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
and the resulting jp2 image has an artifact on the right hand side of the image, about 15 pixels wide,
which corresponds to the width of the right-most tiles.
I also observed that the artifact is present for the default n=6 resolution levels, but not present
for n < 6. So, it looks like a tile of width 15 with 6 resolutions levels is not being encoded properly, probably due to LL0 having zero width.
I also observed that if I tried to do a single tile encode of a 303x179 image with n = 9,
encode fails, complaining that the number of resolutions is too high. So, in the 32x32 tile
case, it looks like the code can't handle n=6 for the boundary tiles, of width 15.
So, to be consistent, encoder should reject these settings, just like it rejects single tile encodings when n is too large.
Encode the following TIF with the encoder settings
-i b2_mono.tif -o b2_mono.jp2 -OutFor jp2 -t 32,32
and the resulting jp2 image has an artifact on the right hand side of the image, about 15 pixels wide,
which corresponds to the width of the right-most tiles.
I also observed that the artifact is present for the default n=6 resolution levels, but not present
for n < 6. So, it looks like a tile of width 15 with 6 resolutions levels is not being encoded properly, probably due to LL0 having zero width.
I also observed that if I tried to do a single tile encode of a 303x179 image with n = 9,
encode fails, complaining that the number of resolutions is too high. So, in the 32x32 tile
case, it looks like the code can't handle n=6 for the boundary tiles, of width 15.
So, to be consistent, encoder should reject these settings, just like it rejects single tile encodings when n is too large.
b2_mono.zip
The text was updated successfully, but these errors were encountered: