Fix GDAL image decoding color problems identified by issue #10089, by: (#10093)
* Fix GDAL image decoding color problems identified by issue #10089, by:
Fixing CV_8UC1 symbol, which should be CV_8UC3 for RGB GDAL color table images.
Fixing image.ptr<VecX>(row,col)[] to be (*image.ptr<VecX>(row,col))[] to correctly access VecX array elements, as ptr<VecX>() returns a pointer to the VecX, not the first element of VecX. This fixes the color problem with color table gif images, and avoids out-of-bounds memory access.
Respecting the color identification of raster bands provided by the GDAL image driver, and produce BGR or BGRA images. Note that color bands of images using the HSL, CMY, CMYK, or YCbCr color space are ignored, rather than converting them to BGR.
* When reading image files using the GDAL decoder, exit with an error if a color band is encountered that isn't used (eg. from CMYK or YCbCbr), rather than silently ignoring the band's data.
Previously, only file-based encoding and decoding were supported with
the libtiff library, leading to the possible use of temporary files.
This fixes issue #8483.
Previously, the return value of fwrite and fclose were not properly
checked, leading to possible silent truncation of the data if writing
failed, e.g. due to lack of disk space.
Fixes issue #9251.
ExifReader::getExif may enter infinite loop with jpeg image which have no EOI.
For example, bytesToSkip may be set to 0 and fseek seems like fseek(f, -2 , SEEK_CUR) for image that end with RST7(FF D7) instead of EOI.
In cases where the signaure string contains a terminating character,
the std::string member function size returns a smaller value than the
allocated string. In these cases, if you then try to use substr,
you will get an out_of_range exception. This patch remedies the problem.
This patch implements the PAM image format as defined at:
http://netpbm.sourceforge.net/doc/pam.html
The PAM format provides a generic means for storing 2 dimensional information.
This is useful for opencv since there are cases where data gets translated into
non standardized formats, which makes it difficult to store and load this information.
UIImages with alpha were ending up with garbage pixels in background (random memory values). Need to initialize matrix pixels before drawing UIImage with alpha on it.
Note: didn’t fix Grayscale image with alpha stripping alpha in UIImage -> Mat conversion.
* master: (468 commits)
added suppression for TBB valgrind issue
update CUDA architecture flags initialization
increase minimal supported CUDA toolkit to 6.5
check the CPU flag correctly
opencv_visualization: check cmdline args
provide better error messages
stop search of markers in Exif reader to prevent infinite loop
Fix calibration fail on python with CALIB_THIN_PRISM_MODEL flag
clarify CUDA arithm operations usage with mask
fixed empty image condition in resize
fixed memory leak in flann tests
fisheye: add CALIB_FIX_PRINCIPAL_POINT
get/put: more type-safety and code unification using templates
py_tutorials: fix cv2.findContours return val
imgproc: speed up threshold of 64F version using NEON and SSE * use NEON under aarch64 only * check 64F version correctly
bigdata: add test, resolve split/merge issue
Improved Carotene library linear resize evaluation precision and enabled it as HAL implementation.
persistence: fixing crash with space-only values
Removed unnecessary check for Android API level and unused flags.
Fix for median blur of 2-channel images
...
- fixed width and height order
- removed unused methods
- simplified signature matching
- rewrote pixel format matching in more compact form
- added dimensions number check (only 2 is allowed)
- added target buffer size check
- added debug messages in all failing points
