determination may have 'binning' round offeffects, the image colors are stretched mathematically, rather than using thehistogram bins. This makes the operator more accurate.

note however that a -linear-stretch of'0' does nothing, while a value of '1' does a nearperfect stretch of the color range.

See also -auto-level for a 'perfect'normalization of mathematical images.

This operator is under review for re-development.


the line width for subsequent draw operations.

-liquid-rescale geometry

rescale image with seam-carving.

See Image Geometry for complete details about the geometry argument.

-list type

Print a list of supported arguments for various options or settings. Choose from these list types:
   Align          Alpha          Boolean        Channel   Class          ClipPath       Coder          Color   Colorspace     Command        Compose        Compress   Configure      DataType       Debug          Decoration   Delegate       Direction      Dispose        Distort   Dither         Endian         Evaluate       FillRule   Filter         Font           Format         Function   Gravity        ImageList      Intent         Interlace   Interpolate    Kernel         Layers         LineCap   LineJoin       List           Locale         LogEvent   Log            Magic          Method         Metric   Mime           Mode           Morphology     Module   Noise          Orientation    Policy         PolicyDomain   PolicyRights   Preview        Primitive      QuantumFormat   Resource       SparseColor    Storage        Stretch   Style          Threshold      Type           Units   Validate       VirtualPixel

These lists vary depending on your version of ImageMagick. Use "-listlist" to get a complete listing of all the "-list" argumentsavailable:

$ identify -list list

-log string

Specify format for debug log.

This option specifies the format for the log printed when the -debug option is active.

You can display the following components by embedding special formatcharacters:

   %d  domain   %e  event   %f  function   %l  line   %m  module   %p  process ID   %r  real CPU time   %t  wall clock time   %u  user CPU time   %%  percent sign   \n  newline   \r  carriage return

For example:

$ convert -debug coders -log "%u %m:%l %e" in.gif out.png

The default behavior is to print all of the components.

-loop iterations

add Netscape loop extension to your GIF animation.

Set iterations to zero to repeat the animation an infinite number of times,otherwise the animation repeats itself up to iterationstimes.

-lowlight-color color

when comparing images, de-emphasize pixel differences with this color.


double the size of the image with pixel art scaling.

-map type

Display image using this type.[animate, display]

Choose from these Standard Colormap types:

   best   default   gray   red   green   blue

The X server must support the StandardColormap you choose, otherwise an error occurs. Use list asthe type and display searches the list of colormap types intop-to-bottom order until one is located. See xstdcmap(1) for one way of creating Standard Colormaps.

-map components

pixel map.[stream]

Here are the valid components of a map:

   r  red pixel component   g  green pixel component   b  blue pixel component   a  alpha pixel component (0 is transparent)   o  opacity pixel component (0 is opaque)   i  grayscale intensity pixel component   c  cyan pixel component   m  magenta pixel component   y  yellow pixel component   k  black pixel component   p  pad component (always 0)

You can specify as many of these components as needed in any order (e.g.bgr). The components can repeat as well (e.g. rgbr).


Prevent updates to image pixels specified by the mask.

This the same as using a mask used for composite masking operations, withgrayscale values causing blended updates of the image the mask is attached to.

Use +mask to remove the mask from images.

Also see -clip-mask which work in the same way,but with strict boolean masking.

-mattecolor color

Specify the color to be used with the -frame option.

The color is specified using the format described under the -fill option.

The default matte color is #BDBDBD, this shade of gray.


return the maximum intensity of an image sequence.

Select the 'maximum' value from all the surrounding pixels.

This is legacy option from the method of the samename.

-median geometry

apply a median filter to the image.

Select the 'middle' value from all the surrounding pixels.

This is legacy option from the method of the samename.

-mean-shift widthxheight{+distance{%}}

image noise removal and color reduction/segmentation (e.g. -mean-shift 7x7+10%).

widthxheight is the window size and distance is the color distance measured in the range 0 to 1 or 0 to 100%

The mean shift algorithm is iterative and thus slower the larger the window size. For each pixel, it gets all the pixels in the window centered at the pixel and excludes those that are outside the radius=sqrt((width-1)(height-1)/4) surrounding the pixel. From those pixels, it finds which of them are within the specified squared color distance from the current mean. It then computes a new x,y centroid from those coordinates and a new mean. This new x,y centroid is used as the center for a new window. This process is iterated until it converges and the final mean is then used to replace the original pixel value. It repeats this process for the next pixel, etc, until it processes all pixels in the image. Results are better when using other colorspaces rather than RGB. Recommend YIQ, YUV or YCbCr, which seem to give equivalent results.

-metric type

Output to STDERR a measure of the differences between images according to the type given metric.

Choose from:

  AE     absolute error count, number of different pixels (-fuzz effected)  FUZZ   mean color distance  MAE    mean absolute error (normalized), average channel error distance  MEPP   mean error per pixel (normalized mean error, normalized peak error)  MSE    mean error squared, average of the channel error squared  NCC    normalized cross correlation  PAE    peak absolute (normalized peak absolute)  PHASH  perceptual hash  PSNR   peak signal to noise ratio  RMSE   root mean squared (normalized root mean squared)

Control the 'AE', or absolute count of pixels that are different,with the -fuzz factor (ignore pixels whichonly changed by a small amount). Use 'PAE' to find thesize of the -fuzz factor needed to make all pixels'similar', while 'MAE' determines the factor neededfor about half the pixels to be similar.

The 'MEPP' metric returns three different metrics('MAE', 'MAE' normalized, and 'PAE'normalized) from a single comparison run.

To print a complete list of metrics, use the -listmetric option.


return the minimum intensity of an image sequence.

Select the 'minimal' value from all the surrounding pixels.

This is legacy option from the method of the samename.

-mode geometry

make each pixel the 'predominant color' of the neighborhood.[convert, mogrify]

-mode value

Mode of operation.[montage]

Choose the value from these styles: Frame,Unframe, or Concatenate

Use the -list option with a 'Mode' argumentfor a list of -mode arguments available in yourImageMagick installation.

-modulate brightness[,saturation,hue]

Vary the brightness, saturation, and hue of an image.

The arguments are given as a percentages of variation. A value of 100 meansno change, and any missing values are taken to mean 100.

The brightness is a multiplier of the overallbrightness of the image, so 0 means pure black, 50 is half as bright, 200 istwice as bright. To invert its meaning -negate the imagebefore and after.

The saturation controls the amount of color in animage. For example, 0 produce a grayscale image, while a large value such as200 produce a very colorful, 'cartoonish' color.

The hue argument causes a "rotation" of the colorswithin the image by the amount specified. For example, 50 results ina counter-clockwise rotation of 90, mapping red shades to purple, and so on.A value of either 0 or 200 results in a complete 180 degree rotation of theimage. Using a value of 300 is a 360 degree rotation resulting in no change tothe original image.

For example, to increase the color brightness by 20% and decrease the colorsaturation by 10% and leave the hue unchanged, use -modulate 120,90.

Use -set attribute of 'option:modulate:colorspace' to specify which colorspace tomodulate. Choose from HCL, HCLp, HSB, HSI, HSL (the default), HSV, HWB, or LCH (LCHuv). For example,

$ convert image.png -set option:modulate:colorspace hsb -modulate 120,90 modulate.png


report image moments and perceptual hash.[identify]


monitor progress.


transform the image to black and white.

-morph frames

morphs an image sequence.

Both the image pixels and size are linearly interpolated to give theappearance of a metamorphosis from one image to the next, over all the imagesin the current image list. The added images are the equivalent of a -blend composition. The framesargument determine how many images to interpolate between each image.


-morphology method kernel

apply a morphology method to the image.

Until I get around to writing an option summary for this, see IM Usage Examples,Morphology.


an simple alias for the -layers method "mosaic"

-motion-blur radius
-motion-blur radiusxsigma+angle

simulate motion blur.

Blur with the given radius, standard deviation (sigma), and angle. Theangle given is the angle toward which the image is blurred. That is thedirection people would consider the object is coming from.

Note that the blur is not uniform distribution, giving the motion adefinite sense of direction of movement.

The -virtual-pixel setting will determine howpixels which are outside the image proper are blurred into the final result.


name an image.


replace each pixel with its complementary color.

The red, green, and blue intensities of an image are negated. White becomesblack, yellow becomes blue, etc. Use +negate to onlynegate the grayscale pixels of the image.

-noise geometry
+noise type

Add or reduce noise in an image.

The principal function of noise peak elimination filter is to smooth theobjects within an image without losing edge information and without creatingundesired structures. The central idea of the algorithm is to replace a pixelwith its next neighbor in value within a pixel window, if this pixel has beenfound to be noise. A pixel is defined as noise if and only if this pixel isa maximum or minimum within the pixel window.

Use -noise radius tospecify the width of the neighborhood when reducing noise. This is equivalentto using a -statistic NonPeak operation,which should be used in preference.

Use +noise followed by a noise type to add noise to an image. Choose from these noisetypes:

   Gaussian   Impulse   Laplacian   Multiplicative   Poisson   Random   Uniform

The amount of noise added can be controled by the -attenuate setting. If unset the value isequivalent to 1.0, or a maximum noise addition.

Note that Random will replace the image with noise rather than add noise to the image. Use Uniform, if you wish to add random noise to the image.

To print a complete list of noises, use the -list noise option.

Also see the -evaluate noise functions that allowsthe use of a controlling value to specify the amount of noise that should beadded to an image.


Increase the contrast in an image by stretching the range of intensity values.

The intensity values are stretched to cover the entire range of possiblevalues. While doing so, black-out at most 2% of the pixels andwhite-out at most 1% of the pixels.

Note that as of ImageMagick 6.4.7-0, -normalizeis equivalent to -contrast-stretch 2%x1%.(Before this version, it was equivalent to -contrast-stretch 2%x99%).

All the channels are normalized in concert by the came amount so as topreserve color integrity, when the default +channelsetting is in use. Specifying any other -channelsetting will normalize the RGB channels independently.

See -contrast-stretch for more details.Also see -auto-level for a 'perfect' normalizationthat is better suited to mathematically generated images.

This operator is under review for re-development.

-opaque color

change this color to the fill color within the image.

The color argument is defined using the formatdescribed under the -fill option. The -fuzz setting can be used to match and replace colors similar to the onegiven.

Use +opaque to paint any pixel that does not matchthe target color.

The -transparent operator is exactly the sameas -opaque but replaces the matching color withtransparency rather than the current -fill color setting.To ensure that it can do this it also ensures that the image has an alphachannel enabled, as per "-alpha set", forthe new transparent colors, and does not require you to modify the -channel to enable alpha channel handling.

-ordered-dither threshold_map{,level...}

dither the image using a pre-defined ordered dither threshold map specified, and a uniform color map with thegiven number of levels per color channel .

You can choose from these standard threshold maps:

   threshold   1x1   Threshold 1x1 (non-dither)   checks      2x1   Checkerboard 2x1 (dither)   o2x2        2x2   Ordered 2x2 (dispersed)   o3x3        3x3   Ordered 3x3 (dispersed)   o4x4        4x4   Ordered 4x4 (dispersed)   o8x8        8x8   Ordered 8x8 (dispersed)   h4x4a       4x1   Halftone 4x4 (angled)   h6x6a       6x1   Halftone 6x6 (angled)   h8x8a       8x1   Halftone 8x8 (angled)   h4x4o             Halftone 4x4 (orthogonal)   h6x6o             Halftone 6x6 (orthogonal)   h8x8o             Halftone 8x8 (orthogonal)   h16x16o           Halftone 16x16 (orthogonal)   c5x5b       c5x5  Circles 5x5 (black)   c5x5w             Circles 5x5 (white)   c6x6b       c6x6  Circles 6x6 (black)   c6x6w             Circles 6x6 (white)   c7x7b       c7x7  Circles 7x7 (black)   c7x7w             Circles 7x7 (white)

The threshold generated a simple 50% threshold of the image.This could be used with level to do the equivalent of -posterize to reduce an image to basic primary colors.

The checks pattern produces a 3 level checkerboard ditherpattern. That is a grayscale will become a pattern of solid black, solidwhite, and mid-tone colors into a checkerboard pattern of black and white.

You can define your own threshold map for ordereddithering and halftoning your images, in either personal or systemthresholds.xml XML file. See Resourcesfor more details of configuration files.

To print a complete list of the thresholds that have been defined, use the-list threshold option.

Note that at this time the same threshold dithering map is used for allcolor channels, no attempt is made to offset or rotate the map for differentchannels is made, to create an offset printing effect. Also as the maps aresimple threshold levels, the halftone and circle maps will create incompletecircles along the edges of a colored area. Also all the effects are purelyon/off boolean effects, without anti-aliasing to make the circles smoothlooking. Large dots can be made to look better with a small amount of blurringafter being created.

-orient image orientation

specify orientation of a digital camera image.

Choose from these orientations:

   bottom-left   bottom-right   left-bottom   left-top   right-bottom   right-top   top-left   top-right   undefined

To print a complete list of orientations, use the -listorientation option.

-page geometry
-page media[offset][{^!<>}]

Set the size and location of an image on the larger virtual canvas.

See Image Geometry for complete details about the geometry argument.

For convenience you can specify the page size using media (see below). Offsets can then be added as with othergeometry arguments (e.g. -page Letter+43+43).

Use media as shorthand to specify the dimensions (widthxheight) of the PostScript page in dots per inch or a TEXT page in pixels.The choices for a PostScript page are:

media width height
11x17 792 1224
Ledger 1224 792
Legal 612 1008
Letter 612 792
LetterSmall 612 792
ArchE 2592 3456
ArchD 1728 2592
ArchC 1296 1728
ArchB 864 1296
ArchA 648 864
A0 2380 3368
A1 1684 2380
A2 1190 1684
A3 842 1190
A4 595 842
A4Small 595 842
A5 421 595
A6 297 421
A7 210 297
A8 148 210
A9 105 148
A10 74 105
B0 2836 4008
B1 2004 2836
B2 1418 2004
B3 1002 1418
B4 709 1002
B5 501 709
C0 2600 3677
C1 1837 2600
C2 1298 1837
C3 918 1298
C4 649 918
C5 459 649
C6 323 459
Flsa 612 936
Flse 612 936
HalfLetter 396 612

This option is also used to place subimages when writing to a multi-imageformat that supports offsets, such as GIF89 and MNG. When used for thispurpose the offsets are always measured from the top left corner of the canvasand are not affected by the -gravity option. Toposition a GIF or MNG image, use -page{+-}x{+-}y (e.g. -page +100+200). When writing to a MNGfile, a -page option appearing ahead of the first image inthe sequence with nonzero width and height defines the width and height valuesthat are written in the MHDR chunk. Otherwise, the MNG width andheight are computed from the bounding box that contains all images in thesequence. When writing a GIF89 file, only the bounding box method is used todetermine its dimensions.

