mirror/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2025-06-11 20:09:23 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
920fd48228
opencv/modules/ocl/src/imgproc.cpp
Roman Donchenko 920fd48228 Merge remote-tracking branch 'origin/master' into merge-2.4 
Conflicts:
	modules/ocl/src/arithm.cpp
2013-11-26 15:32:44 +04:00

1965 lines
91 KiB
C++
Raw Blame History

/*M///////////////////////////////////////////////////////////////////////////////////////
//
// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
//
// By downloading, copying, installing or using the software you agree to this license.
// If you do not agree to this license, do not download, install,
// copy or use the software.
//
//
// License Agreement
// For Open Source Computer Vision Library
//
// Copyright (C) 2010-2012, Institute Of Software Chinese Academy Of Science, all rights reserved.
// Copyright (C) 2010-2012, Advanced Micro Devices, Inc., all rights reserved.
// Copyright (C) 2010-2012, Multicoreware, Inc., all rights reserved.
// Third party copyrights are property of their respective owners.
//
// @Authors
// Niko Li, newlife20080214@gmail.com
// Jia Haipeng, jiahaipeng95@gmail.com
// Shengen Yan, yanshengen@gmail.com
// Rock Li, Rock.Li@amd.com
// Zero Lin, Zero.Lin@amd.com
// Zhang Ying, zhangying913@gmail.com
// Xu Pang, pangxu010@163.com
// Wu Zailong, bullet@yeah.net
// Wenju He, wenju@multicorewareinc.com
// Peng Xiao, pengxiao@outlook.com
// Sen Liu, swjtuls1987@126.com
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
// * Redistribution's of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// * Redistribution's in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other materials provided with the distribution.
//
// * The name of the copyright holders may not be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// This software is provided by the copyright holders and contributors "as is" and
// any express or implied warranties, including, but not limited to, the implied
// warranties of merchantability and fitness for a particular purpose are disclaimed.
// In no event shall the Intel Corporation or contributors be liable for any direct,
// indirect, incidental, special, exemplary, or consequential damages
// (including, but not limited to, procurement of substitute goods or services;
// loss of use, data, or profits; or business interruption) however caused
// and on any theory of liability, whether in contract, strict liability,
// or tort (including negligence or otherwise) arising in any way out of
// the use of this software, even if advised of the possibility of such damage.
//
//M*/
#include "precomp.hpp"
#include "opencl_kernels.hpp"
using namespace cv;
using namespace cv::ocl;
namespace cv
{
namespace ocl
{
////////////////////////////////////OpenCL call wrappers////////////////////////////
template <typename T> struct index_and_sizeof;
template <> struct index_and_sizeof<char>
{
enum { index = 1 };
};
template <> struct index_and_sizeof<unsigned char>
{
enum { index = 2 };
};
template <> struct index_and_sizeof<short>
{
enum { index = 3 };
};
template <> struct index_and_sizeof<unsigned short>
{
enum { index = 4 };
};
template <> struct index_and_sizeof<int>
{
enum { index = 5 };
};
template <> struct index_and_sizeof<float>
{
enum { index = 6 };
};
template <> struct index_and_sizeof<double>
{
enum { index = 7 };
};
/////////////////////////////////////////////////////////////////////////////////////
// threshold
static std::vector<uchar> scalarToVector(const cv::Scalar & sc, int depth, int ocn, int cn)
{
CV_Assert(ocn == cn || (ocn == 4 && cn == 3));
static const int sizeMap[] = { sizeof(uchar), sizeof(char), sizeof(ushort),
sizeof(short), sizeof(int), sizeof(float), sizeof(double) };
int elemSize1 = sizeMap[depth];
int bufSize = elemSize1 * ocn;
std::vector<uchar> _buf(bufSize);
uchar * buf = &_buf[0];
scalarToRawData(sc, buf, CV_MAKE_TYPE(depth, cn));
memset(buf + elemSize1 * cn, 0, (ocn - cn) * elemSize1);
return _buf;
}
static void threshold_runner(const oclMat &src, oclMat &dst, double thresh, double maxVal, int thresholdType)
{
bool ival = src.depth() < CV_32F;
