mirror/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2024-11-30 06:10:02 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
20,771 Commits 6 Branches 126 Tags 3 GiB
91a0270432
Commit Graph

2323 Commits

Author SHA1 Message Date
Li Peng
396921dd23 5x5 gaussian blur optimization 
Add new 5x5 gaussian blur kernel for CV_8UC1 format,
it is 50% ~ 70% faster than current ocl kernel in the perf test.

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-12-06 09:42:37 +08:00
Vadim Pisarevsky
f84dc354f6 Merge pull request #7564 from terfendail:ovx_gaussianfilter 2016-12-02 15:24:03 +00:00
Vitaly Tuzov
afc7396975 Added OpenVX based processing to gaussianBlur 2016-12-02 14:46:12 +03:00
Li Peng
b69cdb2434 Image pyramids upsampling optimization 
Add new ocl kernel for image pyramids upsampling,
It is 35% faster than current OCL kernel in perf test.

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-12-02 13:54:58 +08:00
Vadim Pisarevsky
c99097f89a Merge pull request #7478 from terfendail:ovx_boxfilter 2016-12-01 17:57:46 +00:00
Vadim Pisarevsky
1d45726f47 Merge pull request #7588 from terfendail:ovx_medianblur 2016-12-01 17:54:34 +00:00
Vadim Pisarevsky
e623e710be Merge pull request #7597 from terfendail:ovx_threshold 2016-12-01 17:53:38 +00:00
Vadim Pisarevsky
32f48a8d9c Merge pull request #7631 from savuor:openvx_accumulate 2016-12-01 17:49:18 +00:00
Vadim Pisarevsky
8968318dfc Merge pull request #7668 from savuor:openvx_equalize_hist 2016-12-01 17:43:44 +00:00
Vitaly Tuzov
6d55e99291 Added OpenVX based processing to boxFilter 2016-11-30 11:27:25 +03:00
Li Peng
2ca5a7e862 more optimization for warpAffine and warpPerspective 
Add new OpenCL kernels for bicubic interploation, it is 20% faster
than current warp image kernel with bicubic interploation.

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-11-30 15:43:41 +08:00
Vadim Pisarevsky
c47267ef7f Merge pull request #7538 from Tetragramm:CLAHEfix 2016-11-29 16:42:05 +00:00
Vadim Pisarevsky
57aaec1ee8 Merge pull request #7610 from terfendail:ovx_sobel 2016-11-29 16:40:10 +00:00
Rostislav Vasilikhin
0ac934d09a added OpenVX call to equalizeHist() function 2016-11-29 17:53:47 +03:00
Rostislav Vasilikhin
7c2734ff14 added OpenVX calls to accumulate(), accumulateSquare(), accumulateWeighted() 2016-11-29 17:53:35 +03:00
Vitaly Tuzov
0cd3788624 Added OpenVX based processing to threshold 2016-11-29 15:35:52 +03:00
Vitaly Tuzov
9200db778c Added OpenVX based processing to medianBlur 2016-11-29 15:32:07 +03:00
Vitaly Tuzov
e8f3469096 Added OpenVX based processing to Sobel 2016-11-29 15:29:00 +03:00
Alexander Alekhin
90b52cd9b8 Merge pull request #7726 from pengli:warp_image 2016-11-29 12:19:18 +00:00
Li Peng
b72d196753 optimization for warpAffine and warpPerspective 
Add new ocl kernels for warpAffine and warpPerspective,
The average performance improvemnt is about 30%. The new
ocl kernels require CV_8UC1 format and support nearest
neighbor and bilinear interpolation.

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-11-29 14:55:58 +08:00
Alexander Alekhin
7dc673fd56 imgproc: fix findContours (submatrix input) 2016-11-25 18:25:26 +03:00
Alexander Alekhin
88ef8c996e Merge pull request #7709 from alalek:fix_java_tests 2016-11-24 16:15:56 +00:00
Alexander Alekhin
4fd00c1ae1 Merge pull request #7714 from onixie:master 
Fix docs explaining the returnPoints argument of convexHull.
 2016-11-23 17:44:05 +03:00
Rostislav Vasilikhin
7db43f9fff fixed wrong equivalence in YUV conversion (#7481) 
* fixed wrong equivalence in YUV conversion