For a PostScript page, the image is sized as in -geometry but positioned relative to the lowerleft-hand corner of the page by {+-}xoffset{+-}y offset. Use -page 612x792, for example, to center the image within thepage. If the image size exceeds the PostScript page, it is reduced to fit thepage. The default gravity for the -page option is NorthWest, i.e., positive x and y offset are measured rightward and downward from the top leftcorner of the page, unless the -gravity option ispresent with a value other than NorthWest.

The default page dimensions for a TEXT image is 612x792.

This option is used in concert with -density.

Use +page to remove the page settings for an image.

-paint radius

simulate an oil painting.

Each pixel is replaced by the most frequent color in a circularneighborhood whose width is specified with radius.

-path path

write images to this path on disk.

-pause seconds

Pause between animation loops.[animate]

Pause for the specified number of seconds before repeating the animation.

-pause seconds

Pause between snapshots.[import]

Pause for the specified number of seconds before taking the next snapshot.

-perceptible epsilon

set each pixel whose value is less than |epsilon| to -epsilon or epsilon (whichever is closer) otherwise the pixel value remains unchanged.


efficiently determine image characteristics.

-pointsize value

pointsize of the PostScript, OPTION1, or TrueType font.

-polaroid angle

simulate a Polaroid picture.

Use +polaroid to rotate the image at a random angle between -15 and +15 degrees.

-poly "wt,exp ..."

combines multiple images according to a weighted sum of polynomials; one floating point weight (coefficient) and one floating point polynomial exponent (power) for each image expressed as comma separated pairs.

The weights should typically be fractions between -1 and 1. But the sum of weights should be 1 or at least between 0 and 1 to avoid clamping in non-hdri mode at black and white.

The exponents may be positive, negative or zero. A negative exponent is equivalent to 1 divided by the image raised to the corresponding positive exponent. A zero exponent always produces 1 scaled by quantumrange to white, i.e. wt*white, no matter what the image.

The format is: output = wt1*image1^exp1 + wt2*image2^exp2 ...

Some simple uses are:

  • A weighted sum of each image provided all weights add to unity and all exponents=1. If the the weights are all equal to 1/(number of images), then this is equivalent to -evaluate-sequence mean.
  • The sum of squares of two or more images, provided the weights are equal (and sum to 1 to avoid clamping) and the exponents equal 2.

Note that one may add a constant color to the expression simply by using xc:somecolor for one of the images and specifying the desired weight and exponent equal to 0.

Similarly one may add white to the expression by simply using null: (or xc:white) for one of the images with the appropriate weight and exponent equal to 0.

-posterize levels

reduce the image to a limited number of color levels per channel.

Very low values of levels, e.g., 2, 3, 4, have the mostvisible effect.

-precision value

set the maximum number of significant digits to be printed.

-preview type

image preview type.

Use this option to affect the preview operation of an image (e.g.convert file.png -preview Gamma Preview:gamma.png). Choose fromthese previews:

   Rotate           Shear            Roll             Hue   Saturation       Brightness       Gamma            Spiff   Dull             Grayscale        Quantize         Despeckle   ReduceNoise      Add Noise        Sharpen          Blur   Threshold        EdgeDetect       Spread           Shade   Raise            Segment          Solarize         Swirl   Implode          Wave             OilPaint         CharcoalDrawing   JPEG

To print a complete list of previews, use the -list preview option.

The default preview is JPEG.

-print string

interpret string and print to console.

-process command

process the image with a custom image filter.

The command arguments has the form "module arg1 arg2 arg3 ...argN" where module is the name of the module to invoke (e.g."Analyze") and arg1 arg2 arg3 ... argN are an arbitrary number of arguments topass to the process module.

-profile filename
+profile profile_name

Manage ICM, IPTC, or generic profiles in an image.

Using -profile filename adds anICM (ICC color management), IPTC (newswire information), or a generic profileto the image.

Use +profile profile_name toremove the indicated profile. ImageMagick uses standard filename globbing, sowildcard expressions may be used to remove more than one profile. Here weremove all profiles from the image except for the XMP profile: +profile"!xmp,*".

Use identify -verbose to find out which profiles are in theimage file. Use -strip to remove all profiles (andcomments).

To extract a profile, the -profile option is notused. Instead, simply write the file to an image format such as APP1, 8BIM, ICM, or IPTC.

For example, to extract the Exif data (which is stored in JPEG files in theAPP1 profile), use.

$ convert cockatoo.jpg profile.exif

It is important to note that results may depend on whether or not theoriginal image already has an included profile. Also, keep in mind that -profile is an "operator" (as opposed to a "setting") andtherefore a conversion is made each time it is encountered, in order, in thecommand-line. For instance, in the following example, if the original image isCMYK with profile, a CMYK-CMYK-RGB conversion results.

$ convert CMYK.tif -profile "CMYK.icc" -profile "RGB.icc" RGB.tiff

Furthermore, since ICC profiles are not necessarily symmetric, extraconversion steps can yield unwanted results. CMYK profiles are often veryasymmetric since they involve 3−>4 and 4−>3 channel mapping.

-quality value

JPEG/MIFF/PNG compression level.

For the JPEG and MPEG image formats, quality is 1 (lowest image quality andhighest compression) to 100 (best quality but least effective compression).The default is to use the estimated quality of your input image if it canbe determined, otherwise 92. When the quality is greater than 90, then thechroma channels are not downsampled.Use the -sampling-factor option to specify thefactors for chroma downsampling.

For the MIFF image format, quality/10 is the zlib compression level, whichis 0 (worst but fastest compression) to 9 (best but slowest). It has no effecton the image appearance, since the compression is always lossless.

For the JPEG-2000 image format, quality is mapped using a non-linearequation to the compression ratio required by the Jasper library. Thisnon-linear equation is intended to loosely approximate the quality provided bythe JPEG v1 format. The default quality value 100, a request for non-lossycompression. A quality of 75 results in a request for 16:1 compression.

For the MNG and PNG image formats, the quality value sets the zlibcompression level (quality / 10) and filter-type (quality % 10). The defaultPNG "quality" is 75, which means compression level 7 with adaptive PNGfiltering, unless the image has a color map, in which case it meanscompression level 7 with no PNG filtering.

For compression level 0 (quality value less than 10), the Huffman-onlystrategy is used, which is fastest but not necessarily the worstcompression.

If filter-type is 4 or less, the specified PNG filter-type is used forall scanlines:

   0: none   1: sub   2: up   3: average   4: Paeth

If filter-type is 5, adaptive filtering is used when quality is greaterthan 50 and the image does not have a color map, otherwise no filtering isused.

