| commit | abb3960bd1ac4dd2c9bee85be761b151c83571e8 | |
| author | Stephan Saalfeld <saalfeld@mpi-cbg.de> | |
| Fri, 19 Jun 2009 11:44:28 +0000 (19 13:44 +0200) | ||
| committer | Stephan Saalfeld <saalfeld@mpi-cbg.de> | |
| Fri, 19 Jun 2009 11:44:28 +0000 (19 13:44 +0200) | ||
| tree | d1a6458eecb969d4b246b9af1dc6d495f1fc5f82 | treesnapshot (tar.gz zip) |
| parent | ff1df0fc4db440220d1bab49872a0e92e6cd3129 | commitdiff |
Workaround for a bug which was probably introduced during fixing
of the min/max problems. The underlying problem is, that
the ImageProcessors in Patch.PatchImage do not inherit the min and max
values from the original ImageProcessor during
ImageProcessor.duplicate(). That made transformed images rendered at
wrong contrast for feature extraction.
TODO Find out if making them inherit these values would break the
fragile min/max logic in other places. If not, I suggest to do it that
way---it just makes more sense.
of the min/max problems. The underlying problem is, that
the ImageProcessors in Patch.PatchImage do not inherit the min and max
values from the original ImageProcessor during
ImageProcessor.duplicate(). That made transformed images rendered at
wrong contrast for feature extraction.
TODO Find out if making them inherit these values would break the
fragile min/max logic in other places. If not, I suggest to do it that
way---it just makes more sense.
| mpicbg/trakem2/align/AbstractAffineTile2D.java | diffblobblamehistory |