int cn = src.channels(), vecSize = 4, depth = src.depth();
std::vector<uchar> thresholdValue = scalarToVector(cv::Scalar::all(ival ? cvFloor(thresh) : thresh), dst.depth(),
dst.oclchannels(), dst.channels());
std::vector<uchar> maxValue = scalarToVector(cv::Scalar::all(maxVal), dst.depth(), dst.oclchannels(), dst.channels());
const char * const thresholdMap[] = { "THRESH_BINARY", "THRESH_BINARY_INV", "THRESH_TRUNC",
"THRESH_TOZERO", "THRESH_TOZERO_INV" };
const char * const channelMap[] = { "", "", "2", "4", "4" };
const char * const typeMap[] = { "uchar", "char", "ushort", "short", "int", "float", "double" };
std::string buildOptions = format("-D T=%s%s -D %s", typeMap[depth], channelMap[cn], thresholdMap[thresholdType]);
int elemSize = src.elemSize();
int src_step = src.step / elemSize, src_offset = src.offset / elemSize;
int dst_step = dst.step / elemSize, dst_offset = dst.offset / elemSize;
std::vector< std::pair<size_t, const void *> > args;
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&src.data));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src_offset));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src_step));
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&dst.data));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst_offset));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst_step));
args.push_back( std::make_pair(thresholdValue.size(), (void *)&thresholdValue[0]));
args.push_back( std::make_pair(maxValue.size(), (void *)&maxValue[0]));
int max_index = dst.cols, cols = dst.cols;
if (cn == 1 && vecSize > 1)
{
CV_Assert(((vecSize - 1) & vecSize) == 0 && vecSize <= 16);
cols = divUp(cols, vecSize);
buildOptions += format(" -D VECTORIZED -D VT=%s%d -D VLOADN=vload%d -D VECSIZE=%d -D VSTOREN=vstore%d",
typeMap[depth], vecSize, vecSize, vecSize, vecSize);
int vecSizeBytes = vecSize * dst.elemSize1();
if ((dst.offset % dst.step) % vecSizeBytes == 0 && dst.step % vecSizeBytes == 0)
buildOptions += " -D DST_ALIGNED";
if ((src.offset % src.step) % vecSizeBytes == 0 && src.step % vecSizeBytes == 0)
buildOptions += " -D SRC_ALIGNED";
args.push_back( std::make_pair(sizeof(cl_int), (void *)&max_index));
}
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.rows));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&cols));
size_t localThreads[3] = { 16, 16, 1 };
size_t globalThreads[3] = { cols, dst.rows, 1 };
openCLExecuteKernel(src.clCxt, &imgproc_threshold, "threshold", globalThreads, localThreads, args,
-1, -1, buildOptions.c_str());
}
double threshold(const oclMat &src, oclMat &dst, double thresh, double maxVal, int thresholdType)
{
CV_Assert(thresholdType == THRESH_BINARY || thresholdType == THRESH_BINARY_INV || thresholdType == THRESH_TRUNC
|| thresholdType == THRESH_TOZERO || thresholdType == THRESH_TOZERO_INV);
dst.create(src.size(), src.type());
threshold_runner(src, dst, thresh, maxVal, thresholdType);
return thresh;
}
////////////////////////////////////////////////////////////////////////////////////////////
/////////////////////////////// remap //////////////////////////////////////////////////
////////////////////////////////////////////////////////////////////////////////////////////
void remap( const oclMat &src, oclMat &dst, oclMat &map1, oclMat &map2, int interpolation, int borderType, const Scalar &borderValue )
{
Context *clCxt = src.clCxt;
bool supportsDouble = clCxt->supportsFeature(FEATURE_CL_DOUBLE);
if (!supportsDouble && src.depth() == CV_64F)
{
CV_Error(CV_OpenCLDoubleNotSupported, "Selected device does not support double");
return;
}
if (map1.empty())
map1.swap(map2);
CV_Assert(interpolation == INTER_LINEAR || interpolation == INTER_NEAREST);
CV_Assert((map1.type() == CV_16SC2 && (map2.empty() || (map2.type() == CV_16UC1 || map2.type() == CV_16SC1)) ) ||
(map1.type() == CV_32FC2 && !map2.data) ||
(map1.type() == CV_32FC1 && map2.type() == CV_32FC1));
CV_Assert(!map2.data || map2.size() == map1.size());
CV_Assert(borderType == BORDER_CONSTANT || borderType == BORDER_REPLICATE || borderType == BORDER_WRAP
|| borderType == BORDER_REFLECT_101 || borderType == BORDER_REFLECT);
dst.create(map1.size(), src.type());