If filter-type is 6, adaptive filteringwith minimum-sum-of-absolute-values is used.

Only if the output is MNG, if filter-type is 7, the LOCO colortransformation (intrapixel differencing) and adaptive filteringwith minimum-sum-of-absolute-values are used.

If the filter-type is 8 the zlib Z_RLE compression strategy (or theZ_HUFFMAN_ONLY strategy, when compression level is 0) is used withadaptive PNG filtering.

If the filter-type is 9 the zlib Z_RLE compression strategy (or theZ_HUFFMAN_ONLY strategy, when compression level is 0) is used withno PNG filtering.

The quality setting has no effect on the appearance or signature of PNGand MNG images, since the compression is always lossless.

Not all combinations of compression level, strategy, and PNG filter typecan be obtained using the -quality option. For more precise control,you can use the PNG:compression-level=N, PNG:compression-strategy=N, andPNG:compression-filter=N defines, respectively, instead.See -define. Values from the defines take precedenceover values from the -quality option.

For further information, seethe PNG specification.

-quantize colorspace

reduce colors using this colorspace.

This setting defines the colorspace used to sort out and reduce the numberof colors needed by an image (for later dithering) by operators such as -colors, Note that color reduction also happensautomatically when saving images to color-limited image file formats, such asGIF, and PNG8.


suppress all warning messages. Error messages are still reported.

-radial-blur angle

Blur around the center of the image.

Note that this is actually a rotational blur rather than a radial and assuch actually mis-named.

The -virtual-pixel setting will determine howpixels which are outside the image proper are blurred into the final result.

-raise thickness

Lighten or darken image edges.

This will create a 3-D effect. Use -raise to createa raised effect, otherwise use +raise.

Unlike the similar -frame option, -raise does not alter the dimensions of the image.

-random-threshold lowxhigh

Apply a random threshold to the image.

-red-primary x,y

Set the red chromaticity primary point.


Pay attention to warning messages.

This option causes some warnings in some image formats to be treatedas errors.

-remap filename

Reduce the number of colors in an image to the colors used by this image.

If the -dither setting is enabled (the default) thenthe given colors are dithered over the image as necessary, otherwise the closestcolor (in RGB colorspace) is selected to replace that pixel in the image.

As a side effect of applying a -remap of colors across allimages in the current image sequence, all the images will have the same colortable. That means that when saved to a file format such as GIF, it will usethat color table as a single common or global color table, for all the images,without requiring extra local color tables.

Use +remap to reduce all images in the current imagesequence to use a common color map over all the images. This equivalent toappending all the images together (without extra background colors) and colorreducing those images using -colors with a 256 colorlimit, then -remap those colors over the original list ofimages. This ensures all the images follow a single color map.

If the number of colors over all the images is less than 256, then +remap should not perform any color reduction or dithering, asno color changes are needed. In that case, its only effect is to force the useof a global color table. This recommended after using either -colors or -ordered-dither toreduce the number of colors in an animated image sequence.

Note, the remap image colormap has at most 8-bits of precision. Deeper color maps are automagically coalesced with other colors to meet this requirement.

-region geometry

Set a region in which subsequent operations apply.

The x and y offsets are treatedin the same manner as in -crop.

See Image Geometry for complete details about the geometry argument.


perform a remote operation.

The only command recognized is the name of an image file to load.

If you have more than one display applicationrunning simultaneously, use the window option tospecify which application to control.


render vector operations.

Use +render to turn off rendering vector operations.This useful when saving the result to vector formats such as MVG or SVG.

-repage geometry

Adjust the canvas and offset information of the image.

This option is like -page but acts as an image operatorrather than a setting. You can separately set the canvas size or the offsetof the image on that canvas by only providing those components.

See Image Geometry for complete details about the geometry argument.

If a ! flag is given the offset given is added to the existingoffset to move the image relative to its previous position. This useful foranimation sequences.

A given a canvas size of zero such as '0x0' forces it torecalculate the canvas size so the image (at its current offset) will appearcompletely on that canvas (unless it has a negative offset).

Use +repage to completely remove/reset the virtualcanvas meta-data from the images.

The -set 'page' option can be used todirectly assign virtual canvas meta-data.

-resample horizontalxvertical

Resample image to specified horizontal and vertical resolution.

Resize the image so that its rendered size remains the same as the originalat the specified target resolution. For example, if a 300 DPI image renders at3 inches by 2 inches on a 300 DPI device, when the image has been resampled to72 DPI, it will render at 3 inches by 2 inches on a 72 DPI device. Note thatonly a small number of image formats (e.g. JPEG, PNG, and TIFF) are capable ofstoring the image resolution. For formats which do not support an imageresolution, the original resolution of the image must be specified via -density on the command line prior to specifying theresample resolution.

Note that Photoshop stores and obtains image resolution from a proprietaryembedded profile. If this profile exists in the image, then Photoshop willcontinue to treat the image using its former resolution, ignoring the imageresolution specified in the standard file header.

-resize geometry

Resize an image.

See Image Geometry for complete details about the geometry argument. Offsets, if present in the geometry string, areignored, and the -gravity option has no effect.

If the -filter optionor -define filter:option=value precedes the -resize option, the image is resized with the specifiedfilter.

Many image processing algorithms assume your image is in a linear-lightcoding. If your image is gamma-corrected, you can remove the nonlinear gammacorrection, apply the transform, then restore it like this:

$ convert portrait.jpg -gamma .45455 -resize 25% -gamma 2.2 \
-quality 92 passport.jpg


settings remain in effect until parenthesis boundary.


Reverse the order of images in the current image list.

-roll {+-}x{+-}y

roll an image vertically or horizontally by the amount given.

A negative x offset rolls the image right-to-left.A negative y offset rolls the image bottom-to-top.

-rotate degrees{<}{>}

Apply Paeth image rotation (using shear operations) to the image.

Use > to rotate the image only if its width exceeds theheight. < rotates the image only if its width is lessthan the height. For example, if you specify -rotate "-90>" andthe image size is 480x640, the image is not rotated. However, if the image is640x480, it is rotated by -90 degrees. If you use > or<, enclose it in quotation marks to prevent it from beingmisinterpreted as a file redirection.

Empty triangles in the corners, left over from rotating the image, arefilled with the background color.

See also the -distort operator and specifically the'ScaleRotateTranslate' distort method.

-sample geometry

minify / magnify the image with pixel subsampling and pixel replication, respectively.

Change the image size simply by directly sampling the pixels originalfrom the image. When magnifying, pixels are replicated in blocks. Whenminifying, pixels are sub-sampled (i.e., some rows and columns are skippedover).

The results are thus equivalent to using -resize witha -filter setting of point (nearestneighbour), though -sample is a lot faster, as itavoids all the filter processing of the image. As such it completely ignoresthe current -filter setting.

The key feature of the -sample is that no new colorswill be added to the resulting image, though some colors may disappear.

See Image Geometry for complete details about the geometry argument. Offsets, if present in the geometry string, areignored, unlike -resize.

The actual sampling point is the middle of the sub-region being sampled.As such a single pixel sampling of an image will take the middle pixel, (ortop-left-middle if image has even dimensions). However the -define 'sample:offset' can be set to modifythis position some other location within each sub-region being sampled, asa percentage offset.

By default this value is '50' for the midpoint, but could be setto '0' for top-left, '100' for bottom-right, or withseparate X and Y offsets such as '0x50' for left-middle edge ofsampling sub-region.

-sampling-factor horizontal-factorxvertical-factor

sampling factors used by JPEG or MPEG-2 encoder and YUV decoder/encoder.

This option specifies the sampling factors to be used by the JPEG encoderfor chroma downsampling. If this option is omitted, the JPEG library will useits own default values. When reading or writing the YUV format and whenwriting the M2V (MPEG-2) format, use -sampling-factor 2x1 or -sampling-factor 4:2:2 to specify the 4:2:2downsampling method.

-scale geometry

minify / magnify the image with pixel block averaging and pixel replication, respectively.

Change the image size simply by replacing pixels by averaging pixelstogether when minifying, or replacing pixels when magnifying.

The results are thus equivalent to using -resize witha -filter setting of box. Though it is a lotfaster, as it avoids all the filter processing of the image. As such itcompletely ignores the current -filter setting.

If when shrinking (minifying) images the original image is some integermultiple of the new image size, the number of pixels averaged together toproduce the new pixel color is the same across the whole image. This isa special case known as 'binning' and is often used as a method of reducingnoise in image such as those generated by digital cameras, especially in lowlight conditions.

-scene value

set scene number.

This option sets the scene number of an image or the first image in an image sequence.


specify the screen to capture.

This option indicates that the GetImage request used to obtain the imageshould be done on the root window, rather than directly on the specifiedwindow. In this way, you can obtain pieces of other windows that overlap thespecified window, and more importantly, you can capture menus or other popupsthat are independent windows but appear over the specified window.


seed a new sequence of pseudo-random numbers

-segment cluster-thresholdxsmoothing-threshold

segment the colors of an image.

Segment an image by analyzing the histograms of the color components andidentifying units that are homogeneous with the fuzzy c-means technique. Thisis part of the ImageMagick color quantization routines.

Specify cluster threshold as the number of pixels ineach cluster that must exceed the cluster threshold to be considered valid.Smoothing threshold eliminates noise in the secondderivative of the histogram. As the value is increased, you can expecta smoother second derivative. The default is 1.5.

If the -verbose setting is defined, a detailed reportof the color clusters is returned.

-selective-blur radius
-selective-blur radiusxsigma{+threshold}

Selectively blur pixels within a contrast threshold.

Blurs those pixels that are less than or equal to the threshold incontrast. The threshold may be expressed as a fraction of QuantumRange or as a percentage.


separate an image channel into a grayscale image. Specify the channel with -channel.

-sepia-tone threshold

simulate a sepia-toned photo.

Specify threshold as the percent threshold of the intensity (0 - 99.9%).

This option applies a special effect to the image, similar to the effectachieved in a photo darkroom by sepia toning. Threshold ranges from 0 to QuantumRange and is a measure of the extent of the sepiatoning. A threshold of 80% is a good starting point for a reasonabletone.

-set key value

+set key

sets image attributes and properties for images in the currentimage sequence.

This will assign (or modify) specific settings attached to all the imagesin the current image sequence. Using the +set form of theoption will either remove, or reset that setting to a default state, asappropriate.

For example, it will modify specific well known image meta-data'attributes' such as those normally overridden by: the options -delay, -dispose, and -page, -colorspace; generallyassigned before the image is read in, by using a key ofthe same name.

If the given key does not match a specific known'attribute ', such as shown above, the setting is stored as a a free form'property' string. Such settings are listed in -verbose information ("info:" output format) as "Properties".

This includes string 'properties' that are set by and assigned to imagesusing the options -comment, -label, -caption. These options actually assigna global 'artifact' which are automatically assigned (and any Format PercentEscapes expanded) to images as they are read in. For example:

$ convert rose: -set comment 'Rose is a rose is a rose is a rose' rose.pngidentify -format %c rose.png $ Rose is a rose is a rose is a rose

The set value can also make use of Format and Print ImageProperties in the defined value. For example:

$ convert rose: -set origsize '%wx%h' -resize 50% \ -format 'Old size = %[origsize] New size = %wx%h' info: $ Old size = 70x46 New size = 35x23

Other well known 'properties' that are availible include:'date:create' and 'date:modify' and'signature'.

The -repage operator will also allow you to modifythe 'page' attribute of an image for images already in memory (alsosee -page). However it is designed to provide a finercontrol of the sub-parts of this 'attribute'. The -set pageoption will only provide a direct, unmodified assignment of 'page'attribute.

This option can also associate a colorspace or profile with your image.For example,

$ convert image.psd -set profile ISOcoated_v2_eci.icc image-icc.psd

Some 'properties' must be defined in a specific way to be used. Forexample only 'properties' prefixed with "filename:" can be used tomodify the output filename of an image. For example

$ convert rose: -set filename:mysize '%wx%h' 'rose_%[filename:mysize].png'

If the setting value is prefixed with "option:" the setting willbe saved as a global "Artifact" exactly as if it was set using the -define option. As such settings are global in scope, theycan be used to pass 'attributes' and 'properties' of one specific image,in a way that allows you to use them in a completely different image, even ifthe original image has long since been modified or destroyed. For example:

$ convert rose: -set option:rosesize '%wx%h' -delete 0 \ label:'%[rosesize]' label_size_of_rose.gif

Note that Format Percent Escapes will only matcha 'artifact' if the given key does not match an existing'attribute' or 'property'.

You can set the attributes of the image registry by prefixing the valuewith registry:.

The -set profile option can also be used to injectpreviously-formatted ancillary chunks into the output PNG file, usingthe commandline option as shown below or by setting the profile via aprogramming interface:

$ convert in.png -set profile PNG-chunk-x:<filename> out.png

where x is a location flag andfilename is a file containing the chunkname in the first 4 bytes, then a colon (":"), followed by the chunk data.This encoder will compute the chunk length and CRC, so those must notbe included in the file.

"x" can be "b" (before PLTE), "m" (middle, i.e., between PLTE and IDAT),or "e" (end, i.e., after IDAT). If you want to write multiple chunksof the same type, then add a short unique string after the "x" to preventsubsequent profiles from overwriting the preceding ones, e.g.,

$ convert in.png -set profile PNG-chunk-b01:file01 /
-profile PNG-chunk-b02:file02 out.png

-shade azimuthxelevation

shade the image using a distant light source.

Specify azimuth and elevation asthe position of the light source. Use +shade to returnthe shading results as a grayscale image.

-shadow percent-opacity{xsigma}{+-}x{+-}y{%}

simulate an image shadow.


use shared memory.

This option specifies whether the utility should attempt to use sharedmemory for pixmaps. ImageMagick must be compiled with shared memory support,and the display must support the MIT-SHM extension.Otherwise, this option is ignored. The default is True.

-sharpen radius
-sharpen radiusxsigma

sharpen the image.

Use a Gaussian operator of the given radius and standard deviation (sigma).

-shave geometry

Shave pixels from the image edges.

The size portion of the geometryargument specifies the width of the region to be removed from both sides ofthe image and the height of the regions to be removed from top and bottom.Offsets are ignored.

See Image Geometry for complete details about the geometry argument.

