Re: Is there such thing as a 'soft checksum' tool?
Mel Flynn mel.flynn+fbsd.questi...@mailing.thruhere.net wrote: On Monday 07 September 2009 05:09:53 Michael David Crawford wrote: M I'm looking for a pseudo-checksum tool for use with M cataloging images. One way you could approach it might be to use a blur filter ... Small differences in individual pixels would be blurred away. ... the above does not work, because of compression anyway. Just because you think of an image as a bitmap, does not mean it's stored as such. Certainly it is the decompressed payloads of the JPEG etc. files that are to be compared, rather than the files themselves. It would never have occurred to me that anyone participating in the discussion might have thought otherwise. However, thinking about this inquiry and JPEG in the same sentence has given me an idea that might help the OP: JPEG is a lossy compression, with the degree of loss related to the chosen image quality, so two similar images might become identical -- or at least more similar -- if compressed to a sufficiently low quality using the JPEG algorithm. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is there such thing as a 'soft checksum' tool?
per...@pluto.rain.com wrote: Mel Flynn mel.flynn+fbsd.questi...@mailing.thruhere.net wrote: However, thinking about this inquiry and JPEG in the same sentence has given me an idea that might help the OP: JPEG is a lossy compression, with the degree of loss related to the chosen image quality, so two similar images might become identical -- or at least more similar -- if compressed to a sufficiently low quality using the JPEG algorithm. This seems to be an excellent idea. A similar approach can (successfully) be used to let a computer recognize songs through a micrOphone: the incoming signal is transformmed to MP3 at a rather low quality, which provides a sort of fingerprint of the input. The quality factors shall be adujsted adequately for this application: there is a tradeoff between stability (noise insensitivity) and separation to find. The case of images is much more complicated if one wishes to recognize the same image at two different scales. -- Cheers, Michaël ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is there such thing as a 'soft checksum' tool?
in message 44skf0c6zq@lowell-desk.lan, wrote Lowell Gilbert thusly... Modulok modu...@gmail.com writes: (I am replyin to Lowell's reply for I do not have OP.) I'm not even sure such a tool exists, but it's worth asking: I'm looking for a pseudo-checksum tool for use with catalogging images. For example, a strict checksum algorithm, like the sha family, will produce a dramatically different checksum for two files which differ by only a single bit. I'm looking for something where two images images, which are similar, get a proportionally similar checksum. When I speak of similarities I'm referring to their image patterns. i.e two images of differing sizes, which are otherwise identical, would produce very similar checksums. So the closer the checksums are, the more similar two given images are. Does anyone know of anything like this? See if this ... http://www.stonehenge.com/merlyn/LinuxMag/col50.html ... fits. - parv -- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is there such thing as a 'soft checksum' tool?
Modulok modu...@gmail.com writes: I'm not even sure such a tool exists, but it's worth asking: I'm looking for a pseudo-checksum tool for use with catalogging images. For example, a strict checksum algorithm, like the sha family, will produce a dramatically different checksum for two files which differ by only a single bit. I'm looking for something where two images images, which are similar, get a proportionally similar checksum. When I speak of similarities I'm referring to their image patterns. i.e two images of differing sizes, which are otherwise identical, would produce very similar checksums. So the closer the checksums are, the more similar two given images are. Does anyone know of anything like this? See if this ... http://www.stonehenge.com/merlyn/LinuxMag/col50.html ... fits. - parv *laughs* It makes me feel pretty good after reading how Mr.Schwartz went about it. Before I got any replies I started to think about how I'd do it and began to sketch out an algorithm on the kitchen floor. (Largest black-board in my house.) The general approach was pretty much the same; Recursive bucketing of pixels and generating averaged values down to a user-defined finite limit. Thanks to all who have replied thus far! -Modulok- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is there such thing as a 'soft checksum' tool?
On Monday 07 September 2009 05:09:53 Michael David Crawford wrote: M I'm looking for a pseudo-checksum tool for use with cataloging images. I've seen such tools advertised, but they were proprietary products and only worked on windows. One way you could approach it might be to use a blur filter to blur each of your images, and then to compare the blurred images. Small differences in individual pixels would be blurred away. Did you guys miss Charlie Kester's message? And the above does not work, because of compression anyway. Just because you think of an image as a bitmap, does not mean it's stored as such. -- Mel ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is there such thing as a 'soft checksum' tool?
Modulok modu...@gmail.com writes: I'm not even sure such a tool exists, but it's worth asking: I'm looking for a pseudo-checksum tool for use with catalogging images. For example, a strict checksum algorithm, like the sha family, will produce a dramatically different checksum for two files which differ by only a single bit. I'm looking for something where two images images, which are similar, get a proportionally similar checksum. When I speak of similarities I'm referring to their image patterns. i.e two images of differing sizes, which are otherwise identical, would produce very similar checksums. So the closer the checksums are, the more similar two given images are. Does anyone know of anything like this? It turns out this is a remarkably hard problem. You can look at p5-Image-Compare, but be prepared to experiment before trusting the results. -- Lowell Gilbert, embedded/networking software engineer, Boston area http://be-well.ilk.org/~lowell/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is there such thing as a 'soft checksum' tool?
M I'm looking for a pseudo-checksum tool for use with cataloging images. I've seen such tools advertised, but they were proprietary products and only worked on windows. One way you could approach it might be to use a blur filter to blur each of your images, and then to compare the blurred images. Small differences in individual pixels would be blurred away. Mike -- Michael David Crawford m...@prgmr.com prgmr.com - We Don't Assume You Are Stupid. Xen-Powered Virtual Private Servers: http://prgmr.com/xen ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is there such thing as a 'soft checksum' tool?
On Sat 05 Sep 2009 at 09:33:03 PDT Modulok wrote: List, I'm not even sure such a tool exists, but it's worth asking: I'm looking for a pseudo-checksum tool for use with catalogging images. For example, a strict checksum algorithm, like the sha family, will produce a dramatically different checksum for two files which differ by only a single bit. I'm looking for something where two images images, which are similar, get a proportionally similar checksum. When I speak of similarities I'm referring to their image patterns. i.e two images of differing sizes, which are otherwise identical, would produce very similar checksums. So the closer the checksums are, the more similar two given images are. Does anyone know of anything like this? libpuzzle might be what you're looking for. There's a tool called ftwin that uses libpuzzle to find duplicate or only-slightly-modified files. http://libpuzzle.pureftpd.org/project/libpuzzle http://jok.is-a-geek.net/ftwin.php Both of these are in the portstree. ;-) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Is there such thing as a 'soft checksum' tool?
On Sat, 5 Sep 2009 10:33:03 -0600, Modulok modu...@gmail.com said: M I'm looking for a pseudo-checksum tool for use with cataloging images. M For example, a strict checksum algorithm, like the sha family, will M produce a dramatically different checksum for two files which differ by M only a single bit. I'm looking for something where two images, which M are similar, get a proportionally similar checksum. This works pretty well with text, but I don't know about images. http://ixazon.dynip.com/~cmeclax/nilsimsa.html What's a nilsimsa code? A nilsimsa code is something like a hash, but unlike hashes, a small change in the message results in a small change in the nilsimsa code. Such a function is called a locality-sensitive hash. -- Karl Vogel I don't speak for the USAF or my company Garlic gum is not funny.--written on blackboard by Bart Simpson ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
