Seon-Wook Park
20bf88bad1
spatialGradient: Make kern args const&
2015-07-01 09:34:27 +09:00
Seon-Wook Park
90c398ea67
spatialGradient: Add CV_SSE2 check
2015-07-01 00:42:08 +09:00
Seon-Wook Park
ed38ca5179
spatialGradient: Remove 4 loads in inner loop
2015-06-30 16:08:15 +09:00
Seon-Wook Park
5dddb47863
spatialGradient: Remove pointers caching
2015-06-30 10:51:10 +09:00
Seon-Wook Park
cf0fdfa2bb
spatialGradient: Change ordering of vector loads
2015-06-29 23:50:05 +09:00
Seon-Wook Park
658f96b447
spatialGradient: L/R border handling outside. Kernelize.
2015-06-29 23:12:33 +09:00
Seon-Wook Park
15ea401087
spatialGradient: Move vector decl into loop
2015-06-27 10:01:47 +02:00
Seon-Wook Park
db0cc56c52
spatialGradient: Re-introduce 2-rows at a time
2015-06-27 09:53:42 +02:00
Seon-Wook Park
f958f29c55
spatialGradient: Suppress uninitialised j warnings
2015-06-27 01:03:43 +02:00
Seon-Wook Park
62cad09c64
spatialGradient: Process 1 row at a time in SSE
2015-06-26 17:35:17 +02:00
Seon-Wook Park
8a21726ae5
spatialGradient: Remove an unnecessary branch in nosse code
2015-06-26 17:26:21 +02:00
Seon-Wook Park
febd0f14c6
spatialGradient: Don't dynamically alloc C array, use vector
2015-06-26 17:15:44 +02:00
Seon-Wook Park
f92e2ed57a
spatialGradient: Make nosse version faster
2015-06-26 17:13:23 +02:00
Seon-Wook Park
7b01e32fe8
spatialGradient: HAL-accelerated
...
TODO: Make nosse ver faster than Sobel
Make sse ver faster than Sobel for BORDER_REPLICATE
2015-06-26 16:41:00 +02:00
Seon-Wook Park
6803d1ed28
Support non continuous, BORDER_REPLICATE
...
TODO: HAL-accelerated code
2015-06-26 14:49:31 +02:00
Seon-Wook Park
2ff614dfab
spatialGradient: Per row in outer loop
2015-06-19 19:25:07 +02:00
Seon-Wook Park
e633c991b0
spatialGradient: Doc, fix dangling newline error
2015-06-19 04:57:01 +02:00
Seon-Wook Park
815cd8970d
spatialGradient: Remove unnecessary index calculation
2015-06-19 04:46:17 +02:00
Seon-Wook Park
b5c4355c13
spatialGradient: Add basic perf test
2015-06-19 04:29:28 +02:00
Seon-Wook Park
f9c4c96663
spatialGradient: Reduce temporary vectors
2015-06-19 04:29:07 +02:00
Seon-Wook Park
a2dbd2f10e
spatialGradient: Less vector loads
2015-06-19 04:07:18 +02:00
Seon-Wook Park
88bc88125a
spatialGradient: Vectorise inner area
2015-06-19 03:36:49 +02:00
Seon-Wook Park
770e742e04
spatialGradient: Add non-SSE version
2015-06-19 02:56:51 +02:00
Seon-Wook Park
11fb1f74cc
spatialGradient: Add asserts
2015-06-19 01:23:01 +02:00
Seon-Wook Park
9f1c641199
spatialGradient: Add test class and Sobel proxy method
2015-06-18 17:42:32 +02:00
Vadim Pisarevsky
01c3ed6157
Merge pull request #4108 from Reen:patch-1
2015-06-16 13:26:47 +00:00
Suleyman TURKMEN
caa0058d5b
Update imgproc.hpp
2015-06-11 11:46:42 +03:00
Suleyman TURKMEN
dac715ad1f
Update imgproc.hpp
2015-06-11 11:03:14 +03:00
Suleyman TURKMEN
061131a5ee
Update imgproc.hpp
2015-06-10 13:18:40 +03:00
René
3e2515d735
Fix bug in distanceATS_L1_8u and typos.
