Search found 34 matches

by Wolfgang Woehl
2013-02-17T22:54:19-07:00
Forum: Users
Topic: Smoothing scanned text jagged/rough edges using Hit-And-Miss
Replies: 3
Views: 7407

Re: Smoothing scanned text jagged/rough edges using Hit-And-

Just FYI: Tesseract recognizes all lines in your sample correctly.
by Wolfgang Woehl
2013-02-12T12:48:09-07:00
Forum: Kudos and Rants
Topic: I know they're there...but who the heck is using ImageMagick
Replies: 8
Views: 34055

Re: I know they're there...but who the heck is using ImageMa

Cinemaslides (I'm the author) is a tool to create Digital Cinema Packages (DCPs) from all kinds of images -- slideshows really, including movies (rapid slideshows). ImageMagick is the central tool in there for conforms, resizes, color transforms etc. No way to do it that simple without IM.
by Wolfgang Woehl
2013-01-21T14:48:46-07:00
Forum: Users
Topic: Quality loss with JPEG2000?
Replies: 4
Views: 5657

Re: Quality loss with JPEG2000?

jp2:rate is probably too low.
by Wolfgang Woehl
2013-01-21T12:17:15-07:00
Forum: Users
Topic: Pre-processing for OCR (outlined font)
Replies: 10
Views: 17986

Re: Pre-processing for OCR (outlined font)

Yes, it's an attractively hard problem for any kind of non-OCR approach. From an OCR-centric point-of-view, though, it might be close to trivial, shape recognition and some intelligence towards the concept of outlines.
by Wolfgang Woehl
2013-01-20T14:21:56-07:00
Forum: Users
Topic: Pre-processing for OCR (outlined font)
Replies: 10
Views: 17986

Re: Pre-processing for OCR (outlined font)

Ok, there's one way to do it -- if somewhat cumbersome and, indeed, horrifyingly slow :) Thanks for the effort, snibgo! The problem, I think, with this approach is that it deals with the artefacts of an initial operation (background floodfill with outline color) which is not suitable in the first pl...
by Wolfgang Woehl
2013-01-19T06:42:47-07:00
Forum: Users
Topic: Pre-processing for OCR (outlined font)
Replies: 10
Views: 17986

Re: Pre-processing for OCR (outlined font)

Right, but the kernels are fixed-size. Thus whatever it will match is fixed-size, from what I understand? I found a related topic (area opening and closing): Morphology, area open and close . This is about selecting contiguous areas bigger (or smaller) than a specific amount of pixels. In conjunctio...
by Wolfgang Woehl
2013-01-18T16:54:00-07:00
Forum: Users
Topic: Pre-processing for OCR (outlined font)
Replies: 10
Views: 17986

Re: Pre-processing for OCR (outlined font)

snibgo, very interesting idea indeed. Thanks for the suggestion. Going to experiment with it tomorrow. Hardcoded pixel matches though, yes? That's probably going to be a problem for variable-sized input.
by Wolfgang Woehl
2013-01-18T02:38:57-07:00
Forum: Users
Topic: Pre-processing for OCR (outlined font)
Replies: 10
Views: 17986

Pre-processing for OCR (outlined font)

I'm trying to improve OCR/tesseract legibility of text rendered in an outline font. Text spacing is dense enough that a bunch of letter pairs will touch: http://minus.com/lyPtV8gTWD0SP . Feeding this original through tesseract will not output anything useful (completely garbled). My best shot at it ...
by Wolfgang Woehl
2012-10-07T05:52:55-07:00
Forum: Users
Topic: Calculate hash of image pixels
Replies: 4
Views: 9178

Re: Calculate hash of image pixels

Anthony, in some use cases having a image-data-only hash is extremely useful. Now if the hash algorithm used was configurable ...
by Wolfgang Woehl
2011-04-16T00:44:53-07:00
Forum: Developers
Topic: Resizing properly in a world of {default=sRGB}
Replies: 18
Views: 46276

Re: Resizing properly in a world of {default=sRGB}

magick wrote:It looks like the docs are wrong (we'll correct that). The source code uses 0.04045 for K0.
Ugh, I was grepping an ancient version. Apparently 6.6.3-10 was the last one that used 0.03928f. Sorry for the noise.
by Wolfgang Woehl
2011-04-15T15:34:43-07:00
Forum: Developers
Topic: Resizing properly in a world of {default=sRGB}
Replies: 18
Views: 46276

Re: Resizing properly in a world of {default=sRGB}

magick, I see that magick/colorspace.c is using 0.03928f for K0 (just like the docs at http://www.imagemagick.org/script/command-line-options.php#colorspace section sRGB say.) I'm kind of staring at http://en.wikipedia.org/wiki/SRGB#Theory_of_the_transformation which discusses K0 as 0.04045 in order...
by Wolfgang Woehl
2010-12-27T14:24:31-07:00
Forum: Users
Topic: Montage a whole bunch of tiles?
Replies: 6
Views: 13910

Re: Montage a whole bunch of tiles?

3. Get the coordinates from the filename ... It's step 3 that I have a problem with. There just doesn't seem to be a way to extract the coordinates from the filename. With ruby this is one way to do that: x,y = 'tile_3_-21.png'.match( /(-?\d+)_(-?\d+)/ )[ 1 .. -1 ].collect { |n| n.to_i } which matc...
by Wolfgang Woehl
2010-10-03T11:13:18-07:00
Forum: Users
Topic: MagickSigmoidalContrastImage
Replies: 5
Views: 10067

Re: MagickSigmoidalContrastImage

Moving the center of a sigmoid function (also known as "s-curves") changes rates of both ends. Looking at an "S" makes "shoulder" and "toe" quite obvious.
by Wolfgang Woehl
2010-08-30T11:58:08-07:00
Forum: Users
Topic: Need help to improve special scanned images
Replies: 17
Views: 31885

Re: Need help to improve special scanned images

For an alternative (to Fred's awesome suggestions) look at http://unpaper.berlios.de/
by Wolfgang Woehl
2010-08-07T00:36:24-07:00
Forum: Users
Topic: JPEG2000 Compressed TIFF problem
Replies: 7
Views: 16148

Re: JPEG2000 Compressed TIFF problem

Well, see http://www.digitalpreservation.gov/formats/content/tiff_tags.shtml and http://www.awaresystems.be/imaging/tiff/tifftags/compression.html which says (bottom of the page): LibTiff adds support for some compression schemes that are not part of the specification and are somewhat less common. H...