Open GoogleCodeExporter opened 9 years ago
I am experiencing a problem with pypng, where png.Writer(x, x, compression=XX)
gives
me the same output file sizes for all different compression levels. Opening a
file
with, for example, imagej and compressing it with level nine compression
results in
much smaller files. I guess this bug is responsible for this behaviour, but I
am not
quite sure, as I don't know much about image processing... ( I am using the
0.0.11
release)
Is there a chance to get this old bug fixed (if it covers my problem)?
Original comment by grimmw...@german-borg.de
on 18 Jan 2010 at 2:15
@grimmwolf: Filtering could be the problem, or it could be something else.
Do you have an image that you could attach in a comment? Preferably one output
by imagej, and then I can
see what PyPNG makes of it?
I will in any case investigate whether the compression option actually does
anything (it should affect the zlib
compression, but it's possible that there's a bug there).
I should say though... if PyPNG does ever implement filtering on the PNG files
that it creates it will be a _lot_
slower to save.
Original comment by d...@pobox.com
on 19 Jan 2010 at 8:21
Most ways of implementing filters will be quite slow in Python, however, there
is one idea that might be
reasonably fast (and hence worth implementing):
It should be quick to test whether this scanline is identical to the previous
one (if they are both lists, then A==B
can be used). That being the case, the "up" filter can be applied to give a
constant result.
For most "real" images this is unlikely to give any benefit, but there are
plenty of artificial images where it should
give some benefit.
Original comment by d...@pobox.com
on 5 Mar 2010 at 3:27
Original issue reported on code.google.com by
d...@pobox.com
on 16 Mar 2009 at 12:18