* fixed channel order from YVU to YUV
 2016-11-23 17:39:18 +03:00
Vadim Pisarevsky
b8c875d4a1 Merge pull request #7707 from K-Shinotsuka:issue37 2016-11-23 14:15:02 +00:00
monoid911
b879126584 Fix docs explaining the returnPoints argument of convexHull. 2016-11-23 20:43:52 +09:00
Alexander Alekhin
8019498c6e java tests fixes 2016-11-23 13:53:00 +03:00
k-shinotsuka
284006e336 add NEON code for RGB2Luv_f. 2016-11-23 19:18:41 +09:00
Vadim Pisarevsky
c1ee798213 Merge pull request #7662 from K-Shinotsuka:issue34 2016-11-22 13:57:13 +00:00
Vadim Pisarevsky
af22b6c4a9 Merge pull request #7664 from K-Shinotsuka:issue35 2016-11-22 13:49:14 +00:00
Vadim Pisarevsky
b54bc2db1e Merge pull request #7665 from K-Shinotsuka:issue36 2016-11-22 13:44:42 +00:00
Pavel Vlasov
dfb9c574a4 IPP 2017 filter2D fix; 
filter2D IPP runtime check;
 2016-11-21 17:18:03 +03:00
Li Peng
6cb73356b1 laplacian ocl kernel optimization 
This ocl kernel is 46%~171% faster than current laplacian 3x3
ocl kernel in the perf test, with image format "CV_8UC1".

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-11-17 12:01:02 +08:00
k-shinotsuka
cf5c400ecb add SSE code for RGB2HLS_f. 2016-11-16 00:46:13 +09:00
Alexander Alekhin
c93fb14dd2 Merge pull request #7653 from pengli:deriv 2016-11-14 13:40:04 +00:00
Juha Reunanen
5bb08bae28 Merge pull request #7464 from reunanen:test-7409-7458 
Change contour test images to be very wide (#7464)

* Change contour test images to be very wide (#7409, #7458)

Unfortunately, slows down the tests.

* Decrease the number of contour test cases, in order to (at least partially) offset the test run duration increase caused by making the test images wider

* Don't test with very wide images on 32-bit architectures
 2016-11-14 16:31:29 +03:00
Li Peng
8d4a7d3dcc sobel and scharr ocl kernel optimization 
It improves 108%~230% performance in the perf test
with image format "CV_8UC1" and kernel size 3.

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-11-14 15:34:59 +08:00
k-shinotsuka
a909527a39 add SSE code for HLS2RGB_f. 2016-11-14 01:44:33 +09:00
k-shinotsuka
614ce7743e add SSE code for HSV2RGB_f. 2016-11-13 23:46:21 +09:00
Vadim Pisarevsky
4ebdf4938a Merge pull request #7634 from paroj:undistpt 2016-11-11 14:13:18 +00:00
Alexander Alekhin
b2e0873af1 Merge pull request #7629 from alalek:issue_7626 2016-11-10 16:35:08 +00:00
Vadim Pisarevsky
a2d6d96ef1 Merge pull request #7161 from terfendail:shortline_fix 2016-11-10 16:02:20 +00:00
Pavel Rojtberg
3609343acf undistortPoints: only consider distCoeffs if present 
iters should be 0 if we have no distortion. Also skip tilt distortion in
that case.
Furthermore move variable declarations to usage sites.
 2016-11-09 16:11:26 +01:00
Alexander Alekhin
17ffb28807 Merge pull request #7602 from mshabunin:fix-opencl-warnings 2016-11-09 12:35:00 +00:00
Alexander Alekhin
f1d93cb23b Merge pull request #7624 from pengli:gaussian_blur 2016-11-09 09:25:25 +00:00
Alexander Alekhin
183c75358b imgproc: fix trailingZeros for MSVS 2010 2016-11-08 16:48:41 +03:00
Pavel Vlasov
349d5ba012 --perf_instrument parameter now has int type and 0, 1, 2 modes (1 - simple trees, 2 - expanded trees for functions with same name but different calling address); 
Maximum depth limit var was added to the instrumentation structure;

Trace names output console output fix: improper tree formatting could happen;
Output in case of error was added;

Custom regions improvements;

Improved timing and weight calculation for parallel regions; New TC (threads counter) value to indicate how many different threads accessed particular node;

parallel_for, warnings fixes and ReturnAddress code from Alexander Alekhin;
 2016-11-08 10:18:05 +03:00
Li Peng
8f63f51e81 gaussian blur ocl kernel optimization 
This ocl kernel is for 3x3 kernel size and CV_8UC1 format
It is 115% ~ 300% faster than current ocl path in perf test

python ./modules/ts/misc/run.py -t imgproc --gtest_filter=OCL_GaussianBlurFixture*

