Search found 28 matches

by pipe
2013-05-16T19:28:44-07:00
Forum: Developers
Topic: FFTW benchmarking, single vs. double precision transforms.
Replies: 3
Views: 8004

Re: FFTW benchmarking, single vs. double precision transform

Interesting, I did not know that. I always wondered why it had to be square, because nothing in the actual transforms mandate this. I'm going to investigate this. One minor "problem" is that all textbooks I've read about this (very very few since I'm an electrical engineer and hardly even that) almo...
by pipe
2013-05-16T03:37:56-07:00
Forum: Developers
Topic: FFTW benchmarking, single vs. double precision transforms.
Replies: 3
Views: 8004

FFTW benchmarking, single vs. double precision transforms.

I did some benchmarking with FFTW for another project. I got tired of messing with the graphs, but on my somewhat standard Core2 Quad, switching from double precision FFTW to single precision FFTW would get a speed increase of around 20-40%. The main benefit of switching to single precision is howev...
by pipe
2013-05-10T06:36:09-07:00
Forum: Bugs
Topic: Core dumped with 'stream'
Replies: 1
Views: 3939

Core dumped with 'stream'

I was trying to convert a bunch of huge TIFFs to floats, ran in to a bug. pipe@spacelab:~$ stream -version Version: ImageMagick 6.8.5-5 2013-05-04 Q16 http://www.imagemagick.org Copyright: Copyright (C) 1999-2013 ImageMagick Studio LLC Features: DPC OpenCL OpenMP Delegates: bzlib djvu fftw fontconfi...
by pipe
2013-05-05T08:28:22-07:00
Forum: Users
Topic: [SOLVED]-commands that do not work for me
Replies: 8
Views: 14884

Re: commands that do not work for me

xdpyinfo - display information utility for X
It's part of X11/X.Org. Has nothing to do with Windows or ImageMagick. May work in OSX.
by pipe
2013-05-03T07:30:19-07:00
Forum: Bugs
Topic: CMYK separation problem - old bug?
Replies: 2
Views: 4492

CMYK separation problem - old bug?

I ran in to a problem when trying to separate CMYK channels, I'm not 100% sure it's actually a bug, but the behaviour is off. It has been mentioned about a year ago by Anthony: http://www.imagemagick.org/discourse-server/viewtopic.php?f=1&t=20729#p83381 The main issue is this: convert rose: -colorsp...
by pipe
2013-05-01T07:40:33-07:00
Forum: Digital Image Processing
Topic: Finding optimal colours for separating a scanned print
Replies: 2
Views: 8519

Re: Finding optimal colours for separating a scanned print

I have not printed them myself, so I have no idea about the printer/paper profile, that's my problem here. I should be able to find a color profile for the scanner, it seems pretty high-end. EDIT: I'm not trying to print anything either, I just want to analyze and improve the quality of the scanned ...
by pipe
2013-04-29T10:15:42-07:00
Forum: Digital Image Processing
Topic: Finding optimal colours for separating a scanned print
Replies: 2
Views: 8519

Finding optimal colours for separating a scanned print

I scanned some printed material at a high resolution recently, and I want to reverse the halftoning. I know how to do this using an FFT, but I was thinking a bit about the color channels. It seems to me that I would want to do this in a CMYK colorspace, but is there a way to extract the exact colors...
by pipe
2013-04-27T04:09:06-07:00
Forum: Kudos and Rants
Topic: Thanks for a great product
Replies: 0
Views: 24541

Thanks for a great product

Just wanted to thank you people for ImageMagick, I've been using it for various things over the years. Here's my latest project: http://www.natrox.org/ami/blizzard1260.png (Warning, 16 MB PNG) I had to debug an old circuitboard, and to do this I had to recreate the component traces. I scanned the bo...
by pipe
2013-04-27T03:54:57-07:00
Forum: Bugs
Topic: Issues with -clamp and -depth in HDR
Replies: 17
Views: 17084

Re: Issues with -clamp and -depth in HDR

Maybe I should start using IM7, I think there are a lot of problems with negative values in HDRI in IM6 that would still be in IM7, and I would like to help finding them. Most image processing algorithms that I have read about doesn't need to handle negative image values, so there may be a lot of is...
by pipe
2013-04-25T04:19:18-07:00
Forum: Bugs
Topic: Issues with -clamp and -depth in HDR
Replies: 17
Views: 17084

Re: Issues with -clamp and -depth in HDR

Yeah, what the patch does is simply to test for negative numbers just before applying the operation that would cause negative numbers to break. The level operation is left intact. This is however not the case with the -gamma operator, which works through a lookup-table, which has the effect of both ...
by pipe
2013-04-24T01:23:54-07:00
Forum: Developers
Topic: Selective -clamp may be needed
Replies: 1
Views: 3673

Selective -clamp may be needed

When I was thinking about the problem with negative values and gamma, I realized that a similar problem may show up with other algorithms, and that it would be great to have a -clamp that only clips values below 0.0. If -clamp had not already been used to clamp both the upper and lower range, I woul...
by pipe
2013-04-23T14:06:50-07:00
Forum: Developers
Topic: Decompose 16-bit TIF into two 8-bit PNGs
Replies: 10
Views: 15180

Re: Decompose 16-bit TIF into two 8-bit PNGs

If you really must do this with imagemagick you can use this trick. It assumes that you know the image dimensions and that you are running on a platform with working pipes (that is, linux/bsd/osx/cygwin..). If you run this in windows you can create a temporary file instead. (GRAY:foo.raw instead of ...
by pipe
2013-04-23T10:53:41-07:00
Forum: Bugs
Topic: Issues with -clamp and -depth in HDR
Replies: 17
Views: 17084

Re: Issues with -clamp and -depth in HDR

One problem with the current solution is if someone tries a gamma of 0.5. This will invert every negative value. I can see how this could be troublesome, because most of those pixels will still have a very dark color, so they will not be detected by the naked eye under many viewing conditions. They ...
by pipe
2013-04-23T02:55:21-07:00
Forum: Bugs
Topic: Issues with -clamp and -depth in HDR
Replies: 17
Views: 17084

Re: Issues with -clamp and -depth in HDR

For more headache-inducing problems, try a gamma of exactly 0.5 involving negative pixel values and a HDRI enabled ImageMagick: convert-hdr rose: -scale 400% -level 50,100%,0.5 -depth 8 x: This flips the negative values to positive, because of mathemagic. (Because (-x)^y is defined when y is an inte...
by pipe
2013-04-22T05:05:20-07:00
Forum: Bugs
Topic: Issues with -clamp and -depth in HDR
Replies: 17
Views: 17084

Re: Issues with -clamp and -depth in HDR

No, identify reports TrueColor, and it's straight from the scanner so it's unlikely. I did find the source of the bug though, and it's internal to ImageMagick and relates to how Not-a-Number floats are handled. I performed some additional tests. I got the same thing to break using rose: convert-hdr ...