const char * const typeMap[] = { "uchar", "char", "ushort", "short", "int", "float", "double" };
const char * const channelMap[] = { "", "", "2", "4", "4" };
const char * const interMap[] = { "INTER_NEAREST", "INTER_LINEAR", "INTER_CUBIC", "INTER_LINEAR", "INTER_LANCZOS" };
const char * const borderMap[] = { "BORDER_CONSTANT", "BORDER_REPLICATE", "BORDER_REFLECT", "BORDER_WRAP",
"BORDER_REFLECT_101", "BORDER_TRANSPARENT" };
String kernelName = "remap";
if (map1.type() == CV_32FC2 && map2.empty())
kernelName += "_32FC2";
else if (map1.type() == CV_16SC2)
{
kernelName += "_16SC2";
if (!map2.empty())
kernelName += "_16UC1";
}
else if (map1.type() == CV_32FC1 && map2.type() == CV_32FC1)
kernelName += "_2_32FC1";
else
CV_Error(Error::StsBadArg, "Unsupported map types");
int ocn = dst.oclchannels();
size_t globalThreads[3] = { dst.cols, dst.rows, 1 };
Mat scalar(1, 1, CV_MAKE_TYPE(dst.depth(), ocn), borderValue);
String buildOptions = format("-D %s -D %s -D T=%s%s", interMap[interpolation],
borderMap[borderType], typeMap[src.depth()], channelMap[ocn]);
if (interpolation != INTER_NEAREST)
{
int wdepth = std::max(CV_32F, dst.depth());
buildOptions = buildOptions
+ format(" -D WT=%s%s -D convertToT=convert_%s%s%s -D convertToWT=convert_%s%s"
" -D convertToWT2=convert_%s2 -D WT2=%s2",
typeMap[wdepth], channelMap[ocn],
typeMap[src.depth()], channelMap[ocn], src.depth() < CV_32F ? "_sat_rte" : "",
typeMap[wdepth], channelMap[ocn],
typeMap[wdepth], typeMap[wdepth]);
}
int src_step = src.step / src.elemSize(), src_offset = src.offset / src.elemSize();
int map1_step = map1.step / map1.elemSize(), map1_offset = map1.offset / map1.elemSize();
int map2_step = map2.step / map2.elemSize(), map2_offset = map2.offset / map2.elemSize();
int dst_step = dst.step / dst.elemSize(), dst_offset = dst.offset / dst.elemSize();
std::vector< std::pair<size_t, const void *> > args;
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&src.data));
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&dst.data));
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&map1.data));
if (!map2.empty())
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&map2.data));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src_offset));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst_offset));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&map1_offset));
if (!map2.empty())
args.push_back( std::make_pair(sizeof(cl_int), (void *)&map2_offset));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src_step));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst_step));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&map1_step));
if (!map2.empty())
args.push_back( std::make_pair(sizeof(cl_int), (void *)&map2_step));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src.cols));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src.rows));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.cols));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.rows));
args.push_back( std::make_pair(scalar.elemSize(), (void *)scalar.data));
#ifdef ANDROID
openCLExecuteKernel(clCxt, &imgproc_remap, kernelName, globalThreads, NULL, args, -1, -1, buildOptions.c_str());
#else
size_t localThreads[3] = { 256, 1, 1 };
openCLExecuteKernel(clCxt, &imgproc_remap, kernelName, globalThreads, localThreads, args, -1, -1, buildOptions.c_str());
#endif
}
////////////////////////////////////////////////////////////////////////////////////////////
// resize
static void computeResizeAreaTabs(int ssize, int dsize, double scale, int * const map_tab,
float * const alpha_tab, int * const ofs_tab)
{
int k = 0, dx = 0;
for ( ; dx < dsize; dx++)
{
ofs_tab[dx] = k;
double fsx1 = dx * scale;
double fsx2 = fsx1 + scale;
double cellWidth = std::min(scale, ssize - fsx1);
int sx1 = cvCeil(fsx1), sx2 = cvFloor(fsx2);
sx2 = std::min(sx2, ssize - 1);
sx1 = std::min(sx1, sx2);
if (sx1 - fsx1 > 1e-3)
{
map_tab[k] = sx1 - 1;
alpha_tab[k++] = (float)((sx1 - fsx1) / cellWidth);
}
for (int sx = sx1; sx < sx2; sx++)
{
map_tab[k] = sx;
alpha_tab[k++] = float(1.0 / cellWidth);
}
if (fsx2 - sx2 > 1e-3)
{
map_tab[k] = sx2;