-shear Xdegrees[xYdegrees]

Shear the image along the x-axis and/or y-axis.

The shear angles may be positive, negative, or zero. When Ydegrees is omitted it defaults to 0. When both angles aregiven, the horizontal component of the shear is performed before the verticalcomponent.

Shearing slides one edge of an image along the x-axis or y-axis (i.e.,horizontally or vertically, respectively),creating a parallelogram. The amountof each is controlled by the respective shear angle. For horizontal shears,Xdegrees is measured clockwise relative to "up" (thenegative y-axis), sliding the top edge to the right when 0°<Xdegrees<90° and to the left when 90°<Xdegrees<180°. For vertical shears Ydegrees is measured clockwise relative to "right" (thepositive x-axis), sliding the right edge down when 0°<Ydegrees<90° and up when 90°<Ydegrees<180°.

Empty triangles left over from shearing the image are filled with the colordefined by the -background option. The color is specifiedusing the format described under the -fill option.

The horizontal shear is performed before the vertical part. This isimportant to note, since horizontal and vertical shears do notcommute, i.e., the order matters in a sequence of shears. Forexample, the following two commands are not equivalent.

$ convert logo: -shear 20x0 -shear 0x60 logo-sheared.png $ convert logo: -shear 0x60 -shear 20x0 logo-sheared.png

The first of the two commands above is equivalent to the following, exceptfor the amount of empty space created; the command that follows generatesa smaller image, and so is a better choice in terms of time and space.

$ convert logo: -shear 20x60 logo-sheared.png

-sigmoidal-contrast contrastxmid-point

increase the contrast without saturating highlights or shadows.

Increase the contrast of the image using a sigmoidal transfer functionwithout saturating highlights or shadows. Contrastindicates how much to increase the contrast. For example, 0 is none, 3 istypical and 20 is a lot.

The mid-point indicates where the maximum change'slope' in contrast should fall in the resultant image (0 is white; 50% ismiddle-gray; 100% is black).

By default the image contrast is increased, use +sigmoidal-contrast to decrease the contrast.

To achieve the equivalent of a sigmoidal brightness change (similar to a gamma adjustment), you would use -sigmoidal-contrast{brightness}x0% to increase brightness and +sigmoidal-contrast {brightness}x0% to decrease brightness.Note the use of '0' fo rthe mid-point of the sigmoidal curve.

Using a very high contrast will produce a sort of'smoothed thresholding' of the image. Not as sharp (with high aliasingeffects) of a true threshold, but with tapered gray-levels around the thresholdmid-point.


operate silently.

-similarity-threshold value

minimum RMSE for subimage match.[compare]

If this setting is used, then the search will stop as soon as it finds a match whose metric is less than or equal to the value. A partially filled second output image will result. Using a value of zero, will cause the search to stop at the first perfect match it finds. If this setting is left off, then the search will proceed to completion or as limited by -dissimilarity-threshold.

-size width[xheight][+offset]

set the width and height of the image.

Use this option to specify the width and height of raw images whosedimensions are unknown such as GRAY, RGB, orCMYK. In addition to width and height, use -size with an offset to skip any header information in theimage or tell the number of colors in a MAP image file, (e.g. -size640x512+256).

For Photo CD images, choose from these sizes:

   192x128   384x256   768x512   1536x1024   3072x2048

-sketch radius
-sketch radiusxsigma+angle

simulate a pencil sketch.

Sketch with the given radius, standard deviation (sigma), and angle. Theangle given is the angle toward which the image is sketched. That is thedirection people would consider the object is coming from.

-smush offset

smush an image sequence together.

-snaps value

Set the number of screen snapshots.[import]

Use this option to grab more than one image from the X server screen, to create an animation sequence.

-solarize threshold

negate all pixels above the threshold level.

Specify factor as the percent threshold of the intensity (0 - 99.9%).

This option produces a solarization effect seen whenexposing a photographic film to light during the development process.

-sparse-color method 'x,y color ...'

color the given image using the specified points of color, and filling the other intervening colors using the given methods.
Method Description
barycentric three point triangle of color given 3 points. Giving only 2 points will form a linear gradient between those points. The gradient generated extends beyond the triangle created by those 3 points.
bilinear Like barycentric but for 4 points. Less than 4 points fall back to barycentric.
voronoi Simply map each pixel to the to nearest color point given. The result are polygonal 'cells' of solid color.
shepards Colors points biased on the ratio of inverse distance squared. Generating spots of color in a sea of the average of colors.
inverse Colors points biased on the ratio of inverse distance. This generates sharper points of color rather than rounded spots of 'shepards' Generating spots of color in a sea of the average of colors.

The points are placed according to the images location on the virtualcanvas (-page or -repageoffset), and do not actually have to exist on the given image, but may besome point beyond the edge of the image. All points are floating point values.

Only the color channels defined by the -channel aremodified, which means that by default matte/alpha transparency channel is noteffected. Typically transparency channel is turned off either before or afterthe operation.

Of course if some color points are transparent to generate a transparentgradient, then the image also requires transparency enabled to store thevalues.

All the above methods when given a single point of color will replace allthe colors in the image with the color given, regardless of the point. This islogical, and provides an alternative technique to recolor an image to somedefault value.

-splice geometry

Splice the current background color into the image.

This will add rows and columns of the current -background color into the given image according to thegiven -gravity geometry setting. >See Image Geometry for complete details about the geometry argument. Essentially -splice will divide theimage into four quadrants, separating them by the inserted rows and columns.

If a dimension of geometry is zero no rows or columns will be added for thatdimension. Similarly using a zero offset with the appropriate -gravity setting will add rows and columns to the edges ofthe image, padding the image only along that one edge. Edge padding is what -splice is most commonly used for.

If the exact same geometry and -gravity is later used with -chop theadded added all splices removed.

-spread amount

displace image pixels by a random amount.

The argument amount defines the size of theneighborhood around each pixel from which to choose a candidate pixel toswap.

-statistic type geometry

replace each pixel with corresponding statistic from the neighborhood.[convert, mogrify]

Choose from these statistic types:

   Gradient            maximum difference (max - min) value in neighborhood   Maximum             maximum value per channel in neighborhood   Minimum             minimum value per channel in neighborhood   Mean                average value per channel in neighborhood   Median              median value per channel in neighborhood   Mode                mode (most frequent) value per channel in neighborhood   Nonpeak             value just before or after the median value per channel in neighborhood   RMS                 root mean square value per channel in neighborhood   StandardDeviation   standard deviation value per channel in neighborhood

-stegano offset

hide watermark within an image.

Use an offset to start the image hiding some number of pixels from thebeginning of the image. Note this offset and the image size. You will needthis information to recover the steganographic image (e.g. display -size320x256+35 stegano:image.png).

-stereo +x{+y}

composite two images to create a red / cyan stereo anaglyph.[composite]

The left side of the stereo pair (second image) is saved as the red channel of the output image. The right side (first image) is saved as the green and blue channels. Red-green stereo glasses are required to properly view the stereo image.

-storage-type type

pixel storage type. Here are the valid types:
   char     unsigned characters   double   doubles   float    floats   integer  integers   long     longs   quantum  pixels in the native depth of your ImageMagick distribution   short    unsigned shorts

Float and double types are normalized from 0.0 to 1.0 otherwise the pixelsvalues range from 0 to the maximum value the storage type can support.

-stretch fontStretch

Set a type of stretch style for fonts.

This setting suggests a type of stretch that ImageMagick should try toapply to the currently selected font family. Select fontStretch from the following.

   Any   Condensed   Expanded   ExtraCondensed   ExtraExpanded   Normal   SemiCondensed   SemiExpanded   UltraCondensed   UltraExpanded

To print a complete list of stretch types, use -liststretch.

For other settings that affect fonts, see the options -font, -family, -style, and -weight.


strip the image of any profiles or comments.

-stroke color

color to use when stroking a graphic primitive.

The color is specified using the format described under the -fill option.

See -draw for further details.

-strokewidth value

set the stroke width.

See -draw for further details.

-style fontStyle

Set a font style for text.

This setting suggests a font style that ImageMagick should try to apply tothe currently selected font family. Select fontStyle fromthe following.

   Any   Italic   Normal   Oblique

For other settings that affect fonts, see the options -font, -family, -stretch, and -weight.


search for subimage.[compare]

This option is required to have compare search for the best match locationof a small image within a larger image. This search will produce two images(or two frames). The first is the "difference" image and the second willbe the "match score" image.

The "match-score" image is smaller containing a pixel for ever possibleposition of the top-left corner of the given sub-image. that is its size willbe the size of the larger_image - sub_image + 1. The brightest location inthis image is the location s the locate on the best match that is alsoreported. Note that this may or may not be a perfect match, and the actualbrightness will reflect this. Other bright 'peaks' can be used to locate otherpossible matching locations.

Note that the search will try to compare the sub-image at every possiblelocation in the larger image, as such it can be very slow. The smaller thesub-image the faster this search is.

-swap index,index

Swap the positions of two images in the image sequence.

For example, -swap 0,2 swaps the first and the thirdimages in the current image sequence. Use +swap to switchthe last two images in the sequence.

-swirl degrees

swirl image pixels about the center.

Degrees defines the tightness of the swirl.


synchronize image to storage device.

Set to "true" to ensure all image data is fully flushed and synchronizedto disk. There is a performance penalty, but the benefits include ensuring avalid image file in the event of a system crash and early reporting if thereis not enough disk space for the image pixel cache.


Mark the image as modified.

-text-font name

font for writing fixed-width text.

Specifies the name of the preferred font to use in fixed (typewriter style)formatted text. The default is 14 point Courier.

You can tag a font to specify whether it is a PostScript, TrueType, orOPTION1 font. For example, Courier.ttf is a TrueType font andx:fixed is OPTION1.

-texture filename

name of texture to tile onto the image background.

-threshold value{%}

Apply simultaneous black/white threshold to the image.

Any pixel values (more specifically, those channels set using ‑channel) that exceed the specified threshold are reassigned themaximum channel value, while all other values are assigned the minimum.

The threshold value can be given as a percentage or as an absolute integervalue corresponding to the desired channel value. When given as an integer,the minimum attainable value is 0 (corresponding to black when all channelsare affected), but the maximum value (corresponding to white) is that of thequantum depth of the particular build of ImageMagick, and istherefore dependent on the installation. For that reason, a reasonablerecommendation for most applications is to specify the threshold values asa percentage.

The following would force pixels with red values above 50% to have 100%red values, while those at or below 50% red would be set to 0 in the redchannel. The green, blue, and alpha channels (if present) would be unchanged.

$ convert in.png -channel red -threshold 50% out.png

As (possibly) impractical but instructive examples, the following wouldgenerate an all-black and an all-white image with the same dimensions as theinput image.

$ convert in.png -threshold 100% black.png $ convert in.png -threshold -1 white.png

Note that the values of the transparency channel is treated as 'matte'values (0 is opaque) and not as 'alpha' values (0 is transparent).

See also ‑black‑threshold and ‑white‑threshold.

-thumbnail geometry

Create a thumbnail of the image.

This is similar to -resize, except it is optimizedfor speed and any image profile, other than a color profile, is removed toreduce the thumbnail size. To strip the color profiles as well, add -strip just before of after this option.

See Image Geometry for complete details about the geometry argument.

-tile filename

Set the tile image used for filling a subsequent graphic primitive.

-tile geometry

Specify the layout of images .[montage]

See Image Geometry for complete details about the geometry argument.


Specifies that a subsequent composite operation is repeated across and down image.[composite]

-tile-offset {+-}x{+-}y

Specify the offset for tile images, relative to the background image it is tiled on.

This should be set before the tiling image is set by -tile or -texture, or directly applied forcreating a tiled canvas using TILE: or PATTERN: inputformats.

Internally ImageMagick does a -roll of the tile imageby the arguments given when the tile image is set.

-tint value

Tint the image with the fill color.

Tint the image with the fill color.

Specify the amount of tinting as a percentage. Pure colors like black,white red, yellow, will not be affected by -tint. Only mid-range colors suchas the various shades of grey.

-title string

Assign a title to displayed image.[animate, display, montage]

Use this option to assign a specific title to the image. This assigned tothe image window and is typically displayed in the window title bar.Optionally you can include the image filename, type, width, height, Exif data,or other image attribute by embedding special format characters describedunder the -format option.

For example,

-title "%m:%f %wx%h"

produces an image title of MIFF:bird.miff 512x480 for an imagetitled bird.miff and whose width is 512 and height is 480.


transform the image.

This option applies the transformation matrix from a previous -affine option.

$ convert -affine 2,2,-2,2,0,0 -transform bird.ppm bird.jpg

This operator has been now been superseded by the -distort 'AffineProjection' method.

-transparent color

Make this color transparent within the image.

The color argument is defined using the formatdescribed under the -fill option. The -fuzz setting can be used to match and replace colors similar to the onegiven.

Use +transparent to invert the pixels matched.that is make all non-matching colors transparent.

The -opaque operator is exactly the same as -transparent but replaces the matching color with thecurrent -fill color setting, rather than transparent.However the -transparent operator also ensuresthat the image has an alpha channel enabled, as per "-alpha set", and does not require you to modify the -channel to enable alpha channel handling.

Note that this does not define the color as being the 'transparency color'used for color-mapped image formats, such as GIF. For that use -transparent-color

-transparent-color color

Set the transparent color.

Sometimes this is used for saving to image formats such asGIF and PNG8 which uses this color to represent boolean transparency. Thisdoes not make a color transparent, it only defines what color the transparentcolor is in the color palette of the saved image. Use -transparent to make an opaque color transparent.

This option allows you to have both an opaque visible color, as well as atransparent color of the same color value without conflict. That is, you canuse the same color for both the transparent and opaque color areas within animage. This, in turn, frees to you to select a transparent color that isappropriate when an image is displayed by an application that does not handle atransparent color index, while allowing ImageMagick to correctly handle images of thistype.

The default transparent color is #00000000, which is fully transparent black.


Mirror the image along the top-left to bottom-right diagonal.

This option mathematically transposes the pixel array. It is equivalent to the sequence -flip -rotate 90.


Mirror the image along the images bottom-left top-right diagonal. Equivalent to the operations -flop -rotate 90.