...
The inner loop of the backward scan got the wrong initial "a".
2015-06-09 17:23:22 +02:00
Vadim Pisarevsky
b46719b093
Merge pull request #4074 from vpisarev:objdetect_fixes
2015-05-28 19:43:51 +00:00
Vadim Pisarevsky
5a94a95fbf
improvements in Haar CascadeClassifier: 1) use CV_32S instead of CV_32F for the integral of squares (which is more accurate and more efficient); 2) skip the window if its contrast is too low
2015-05-28 19:33:21 +03:00
S. Garrido
bd2c2f3bd1
fix isContourConvex
2015-05-27 20:51:53 +02:00
Zhigang Gong
0f7de40e66
Fixed the race condition between inc and dec on the l_counter.
...
Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
2015-05-26 22:06:18 +08:00
Zhigang Gong
3c85200989
Avoid negative index for a local buffer in Canny.cl.
...
int pix_per_thr = l_counter / LOCAL_TOTAL + ((lid < mod) ? 1 : 0);
The pix_per_thr * LOCAL_TOTAL may be larger than l_counter.
Thus the index of l_stack may be negative which may cause serious
problems. Let's skip the loop when we get negative index and we need
to add back the lcounter to keep its balance and avoid potential
negative counter.
Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
2015-05-26 08:48:05 +08:00
Andreas Stührk
3ec0e0943c
TIFF loader: Allocate large enough buffer when (bpp * ncn) > 8.
...
Conflicts:
modules/highgui/src/grfmt_tiff.cpp
floodfill return value removed
2015-05-24 14:06:33 +02:00
Alexander Smorkalov
6e1a595342
GCC 4.8 warning array subscript is above array bounds fixed.
...
tiff test fixed
2015-05-24 13:31:01 +02:00
Dikay900
7ca0557b40
typos in comments
2015-05-23 17:36:45 +02:00
Vadim Pisarevsky
e0136e39f4
Merge pull request #4011 from cr333:parula
2015-05-16 19:11:32 +00:00
Vadim Pisarevsky
a85bb091c8
Merge pull request #4018 from vpisarev:imgproc_video_fixes
2015-05-15 16:15:01 +00:00
Vadim Pisarevsky
c52092e6f8
Merge pull request #3992 from nomarek:master
2015-05-14 18:24:09 +00:00
Vadim Pisarevsky
ffabbfa778
added test to prove that remap does not leak memory ( http://code.opencv.org/issues/2502 ). disabled the test for now to save execution time.
2015-05-14 18:21:26 +03:00
Vadim Pisarevsky
ca90667723
fixed compile warnings on Linux and Windows
2015-05-14 16:42:51 +03:00
Vadim Pisarevsky
05d888316a
added test for http://code.opencv.org/issues/2957
2015-05-14 16:25:18 +03:00
Vadim Pisarevsky
feb5b6aa93
increased singularity epsilon in LU decomposition. This solved singular case from http://code.opencv.org/issues/3305 . Added the respective test.
2015-05-14 10:42:55 +03:00
Vadim Pisarevsky
2e7e754032
added support for n-channel (n>4) images in warpAffine/warpPerspective/remap: http://code.opencv.org/issues/4190
2015-05-14 08:06:46 +03:00
Vadim Pisarevsky
d3b0cb878a
added test for http://code.opencv.org/issues/2736
2015-05-13 22:16:35 +03:00
Vadim Pisarevsky
71002e0e79
some fixes from http://code.opencv.org/issues/3733
2015-05-13 22:00:02 +03:00
Marek Nogacki
25bcca2edb
fixed http://code.opencv.org/issues/4276 - set drawing thickness limit to 32767
2015-05-13 18:54:25 +02:00
Maksim Shabunin
61293a09ff
Fixed RGB-to-HLS conversion formula in documentation
2015-05-13 18:46:21 +03:00