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-11-08 11:22:26 +08:00
Alexander Alekhin
442380bfac Merge pull request #7585 from pengli:morph_filter 2016-11-07 17:11:32 +00:00
mshabunin
3e28d51779 Fixed several OpenCL compiler warnings 2016-11-07 16:49:12 +03:00
Li Peng
35198b84a4 morph ocl kernel for erode and dilate filter 
This kernel is for CV_8UC1 format and 3x3 kernel size,
It is about 33% ~ 55% faster than current ocl kernel with below perf test

python ./modules/ts/misc/run.py -t imgproc --gtest_filter=OCL_ErodeFixture*
python ./modules/ts/misc/run.py -t imgproc --gtest_filter=OCL_DilateFixture*

Also add accuracy test cases for this kernel, the test command is

./bin/opencv_test_imgproc --gtest_filter=OCL_Filter/MorphFilter3x3*

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-11-04 12:24:24 +08:00
Tetragramm
17df65e666 Fix the OpenCL portion to match the c++ code. 
Fix an undiscovered bug in the c++ code.
 2016-11-03 20:41:16 -05:00
Vadim Pisarevsky
3dc022ae84 Merge pull request #7423 from grahamfyffe:leading_edges 2016-11-03 11:20:43 +00:00
Vadim Pisarevsky
6fceb1dd59 Merge pull request #7601 from sovrasov:remap_big_img_assert 2016-11-02 14:17:04 +00:00
Vadim Pisarevsky
bebd49d91b Merge pull request #7467 from tomoaki0705:featureCheckSimdUniversal 2016-11-02 12:30:55 +00:00
Vladislav Sovrasov
af01267f23 Add assertion to prevent processing of large images in remap 2016-11-02 14:17:56 +03:00
Vadim Pisarevsky
8944875258 Merge pull request #7526 from alalek:fix_arm_builds 2016-11-02 10:49:43 +00:00
Vadim Pisarevsky
94d879737b Merge pull request #7582 from K-Shinotsuka:issue32 2016-11-02 10:48:29 +00:00
Vadim Pisarevsky
01f355dfaa Merge pull request #7583 from K-Shinotsuka:issue33 2016-11-02 10:47:33 +00:00
k-shinotsuka
a7db950b66 add SSE code for Lab2RGB_f. 2016-11-01 22:33:58 +09:00
abratchik
f978ee613e fix for #7510, #7511 and #7512 2016-10-31 11:07:56 +04:00
Alexander Alekhin
4c66772783 Merge pull request #7516 from sovrasov:find_contours_fix 2016-10-30 08:33:12 +00:00
k-shinotsuka
c1c7c7f83e add SSE code for Luv2RGB_f. 2016-10-30 14:50:18 +09:00
Vadim Pisarevsky
7b532cea89 Merge pull request #7540 from K-Shinotsuka:issue31 2016-10-29 21:28:25 +00:00
Vadim Pisarevsky
2b7866f21b Merge pull request #7503 from pengli:box_filter_v2 2016-10-29 21:20:06 +00:00
k-shinotsuka
992795d47d add SSE code for RGB2Luv_f. 2016-10-29 15:28:31 +09:00
Tomoaki Teshima
cba22349b7 add universal hardware support check function 
* use hasSIMD128 rather than calling checkHardwareSupport
  * add SIMD check in spartialgradient.cpp
  * add SIMD check in stereosgbm.cpp
  * add SIMD check in canny.cpp
 2016-10-29 13:24:31 +09:00
Vladislav Sovrasov
56c133d459 Disable redundant border initialization in findContours 2016-10-28 11:05:54 +03:00
Pavel Vlasov
f07525031b BORDER_ISOLATED support for Laplacian, boxFilter and morphology; 2016-10-27 09:53:19 +03:00
Li Peng
3607da9f6b ocl kernel performance optimization for box filter 
The optimization is for CV_8UC1 format and 3x3 box filter,
it is 15%~87% faster than current ocl kernel with below perf test

./modules/ts/misc/run.py -t imgproc --gtest_filter=OCL_BlurFixture*

Also add test cases for this ocl kernel.