alpha_tab[k++] = (float)(std::min(std::min(fsx2 - sx2, 1.), cellWidth) / cellWidth);
}
}
ofs_tab[dx] = k;
}
static void computeResizeAreaFastTabs(int * dmap_tab, int * smap_tab, int scale, int dcols, int scol)
{
for (int i = 0; i < dcols; ++i)
dmap_tab[i] = scale * i;
for (int i = 0, size = dcols * scale; i < size; ++i)
smap_tab[i] = std::min(scol - 1, i);
}
static void resize_gpu( const oclMat &src, oclMat &dst, double ifx, double ify, int interpolation)
{
float ifxf = (float)ifx, ifyf = (float)ify;
int src_step = src.step / src.elemSize(), src_offset = src.offset / src.elemSize();
int dst_step = dst.step / dst.elemSize(), dst_offset = dst.offset / dst.elemSize();
int ocn = dst.oclchannels(), depth = dst.depth();
const char * const interMap[] = { "NN", "LN", "CUBIC", "AREA", "LAN4" };
std::string kernelName = std::string("resize") + interMap[interpolation];
const char * const typeMap[] = { "uchar", "char", "ushort", "short", "int", "float", "double" };
const char * const channelMap[] = { "" , "", "2", "4", "4" };
std::string buildOption = format("-D %s -D T=%s%s", interMap[interpolation], typeMap[depth], channelMap[ocn]);
int wdepth = std::max(src.depth(), CV_32F);
// check if fx, fy is integer and then we have inter area fast mode
int iscale_x = saturate_cast<int>(ifx);
int iscale_y = saturate_cast<int>(ify);
bool is_area_fast = std::abs(ifx - iscale_x) < DBL_EPSILON &&
std::abs(ify - iscale_y) < DBL_EPSILON;
if (is_area_fast)
wdepth = std::max(src.depth(), CV_32S);
if (interpolation != INTER_NEAREST)
{
buildOption += format(" -D WT=%s -D WTV=%s%s -D convertToWTV=convert_%s%s -D convertToT=convert_%s%s%s",
typeMap[wdepth], typeMap[wdepth], channelMap[ocn],
typeMap[wdepth], channelMap[ocn],
typeMap[src.depth()], channelMap[ocn], src.depth() <= CV_32S ? "_sat_rte" : "");
}
#ifdef ANDROID
size_t blkSizeX = 16, blkSizeY = 8;
#else
size_t blkSizeX = 16, blkSizeY = 16;
#endif
size_t glbSizeX;
if (src.type() == CV_8UC1 && interpolation == INTER_LINEAR)
{
size_t cols = (dst.cols + dst.offset % 4 + 3) / 4;
glbSizeX = cols % blkSizeX == 0 && cols != 0 ? cols : (cols / blkSizeX + 1) * blkSizeX;
}
else
glbSizeX = dst.cols;
oclMat alphaOcl, mapOcl, tabofsOcl;
if (interpolation == INTER_AREA)
{
if (is_area_fast)
{
kernelName += "_FAST";
int wdepth2 = std::max(CV_32F, src.depth());
buildOption += format(" -D WT2V=%s%s -D convertToWT2V=convert_%s%s -D AREA_FAST -D XSCALE=%d -D YSCALE=%d -D SCALE=%f",
typeMap[wdepth2], channelMap[ocn], typeMap[wdepth2], channelMap[ocn],
iscale_x, iscale_y, 1.0f / (iscale_x * iscale_y));
int smap_tab_size = dst.cols * iscale_x + dst.rows * iscale_y;
AutoBuffer<int> dmap_tab(dst.cols + dst.rows), smap_tab(smap_tab_size);
int * dxmap_tab = dmap_tab, * dymap_tab = dxmap_tab + dst.cols;
int * sxmap_tab = smap_tab, * symap_tab = smap_tab + dst.cols * iscale_y;
computeResizeAreaFastTabs(dxmap_tab, sxmap_tab, iscale_x, dst.cols, src.cols);
computeResizeAreaFastTabs(dymap_tab, symap_tab, iscale_y, dst.rows, src.rows);
tabofsOcl = oclMat(1, dst.cols + dst.rows, CV_32SC1, (void *)dmap_tab);
mapOcl = oclMat(1, smap_tab_size, CV_32SC1, (void *)smap_tab);
}
else
{
Size ssize = src.size(), dsize = dst.size();
int xytab_size = (ssize.width + ssize.height) << 1;
int tabofs_size = dsize.height + dsize.width + 2;
AutoBuffer<int> _xymap_tab(xytab_size), _xyofs_tab(tabofs_size);
AutoBuffer<float> _xyalpha_tab(xytab_size);
int * xmap_tab = _xymap_tab, * ymap_tab = _xymap_tab + (ssize.width << 1);
float * xalpha_tab = _xyalpha_tab, * yalpha_tab = _xyalpha_tab + (ssize.width << 1);
int * xofs_tab = _xyofs_tab, * yofs_tab = _xyofs_tab + dsize.width + 1;
computeResizeAreaTabs(ssize.width, dsize.width, ifx, xmap_tab, xalpha_tab, xofs_tab);
computeResizeAreaTabs(ssize.height, dsize.height, ify, ymap_tab, yalpha_tab, yofs_tab);
// loading precomputed arrays to GPU
alphaOcl = oclMat(1, xytab_size, CV_32FC1, (void *)_xyalpha_tab);
mapOcl = oclMat(1, xytab_size, CV_32SC1, (void *)_xymap_tab);
tabofsOcl = oclMat(1, tabofs_size, CV_32SC1, (void *)_xyofs_tab);
}
}
size_t globalThreads[3] = { glbSizeX, dst.rows, 1 };
size_t localThreads[3] = { blkSizeX, blkSizeY, 1 };
std::vector< std::pair<size_t, const void *> > args;
