commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gary Lucas (JIRA)" <>
Subject [jira] [Commented] (IMAGING-141) BufferedImage from TIFF is always 24 or 32 bpp
Date Tue, 06 Jan 2015 20:28:34 GMT


Gary Lucas commented on IMAGING-141:

Let me clarify.  I am not yet sure, but I have seen some indications that the performance
for 1-bit images might be a bit slow compared to 8-bit images on a modern graphics system
(1 bit images no longer being a priority for most GPU developers).  If that were to turn out
to be the case, one potential solution would be to store the data as 8-bit images. It would
not be sloppy. It would not be incorrect. It would be a compromise with system performance
issues. On the other hand, if your images are so large that the memory requirements for an
8-bit representation are prohibitive, then a 1-bit format would be required even if it were

> BufferedImage from TIFF is always 24 or 32 bpp
> ----------------------------------------------
>                 Key: IMAGING-141
>                 URL:
>             Project: Commons Imaging
>          Issue Type: Bug
>          Components: Format: TIFF
>            Reporter: Josh Highley
>             Fix For: Patch Needed
> When getting a BufferedImage from a TIFF, TiffImageParser.getBufferedImage gets the TIFF's
BPP from the TIFF tags, but at the end ImageBuilder.getBufferedImage() will always set the
BufferedImage ColorModel BPP to 32 for images with alpha and 24 bpp otherwise.  Many TIFFs
are 1 bpp (G3 and G4) so the BufferedImage should also be 1 bpp.

This message was sent by Atlassian JIRA

View raw message