Search found 15 matches

by foxyshadis
2011-02-10T22:34:14-07:00
Forum: Bugs
Topic: High-bit YUV input no longer working?
Replies: 0
Views: 3179

High-bit YUV input no longer working?

I've just started using the 16-bit functionality of a video encoder I'm testing out. Unfortunately, it looks like IM can't read the 16-bit files that it outputs. My command lines: convert -size 1280x1264 -depth 8 -sampling-factor 4:2:0 out.yuv out.png convert -size 1280x1264 -depth 16 -sampling-fact...
by foxyshadis
2011-01-20T15:26:52-07:00
Forum: MagickWand
Topic: MorphologyApply available for Wand?
Replies: 2
Views: 10700

Re: MorphologyApply available for Wand?

Thanks, I did miss that. I added it to my PyMagickWand set.

BTW, there's a problem in the morphology docs: ScaleGeometryKernelInfo doesn't have its signature shown, instead ScaleKernelInfo is listed twice.
by foxyshadis
2011-01-18T20:46:42-07:00
Forum: MagickWand
Topic: MorphologyApply available for Wand?
Replies: 2
Views: 10700

MorphologyApply available for Wand?

I would like to use Morphology in my current project, there's some really powerful stuff in it. Is there a MagickWand interface to it yet? I suspect there's not, since I looked over the code. If not, can you post a short snippet demonstrating how to combine Wand and Core APIs without accidentally le...
by foxyshadis
2011-01-18T20:20:20-07:00
Forum: Users
Topic: How to speed up to load enlarge TIFF File
Replies: 4
Views: 8211

Re: How to speed up to load enlarge TIFF File

In this case, the Q8 version would be better than the Q16, since it'll halve the memory and some of the processing required. (Everything is natively stored in expanded 8- or 16-bit in memory, even if it's only 1-bit packed on disk.) You might have to use some form of partitioning, but it would be ug...
by foxyshadis
2011-01-06T22:30:08-07:00
Forum: Users
Topic: Reduce time for zoom image without loosing image quality..
Replies: 3
Views: 6635

Re: Reduce time for zoom image without loosing image quality

Many zooms in and out will always give you distortion. That's just how it works. Why on earth would you expect it to look good when you throw away 50-90% of the information and then try to regenerate it with an algorithm, 25 times in a row? That's unrealistic. If you can perform any kind of windowin...
by foxyshadis
2011-01-06T20:07:32-07:00
Forum: Users
Topic: Neighborhood Maximum/Minimum in IM?
Replies: 2
Views: 5641

Re: Neighborhood Maximum/Minimum in IM?

Thanks, the Hit/Miss & Pattern Matching methods on the page help solve another problem I was about to start experimenting with. Awesome!
by foxyshadis
2011-01-06T16:15:52-07:00
Forum: Users
Topic: Neighborhood Maximum/Minimum in IM?
Replies: 2
Views: 5641

Neighborhood Maximum/Minimum in IM?

I looked through the documentation and the filter list and didn't see this, so I'm wondering if it's something that's in IM or that someone has implemented it in script. Many image-editing programs have a local min/max filter that uses 3x3, 5x5 etc, and it's handy in some instances. (For instance, o...
by foxyshadis
2010-12-29T22:11:21-07:00
Forum: MagickWand
Topic: Stack Overflow in MagickWriteImage after deskew - fixed
Replies: 1
Views: 7423

Re: Stock Overflow in MagickWriteImage after deskew

Well, it turns out that this was because there was another, older, set of IM dlls in my path. By dynamically setting the correct IM path to be FIRST instead of last, it turns out that it works fine. Sorry about that.
by foxyshadis
2010-12-29T21:03:35-07:00
Forum: MagickWand
Topic: Stack Overflow in MagickWriteImage after deskew - fixed
Replies: 1
Views: 7423

Stack Overflow in MagickWriteImage after deskew - fixed

I'm loading an image, reading the properties, deskewing, and saving. I've extended sickos' PyMagickWand to work with the current version and I haven't had any trouble on previous IM-using projects up until I tried to use Deskew. Here's my results, please let me know if I'm doing something wrong: b'T...
by foxyshadis
2010-04-30T13:43:03-07:00
Forum: Bugs
Topic: Crash opening YCbCr Tiff in x64
Replies: 4
Views: 8924

Re: Crash opening YCbCr Tiff in x64

I'm on Win7 x64, using IM 6.6.1-6 2010-04-23 Q16 (no OpenMP). IM is one of the very few programs that can properly handle YCbCr in tiff. Some software (like XnView) converts the colors incorrectly, most won't open it at all. It's created with ffmpeg, which added the capability in 2007: ffmpeg -i sam...
by foxyshadis
2010-04-29T23:32:12-07:00
Forum: Bugs
Topic: Crash opening YCbCr Tiff in x64
Replies: 4
Views: 8924

Re: Crash opening YCbCr Tiff in x64

Not sure why, I created it in 7zip and can open it in regular zip software, but I uploaded another one created in windows explorer.
by foxyshadis
2010-04-29T21:43:09-07:00
Forum: Bugs
Topic: Crash opening YCbCr Tiff in x64
Replies: 4
Views: 8924

Crash opening YCbCr Tiff in x64

This image causes a crash, only in x64 windows compiles, not either 32-bit. If I get a chance I'll further debug it.
by foxyshadis
2010-04-29T21:04:38-07:00
Forum: Bugs
Topic: Rounding error in -compose subtract
Replies: 3
Views: 7256

Re: Rounding error in -compose subtract

fmw42 wrote:-compose subtract will wrap. try using -compose minus. see http://www.imagemagick.org/Usage/compose/#subtract and http://www.imagemagick.org/Usage/compose/#minus
That's the intent. Check my command lines again - if I used minus, it'll destroy the recreated image.
by foxyshadis
2010-04-29T21:03:37-07:00
Forum: Bugs
Topic: Rounding error in -compose subtract
Replies: 3
Views: 7256

Re: Rounding error in -compose subtract

I installed the 32-bit, Q8 version, and the bug does not appear. I was originally using the x64 Q16 version. I've just verified that the 32-bit Q16 version is incorrect as well. It looks like this line in composite.c: if (pixel < 0.0) pixel+=(QuantumRange+1.0); It looks like the floating point arith...
by foxyshadis
2010-04-29T20:11:40-07:00
Forum: Bugs
Topic: Rounding error in -compose subtract
Replies: 3
Views: 7256

Rounding error in -compose subtract

Hello, I'm trying to use subtract for a research project. It mostly works quite well, the original image is reconstructed perfectly, but in a few spots huge errors manifest - where it seems to create a difference of 255 in one or more channels. In all other cases I've verified that the difference is...