Main Area and Open Discussion > General Software Discussion

photo duplicate scanner

<< < (22/25) > >>

cranioscopical:
Hmm... maybe I should just try to learn proper English?
-Curt (November 11, 2010, 05:14 AM)
--- End quote ---
What a fantamazing idea!  ;)

tomos:
Hmm... maybe I should just try to learn proper English?
-Curt (November 11, 2010, 05:14 AM)
--- End quote ---

OTOH wouldnt it be nice to have a new English word originating here at dc
(or there in dk) :)

sajman99:
Seems like both AntiDupl & UniqueFiler have their bugs, AD finds a bunch of quasi-equal images (few pixels different for a 1280*1024 image) that UF didn't catch, while the exact same happens the other way around. :(
-Szero (November 11, 2010, 01:27 AM)
--- End quote ---

This finding is fairly common, and it's why I don't rely on a single program to detect similar images. All of these duplicate image finders have strengths and weaknesses based on algorithm.

For example, I recently had a pic which was much darker than its duplicate. I could easily see they were the same, but many of these tools were blind to the match. Also, inaccuracies will occur with dark/light backgrounds and captioned images; many algorithms aren't smart enough to deal with these situations. It's sometimes really surprising when a particular software is unable to "see" what is in fact a visually similar image.

sajman99:
Greetings,

The author of Prismatic Dup Detector has released version 3.30.
http://www.prismaticsoftware.com/dupdetector/dupdetector.html

Wow, it's been six years since the last version of this great freeware.

Curt:
Dup Detector used 22% CPU for 22 minutes, in order to "Build" a data file from my 5.772 medium sized JPGs. The following analysis was both fast (a few seconds) and exact. This is a keeper!

Thanks for telling  :up:

Navigation

[0] Message Index

[#] Next page

[*] Previous page