Signed-off-by: Li Peng <peng.li@intel.com>
 2016-10-26 11:56:11 +08:00
Vladislav Sovrasov
34df4ae02a Add a regression test, fix documentation 2016-10-25 11:38:07 +03:00
Vladislav Sovrasov
1f4f50b9dc findContours: integrate workaroung to allow contours detection on image border 2016-10-25 11:38:07 +03:00
Vladislav Sovrasov
f5592fd21b Fix wrong default mask value in floodFill 2016-10-25 11:16:59 +03:00
Tetragramm
c6772a8f5d Fix CLAHE distribution. 
Fix CLAHE for 16-bit images.
 2016-10-21 19:54:42 -05:00
Alexander Alekhin
7a9ed39655 test: update HoughLines perf test 2016-10-20 16:54:35 +03:00
LukeZhu
ef47bcc88b Fix the problem: filterSmall.cl report error with double 2016-10-17 15:12:42 +08:00
Vitaly Tuzov
26c9889c6b Fix for incorrect line drawing beyond 32768 row or column 2016-10-14 18:37:59 +03:00
Vadim Pisarevsky
4acda3f609 Merge pull request #7428 from alalek:cmake_fix_compiler_flags_detection 2016-10-12 19:38:25 +00:00
Alexander Alekhin
994815fbd4 Merge pull request #7413 from tomoaki0705:featureUniversalThreshold 2016-10-12 10:01:37 +00:00
Juha Reunanen
0f387cda8f Merge pull request #7451 from reunanen:issue-7409 
Fix findContours crash for very large images (#7451)

* Cast step to size_t in order to avoid integer overflow when processing very large images

* Change assert to CV_Assert
 2016-10-12 10:34:51 +03:00
Tomoaki Teshima
ea6410d1e7 use universal intrinsic in threshold 
* add performance test for 32F and 64F threshold
  * requires update of opencv_extra
 2016-10-11 18:00:41 +09:00
Alexander Alekhin
218f7574f9 Merge pull request #7440 from mschoeneck:Prefer_OCL_befor_IPP_for_Sobel_and_Scharr 2016-10-10 10:45:27 +00:00
StevenPuttemans
947f32d2e0 add border replicate note 2016-10-10 09:55:13 +02:00
matze
d9f6a06773 Adding check for IPP if UMat and OpenCL is available in Sobel and Scharr. 2016-10-09 15:37:22 +02:00
Alexander Alekhin
5da8d65371 eliminate compiler warnings 2016-10-08 02:19:43 +03:00
Graham Fyffe
57be99ff18 Added getLeadingEdges function to subdivision2d 2016-10-06 18:01:34 -07:00
Vadim Pisarevsky
84699e0e18 Merge pull request #7399 from K-Shinotsuka:issue27 2016-10-06 13:31:06 +00:00
Vadim Pisarevsky
294bea11a5 Merge pull request #7401 from K-Shinotsuka:issue28 2016-10-06 13:30:31 +00:00
Vadim Pisarevsky
8ee459207e Merge pull request #7402 from K-Shinotsuka:issue29 2016-10-06 13:29:56 +00:00
Vadim Pisarevsky
285e6ad15b Merge pull request #7404 from K-Shinotsuka:issue30 2016-10-06 13:29:00 +00:00
k-shinotsuka
880ace2eb3 add simd code when dcn equals 4 at HSV2RGB_b(). 2016-10-06 00:06:02 +09:00
k-shinotsuka
db8b3252bf add simd code when scn equals 4 at RGB2HLS_b(). 2016-10-05 22:01:33 +09:00
k-shinotsuka
c1745539c4 add simd code when dcn equals 4 at HLS2RGB_b(). 2016-10-05 21:02:53 +09:00
Vadim Pisarevsky
42fbfb9be0 Merge pull request #7311 from fran6co:lsd 2016-10-05 11:27:43 +00:00
Vadim Pisarevsky
43b60f4617 Merge pull request #6961 from K-Shinotsuka:issue5 2016-10-05 11:23:35 +00:00
Vadim Pisarevsky
33ce36397c Merge pull request #7222 from mschoeneck:Improve_findContours_with_SSE2 2016-10-05 11:21:37 +00:00
Vadim Pisarevsky
2b01e316b3 Merge pull request #7392 from K-Shinotsuka:issue25 2016-10-05 11:19:43 +00:00
k-shinotsuka
a5f493a763 add simd code when scn equals 4 at Lab2RGB_b(). 2016-10-05 20:14:02 +09:00
k-shinotsuka
47c792ecc0 add simd code when scn equals 4 at RGB2Luv_b(). 2016-10-05 01:18:43 +09:00
Vadim Pisarevsky
eddd417d74 Merge pull request #7282 from alalek:fix_calculations 2016-10-04 15:05:42 +00:00