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&dst.data));
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&src.data));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst_offset));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src_offset));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst_step));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src_step));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src.cols));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&src.rows));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.cols));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.rows));
if (wdepth == CV_64F)
{
args.push_back( std::make_pair(sizeof(cl_double), (void *)&ifx));
args.push_back( std::make_pair(sizeof(cl_double), (void *)&ify));
}
else
{
args.push_back( std::make_pair(sizeof(cl_float), (void *)&ifxf));
args.push_back( std::make_pair(sizeof(cl_float), (void *)&ifyf));
}
// precomputed tabs
if (!tabofsOcl.empty())
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&tabofsOcl.data));
if (!mapOcl.empty())
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&mapOcl.data));
if (!alphaOcl.empty())
args.push_back( std::make_pair(sizeof(cl_mem), (void *)&alphaOcl.data));
ocn = interpolation == INTER_LINEAR ? ocn : -1;
depth = interpolation == INTER_LINEAR ? depth : -1;
openCLExecuteKernel(src.clCxt, &imgproc_resize, kernelName, globalThreads, localThreads, args,
ocn, depth, buildOption.c_str());
}
void resize(const oclMat &src, oclMat &dst, Size dsize, double fx, double fy, int interpolation)
{
if (!src.clCxt->supportsFeature(FEATURE_CL_DOUBLE) && src.depth() == CV_64F)
{
CV_Error(CV_OpenCLDoubleNotSupported, "Selected device does not support double");
return;
}
CV_Assert(src.type() == CV_8UC1 || src.type() == CV_8UC3 || src.type() == CV_8UC4
|| src.type() == CV_32FC1 || src.type() == CV_32FC3 || src.type() == CV_32FC4);
CV_Assert(dsize.area() > 0 || (fx > 0 && fy > 0));
if (dsize.area() == 0)
{
dsize = Size(saturate_cast<int>(src.cols * fx), saturate_cast<int>(src.rows * fy));
CV_Assert(dsize.area() > 0);
}
else
{
fx = (double)dsize.width / src.cols;
fy = (double)dsize.height / src.rows;
}
double inv_fy = 1 / fy, inv_fx = 1 / fx;
CV_Assert(interpolation == INTER_LINEAR || interpolation == INTER_NEAREST ||
(interpolation == INTER_AREA && inv_fx >= 1 && inv_fy >= 1));
dst.create(dsize, src.type());
resize_gpu( src, dst, inv_fx, inv_fy, interpolation);
}
////////////////////////////////////////////////////////////////////////
// medianFilter
void medianFilter(const oclMat &src, oclMat &dst, int m)
{
CV_Assert( m % 2 == 1 && m > 1 );
CV_Assert( (src.depth() == CV_8U || src.depth() == CV_32F) && (src.channels() == 1 || src.channels() == 4));
dst.create(src.size(), src.type());
int srcStep = src.step / src.elemSize(), dstStep = dst.step / dst.elemSize();
int srcOffset = src.offset / src.elemSize(), dstOffset = dst.offset / dst.elemSize();
Context *clCxt = src.clCxt;
std::vector< std::pair<size_t, const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&src.data));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&dst.data));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&srcOffset));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dstOffset));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.cols));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.rows));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&srcStep));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dstStep));
size_t globalThreads[3] = {(src.cols + 18) / 16 * 16, (src.rows + 15) / 16 * 16, 1};
size_t localThreads[3] = {16, 16, 1};
if (m == 3)
{
String kernelName = "medianFilter3";
openCLExecuteKernel(clCxt, &imgproc_median, kernelName, globalThreads, localThreads, args, src.oclchannels(), src.depth());
}
else if (m == 5)
{
String kernelName = "medianFilter5";
openCLExecuteKernel(clCxt, &imgproc_median, kernelName, globalThreads, localThreads, args, src.oclchannels(), src.depth());
}
else
CV_Error(Error::StsBadArg, "Non-supported filter length");
}
////////////////////////////////////////////////////////////////////////
// copyMakeBorder
void copyMakeBorder(const oclMat &src, oclMat &dst, int top, int bottom, int left, int right, int bordertype, const Scalar &scalar)
{
if (!src.clCxt->supportsFeature(FEATURE_CL_DOUBLE) && src.depth() == CV_64F)
{
CV_Error(Error::OpenCLDoubleNotSupported, "Selected device does not support double");
return;
}
oclMat _src = src;
CV_Assert(top >= 0 && bottom >= 0 && left >= 0 && right >= 0);
if( (_src.wholecols != _src.cols || _src.wholerows != _src.rows) && (bordertype & BORDER_ISOLATED) == 0 )
{
Size wholeSize;
Point ofs;
_src.locateROI(wholeSize, ofs);
int dtop = std::min(ofs.y, top);
int dbottom = std::min(wholeSize.height - _src.rows - ofs.y, bottom);
int dleft = std::min(ofs.x, left);
int dright = std::min(wholeSize.width - _src.cols - ofs.x, right);
_src.adjustROI(dtop, dbottom, dleft, dright);
top -= dtop;
left -= dleft;
bottom -= dbottom;
right -= dright;
}
bordertype &= ~cv::BORDER_ISOLATED;
dst.create(_src.rows + top + bottom, _src.cols + left + right, _src.type());
int srcStep = _src.step / _src.elemSize(), dstStep = dst.step / dst.elemSize();
int srcOffset = _src.offset / _src.elemSize(), dstOffset = dst.offset / dst.elemSize();
int depth = _src.depth(), ochannels = _src.oclchannels();
int __bordertype[] = { BORDER_CONSTANT, BORDER_REPLICATE, BORDER_REFLECT, BORDER_WRAP, BORDER_REFLECT_101 };
const char *borderstr[] = { "BORDER_CONSTANT", "BORDER_REPLICATE", "BORDER_REFLECT", "BORDER_WRAP", "BORDER_REFLECT_101" };
int bordertype_index = -1;
for (int i = 0, end = sizeof(__bordertype) / sizeof(int); i < end; i++)
if (__bordertype[i] == bordertype)
{
bordertype_index = i;
break;
}
if (bordertype_index < 0)
CV_Error(Error::StsBadArg, "Unsupported border type");
size_t localThreads[3] = { 16, 16, 1 };
size_t globalThreads[3] = { dst.cols, dst.rows, 1 };
std::vector< std::pair<size_t, const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&_src.data));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&dst.data));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst.cols));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst.rows));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&_src.cols));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&_src.rows));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&srcStep));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&srcOffset));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dstStep));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dstOffset));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&top));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&left));
const char * const typeMap[] = { "uchar", "char", "ushort", "short", "int", "float", "double" };
const char * const channelMap[] = { "", "", "2", "4", "4" };
std::string buildOptions = format("-D GENTYPE=%s%s -D %s",
typeMap[depth], channelMap[ochannels],
borderstr[bordertype_index]);
int cn = src.channels(), ocn = src.oclchannels();
int bufSize = src.elemSize1() * ocn;
AutoBuffer<uchar> _buf(bufSize);
uchar * buf = (uchar *)_buf;
scalarToRawData(scalar, buf, dst.type());
memset(buf + src.elemSize1() * cn, 0, (ocn - cn) * src.elemSize1());
args.push_back( std::make_pair( bufSize , (void *)buf ));
openCLExecuteKernel(src.clCxt, &imgproc_copymakeboder, "copymakeborder", globalThreads,
localThreads, args, -1, -1, buildOptions.c_str());
}
////////////////////////////////////////////////////////////////////////
// warp
namespace
{
#define F double
void convert_coeffs(F *M)
{
double D = M[0] * M[4] - M[1] * M[3];
D = D != 0 ? 1. / D : 0;
double A11 = M[4] * D, A22 = M[0] * D;
M[0] = A11;
M[1] *= -D;
M[3] *= -D;
M[4] = A22;
double b1 = -M[0] * M[2] - M[1] * M[5];
double b2 = -M[3] * M[2] - M[4] * M[5];
M[2] = b1;
M[5] = b2;
}
double invert(double *M)
{
#define Sd(y,x) (Sd[y*3+x])
#define Dd(y,x) (Dd[y*3+x])
#define det3(m) (m(0,0)*(m(1,1)*m(2,2) - m(1,2)*m(2,1)) - \
m(0,1)*(m(1,0)*m(2,2) - m(1,2)*m(2,0)) + \
m(0,2)*(m(1,0)*m(2,1) - m(1,1)*m(2,0)))
double *Sd = M;
double *Dd = M;
double d = det3(Sd);
double result = 0;
if ( d != 0)
{
double t[9];
result = d;
d = 1. / d;
t[0] = (Sd(1, 1) * Sd(2, 2) - Sd(1, 2) * Sd(2, 1)) * d;
t[1] = (Sd(0, 2) * Sd(2, 1) - Sd(0, 1) * Sd(2, 2)) * d;
