You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi. I have weird bug. I am working on some software and my idea was to let users upload panorama photos straight from their iPhone and use jpegotpim to optimize it.
It does work but sometimes it degrades quality of sky and leaves weird colorized-like effect on the clouds (probably because it place of lowest entropy on image).
I will probably use another approach (change resolution down before feeding it into jpegotpim) but I am interested in explanation what's happening there and how to detect and prevent it.
I use this command jpegoptim --all-progressive --size=2000 -- filename.jpg. My target is to be 2MB per photo.
severak@kapybara:~$ jpegoptim --version
jpegoptim v1.4.6 x86_64-pc-linux-gnu
Copyright (C) 1996-2018, Timo Kokkonen
This program comes with ABSOLUTELY NO WARRANTY. This is free software,
and you are welcome to redistirbute it under certain conditions.
See the GNU General Public License for more details.
libjpeg version: 6b 27-Mar-1998
Copyright (C) 1991-2020 The libjpeg-turbo Project and many others
@severak, interesting that it seems to mainly affect the sky (probably would affect any larger areas without fine detail).
I think "problem" is that you're setting fixed target file size, instead of -m option to set the quality factor.
With size option jpegoptim will do simple binary search to find quality factor value that produces file closest to the target size set. But this can lead to it using quality factor value that won't produce "usable" results....
I guess, could add command line option to allow specifying the minimum quality factor that jpegoptim will use when it tries to find quality factor producing file of specific size...
Hi. I have weird bug. I am working on some software and my idea was to let users upload panorama photos straight from their iPhone and use
jpegotpim
to optimize it.It does work but sometimes it degrades quality of sky and leaves weird colorized-like effect on the clouds (probably because it place of lowest entropy on image).
See this example or another one. More examples possibly here.
I will probably use another approach (change resolution down before feeding it into jpegotpim) but I am interested in explanation what's happening there and how to detect and prevent it.
I use this command
jpegoptim --all-progressive --size=2000 -- filename.jpg
. My target is to be 2MB per photo.For the reference original of the first picture (16MB) and source code of my resizer.
The text was updated successfully, but these errors were encountered: