Namgoo Lee fb8e652c3f Add CV_16UC1 support for cuda::CLAHE ...

Due to size limit of shared memory, histogram is built on
the global memory for CV_16UC1 case.

The amount of memory needed for building histogram is:

    65536 * 4byte = 256KB

and shared memory limit is 48KB typically.

Added test cases for CV_16UC1 and various clip limits.
Added perf tests for CV_16UC1 on both CPU and CUDA code.

There was also a bug in CV_8UC1 case when redistributing
"residual" clipped pixels. Adding the test case where clip
limit is 5.0 exposes this bug.

2019-02-06 17:21:55 +00:00

..

cuda

Add CV_16UC1 support for cuda::CLAHE

2019-02-06 17:21:55 +00:00

bilateral_filter.cpp

renamed gpuimgproc -> cudaimgproc

2013-09-02 14:00:43 +04:00

blend.cpp

renamed gpuimgproc -> cudaimgproc

2013-09-02 14:00:43 +04:00

canny.cpp

build: don't use nullptr without C++11 guard

2018-09-05 06:19:02 +00:00

color.cpp

Utilize CV_UNUSED macro

2018-09-07 20:33:52 +09:00

corners.cpp

Merge branch 'master' into gpu-cuda-rename

2013-09-06 15:44:44 +04:00

cvt_color_internal.h

rewrite cuda::cvtColor with new device layer and fix test failures

2014-12-25 19:23:15 +03:00

generalized_hough.cpp

Utilize CV_UNUSED macro

2018-09-07 20:33:52 +09:00

gftt.cpp

Thrust allocator usage.

2016-03-15 11:14:47 -04:00

histogram.cpp

Add CV_16UC1 support for cuda::CLAHE

2019-02-06 17:21:55 +00:00

hough_circles.cpp

Utilize CV_UNUSED macro

2018-09-07 20:33:52 +09:00

hough_lines.cpp

Utilize CV_UNUSED macro

2018-09-07 20:33:52 +09:00

hough_segments.cpp

Utilize CV_UNUSED macro

2018-09-07 20:33:52 +09:00

match_template.cpp

refactor cudaarithm reductions:

2014-12-30 11:06:33 +03:00

mean_shift.cpp

replaced GPU -> CUDA

2013-09-02 14:00:44 +04:00

mssegmentation.cpp

Misc. typos

2018-07-31 18:44:23 +03:00

precomp.hpp

Update precomp.hpp

2014-02-17 09:48:51 +01:00