t[2] = (Sd(0, 1) * Sd(1, 2) - Sd(0, 2) * Sd(1, 1)) * d;
t[3] = (Sd(1, 2) * Sd(2, 0) - Sd(1, 0) * Sd(2, 2)) * d;
t[4] = (Sd(0, 0) * Sd(2, 2) - Sd(0, 2) * Sd(2, 0)) * d;
t[5] = (Sd(0, 2) * Sd(1, 0) - Sd(0, 0) * Sd(1, 2)) * d;
t[6] = (Sd(1, 0) * Sd(2, 1) - Sd(1, 1) * Sd(2, 0)) * d;
t[7] = (Sd(0, 1) * Sd(2, 0) - Sd(0, 0) * Sd(2, 1)) * d;
t[8] = (Sd(0, 0) * Sd(1, 1) - Sd(0, 1) * Sd(1, 0)) * d;
Dd(0, 0) = t[0];
Dd(0, 1) = t[1];
Dd(0, 2) = t[2];
Dd(1, 0) = t[3];
Dd(1, 1) = t[4];
Dd(1, 2) = t[5];
Dd(2, 0) = t[6];
Dd(2, 1) = t[7];
Dd(2, 2) = t[8];
}
return result;
}
void warpAffine_gpu(const oclMat &src, oclMat &dst, F coeffs[2][3], int interpolation)
{
CV_Assert( (src.oclchannels() == dst.oclchannels()) );
int srcStep = src.step1();
int dstStep = dst.step1();
float float_coeffs[2][3];
cl_mem coeffs_cm;
Context *clCxt = src.clCxt;
String s[3] = {"NN", "Linear", "Cubic"};
String kernelName = "warpAffine" + s[interpolation];
if (src.clCxt->supportsFeature(FEATURE_CL_DOUBLE))
{
cl_int st;
coeffs_cm = clCreateBuffer(*(cl_context*)clCxt->getOpenCLContextPtr(), CL_MEM_READ_WRITE, sizeof(F) * 2 * 3, NULL, &st );
openCLVerifyCall(st);
openCLSafeCall(clEnqueueWriteBuffer(*(cl_command_queue*)clCxt->getOpenCLCommandQueuePtr(), (cl_mem)coeffs_cm, 1, 0,
sizeof(F) * 2 * 3, coeffs, 0, 0, 0));
}
else
{
cl_int st;
for(int m = 0; m < 2; m++)
for(int n = 0; n < 3; n++)
float_coeffs[m][n] = coeffs[m][n];
coeffs_cm = clCreateBuffer(*(cl_context*)clCxt->getOpenCLContextPtr(), CL_MEM_READ_WRITE, sizeof(float) * 2 * 3, NULL, &st );
openCLSafeCall(clEnqueueWriteBuffer(*(cl_command_queue*)clCxt->getOpenCLCommandQueuePtr(), (cl_mem)coeffs_cm,
1, 0, sizeof(float) * 2 * 3, float_coeffs, 0, 0, 0));
}
//TODO: improve this kernel
#ifdef ANDROID
size_t blkSizeX = 16, blkSizeY = 4;
#else
size_t blkSizeX = 16, blkSizeY = 16;
#endif
size_t glbSizeX;
size_t cols;
if (src.type() == CV_8UC1 && interpolation != 2)
{
cols = (dst.cols + dst.offset % 4 + 3) / 4;
glbSizeX = cols % blkSizeX == 0 ? cols : (cols / blkSizeX + 1) * blkSizeX;
}
else
{
cols = dst.cols;
glbSizeX = dst.cols % blkSizeX == 0 ? dst.cols : (dst.cols / blkSizeX + 1) * blkSizeX;
}
size_t glbSizeY = dst.rows % blkSizeY == 0 ? dst.rows : (dst.rows / blkSizeY + 1) * blkSizeY;
size_t globalThreads[3] = {glbSizeX, glbSizeY, 1};
size_t localThreads[3] = {blkSizeX, blkSizeY, 1};
std::vector< std::pair<size_t, const void *> > args;
args.push_back(std::make_pair(sizeof(cl_mem), (void *)&src.data));
args.push_back(std::make_pair(sizeof(cl_mem), (void *)&dst.data));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&src.cols));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&src.rows));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&dst.cols));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&dst.rows));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&srcStep));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&dstStep));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&src.offset));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&dst.offset));
args.push_back(std::make_pair(sizeof(cl_mem), (void *)&coeffs_cm));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&cols));
openCLExecuteKernel(clCxt, &imgproc_warpAffine, kernelName, globalThreads, localThreads, args, src.oclchannels(), src.depth());
openCLSafeCall(clReleaseMemObject(coeffs_cm));
}
void warpPerspective_gpu(const oclMat &src, oclMat &dst, double coeffs[3][3], int interpolation)
{
CV_Assert( (src.oclchannels() == dst.oclchannels()) );
int srcStep = src.step1();
int dstStep = dst.step1();
float float_coeffs[3][3];
cl_mem coeffs_cm;
Context *clCxt = src.clCxt;
String s[3] = {"NN", "Linear", "Cubic"};
String kernelName = "warpPerspective" + s[interpolation];
if (src.clCxt->supportsFeature(FEATURE_CL_DOUBLE))
{
cl_int st;
coeffs_cm = clCreateBuffer(*(cl_context*)clCxt->getOpenCLContextPtr(), CL_MEM_READ_WRITE, sizeof(double) * 3 * 3, NULL, &st );
openCLVerifyCall(st);
openCLSafeCall(clEnqueueWriteBuffer(*(cl_command_queue*)clCxt->getOpenCLCommandQueuePtr(), (cl_mem)coeffs_cm, 1, 0,
sizeof(double) * 3 * 3, coeffs, 0, 0, 0));
}
else
{
cl_int st;
for(int m = 0; m < 3; m++)