-treedepth value

tree depth for the color reduction algorithm.

Normally, this integer value is zero or one. A value of zero or one causesthe use of an optimal tree depth for the color reduction algorithm.

An optimal depth generally allows the best representation of the sourceimage with the fastest computational speed and the least amount of memory.However, the default depth is inappropriate for some images. To assure thebest representation, try values between 2 and 8 for this parameter. Refer tothe color reduction algorithm for more details.

The -colors or -monochromeoption, or writing to an image format which requires color reduction, isrequired for this option to take effect.


trim an image.

This option removes any edges that are exactly the same color as the cornerpixels. Use -fuzz to make -trim removeedges that are nearly the same color as the corner pixels.

The page or virtual canvas information of the image is preserved allowingyou to extract the result of the -trim operation from theimage. Use a +repage to remove the virtual canvas pageinformation if it is unwanted.

If the trimmed image 'disappears' an warning is produced, and a specialsingle pixel transparent 'missed' image is returned, in the same way as when a-crop operation 'misses' the image proper.

-type type

the image type.

Choose from: Bilevel,Grayscale, GrayscaleMatte, Palette,PaletteMatte, TrueColor, TrueColorMatte,ColorSeparation, or ColorSeparationMatte.

Normally, when a format supports different subformats such as grayscale andtruecolor, the encoder will try to choose an efficient subformat. The -type option can be used to override this behavior. Forexample, to prevent a JPEG from being written in grayscale format even thoughonly gray pixels are present, use.

$ convert bird.png -type TrueColor bird.jpg

Similarly, use -type TrueColorMatte to force theencoder to write an alpha channel even though the image is opaque, if theoutput format supports transparency.

Use -type optimize to ensure the image is written in the smallest possible file size.

-undercolor color

set the color of the annotation bounding box.

The color is specified using the format described under the -fill option.

See -draw for further details.

-update seconds

detect when image file is modified and redisplay.

Suppose that while you are displaying an image the file that is currentlydisplayed is over-written. display will automagically detect thatthe input file has been changed and update the displayed imageaccordingly.


discard all but one of any pixel color.

-units type

the units of image resolution.

Choose from: Undefined, PixelsPerInch, orPixelsPerCentimeter. This option is normally used in conjunctionwith the -density option.

-unsharp radius
-unsharp radiusxsigma{+gain}{+threshold}

sharpen the image with an unsharp mask operator.

The -unsharp option sharpens an image. The image isconvolved with a Gaussian operator of the given radius and standard deviation(sigma). For reasonable results, radius should be larger than sigma. Usea radius of 0 to have the method select a suitable radius.

The parameters are:

   radius     The radius of the Gaussian, in pixels,  not counting the center              pixel (default 0).   sigma      The standard deviation of the Gaussian, in pixels (default 1.0).   gain       The fraction of the difference between the original and the blur              image that is added back into the original (default 1.0).   threshold  The threshold, as a fraction of QuantumRange, needed to apply the              difference amount (default 0.05).


print detailed information about the image when this optionprecedes the -identify option orinfo:.


print ImageMagick version string and exit.

-view string

FlashPix viewing parameters.

-vignette radius{xsigma}{+-}x{+-}y{%}

soften the edges of the image in vignette style.

The vignette effect rolloff is controlled by radiusxsigma. For nominal rolloff, this would be set to 0xsigma. A value of 0x0 will produce a circle/ellipse with no rolloff. The arguments x and y control the size of the circle. Larger values decrease the radii and smaller values increase the radii. Values of +0+0 will generate a circle/ellipse the same size as the image. The default values for x and y are 10% of the corresponding image dimension. Thus, the radii will be decreased by 10%, i.e., the diameters of the circle/ellipse will be 80% of the corresponding image dimension.

-virtual-pixel method

Specify contents of virtual pixels.

This option defines what color source should be used if and when a colorlookup completely 'misses' the source image. The color(s) that appear tosurround the source image. Generally this color is derived from the sourceimage, but could also be set to a specify background color.

Choose from these methods:

   background            the area surrounding the image is the background color   black                 the area surrounding the image is black   checker-tile          alternate squares with image and background color   dither                non-random 32x32 dithered pattern   edge                  extend the edge pixel toward infinity   gray                  the area surrounding the image is gray   horizontal-tile       horizontally tile the image, background color above/below   horizontal-tile-edge  horizontally tile the image and replicate the side edge pixels   mirror                mirror tile the image   random                choose a random pixel from the image   tile                  tile the image (default)   transparent           the area surrounding the image is transparent blackness   vertical-tile         vertically tile the image, sides are background color   vertical-tile-edge    vertically tile the image and replicate the side edge pixels   white                 the area surrounding the image is white

The default value is "edge".

This most important for distortion operators such as -distort, -implode, and -fx.However it also effects operations that may access pixels just outside theimage proper, such as -convolve, -blur, and -sharpen.

To print a complete list of virtual pixel types, use the -list virtual-pixel option.

-visual type

Animate images using this X visual type.[animate, display]

Choose from these visual classes:

   StaticGray   GrayScale   StaticColor   PseudoColor   TrueColor   DirectColor   default   visual id

The X server must support the visual you choose, otherwise an error occurs.If a visual is not specified, the visual class that can display the mostsimultaneous colors on the default screen is chosen.

-watermark brightnessxsaturation

Watermark an image using the given percentages of brightness andsaturation.[composite]

Take a grayscale image (with alpha mask) and modify the destination image'sbrightness according to watermark image's grayscale value and the brightness percentage. The destinations color saturationattribute is just direct modified by the saturationpercentage, which defaults to 100 percent (no color change).

-wave amplitude
-wave amplitudexwavelength

Shear the columns of an image into a sine wave.

Specify amplitude and wavelengthof the wave.

-weight fontWeight

Set a font weight for text.

This setting suggests a font weight that ImageMagick should try to apply tothe currently selected font family. Use a positive integer for fontWeight or select from the following.

fontWeight Description
All No effect.
Bold Same as fontWeight = 700.
Bolder Add 100 to font weight if currently ≤ 800.
Lighter Subtract 100 to font weight if currently ≤ 100.
Normal Same as fontWeight = 400.

To print a complete list of weight types, use -listweight.

For other settings that affect fonts, see the options -font, -family, -stretch, and -style.

-white-point x,y

chromaticity white point.

-white-threshold value{%}

Force to white all pixels above the threshold while leaving allpixels at or below the threshold unchanged.

The threshold value can be given as a percentage or as an absolute integervalue within [0, QuantumRange] corresponding to thedesired ‑channel value. See ‑thresholdfor more details on thresholds and resulting values.

-window id

Make the image the background of a window.[animate, display]

id can be a window id or name. Specify rootto select X's root window as the target window.

By default the image is tiled onto the background of the target window. Ifbackdrop or -resize are specified, theimage is surrounded by the background color. Refer to X RESOURCESfor details.

The image will not display on the root window if the image has more uniquecolors than the target window colormap allows. Use -colors to reduce the number of colors.


specify the window group.

-write filename

write an image sequence.

The image sequence preceding the -write filename option is written out, and processing continues with the same image in its current state if there are additional options. To restore the image to its original state after writing it, use the +write filename option.

Use -compress to specify the type of image compression.