for(int n = 0; n < 3; n++)
float_coeffs[m][n] = coeffs[m][n];
coeffs_cm = clCreateBuffer(*(cl_context*)clCxt->getOpenCLContextPtr(), CL_MEM_READ_WRITE, sizeof(float) * 3 * 3, NULL, &st );
openCLVerifyCall(st);
openCLSafeCall(clEnqueueWriteBuffer(*(cl_command_queue*)clCxt->getOpenCLCommandQueuePtr(), (cl_mem)coeffs_cm, 1, 0,
sizeof(float) * 3 * 3, float_coeffs, 0, 0, 0));
}
//TODO: improve this kernel
#ifdef ANDROID
size_t blkSizeX = 16, blkSizeY = 8;
#else
size_t blkSizeX = 16, blkSizeY = 16;
#endif
size_t glbSizeX;
size_t cols;
if (src.type() == CV_8UC1 && interpolation == 0)
{
cols = (dst.cols + dst.offset % 4 + 3) / 4;
glbSizeX = cols % blkSizeX == 0 ? cols : (cols / blkSizeX + 1) * blkSizeX;
}
else
{
cols = dst.cols;
glbSizeX = dst.cols % blkSizeX == 0 ? dst.cols : (dst.cols / blkSizeX + 1) * blkSizeX;
}
size_t glbSizeY = dst.rows % blkSizeY == 0 ? dst.rows : (dst.rows / blkSizeY + 1) * blkSizeY;
size_t globalThreads[3] = {glbSizeX, glbSizeY, 1};
size_t localThreads[3] = {blkSizeX, blkSizeY, 1};
std::vector< std::pair<size_t, const void *> > args;
args.push_back(std::make_pair(sizeof(cl_mem), (void *)&src.data));
args.push_back(std::make_pair(sizeof(cl_mem), (void *)&dst.data));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&src.cols));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&src.rows));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&dst.cols));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&dst.rows));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&srcStep));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&dstStep));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&src.offset));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&dst.offset));
args.push_back(std::make_pair(sizeof(cl_mem), (void *)&coeffs_cm));
args.push_back(std::make_pair(sizeof(cl_int), (void *)&cols));
openCLExecuteKernel(clCxt, &imgproc_warpPerspective, kernelName, globalThreads, localThreads, args, src.oclchannels(), src.depth());
openCLSafeCall(clReleaseMemObject(coeffs_cm));
}
}
void warpAffine(const oclMat &src, oclMat &dst, const Mat &M, Size dsize, int flags)
{
int interpolation = flags & INTER_MAX;
CV_Assert((src.depth() == CV_8U || src.depth() == CV_32F) && src.oclchannels() != 2 && src.oclchannels() != 3);
CV_Assert(interpolation == INTER_NEAREST || interpolation == INTER_LINEAR || interpolation == INTER_CUBIC);
dst.create(dsize, src.type());
CV_Assert(M.rows == 2 && M.cols == 3);
int warpInd = (flags & WARP_INVERSE_MAP) >> 4;
F coeffs[2][3];
double coeffsM[2*3];
Mat coeffsMat(2, 3, CV_64F, (void *)coeffsM);
M.convertTo(coeffsMat, coeffsMat.type());
if (!warpInd)
convert_coeffs(coeffsM);
for(int i = 0; i < 2; ++i)
for(int j = 0; j < 3; ++j)
coeffs[i][j] = coeffsM[i*3+j];
warpAffine_gpu(src, dst, coeffs, interpolation);
}
void warpPerspective(const oclMat &src, oclMat &dst, const Mat &M, Size dsize, int flags)
{
int interpolation = flags & INTER_MAX;
CV_Assert((src.depth() == CV_8U || src.depth() == CV_32F) && src.oclchannels() != 2 && src.oclchannels() != 3);
CV_Assert(interpolation == INTER_NEAREST || interpolation == INTER_LINEAR || interpolation == INTER_CUBIC);
dst.create(dsize, src.type());
CV_Assert(M.rows == 3 && M.cols == 3);
int warpInd = (flags & WARP_INVERSE_MAP) >> 4;
double coeffs[3][3];
double coeffsM[3*3];
Mat coeffsMat(3, 3, CV_64F, (void *)coeffsM);
M.convertTo(coeffsMat, coeffsMat.type());
if (!warpInd)
invert(coeffsM);
for(int i = 0; i < 3; ++i)
for(int j = 0; j < 3; ++j)
coeffs[i][j] = coeffsM[i*3+j];
warpPerspective_gpu(src, dst, coeffs, interpolation);
}
////////////////////////////////////////////////////////////////////////
// integral
void integral(const oclMat &src, oclMat &sum, oclMat &sqsum, int sdepth)
{
CV_Assert(src.type() == CV_8UC1);
if (!src.clCxt->supportsFeature(ocl::FEATURE_CL_DOUBLE) && src.depth() == CV_64F)
{
CV_Error(Error::OpenCLDoubleNotSupported, "Select device doesn't support double");
return;
}
if( sdepth <= 0 )
sdepth = CV_32S;
sdepth = CV_MAT_DEPTH(sdepth);
int type = CV_MAKE_TYPE(sdepth, 1);
int vlen = 4;
int offset = src.offset / vlen;
int pre_invalid = src.offset % vlen;
int vcols = (pre_invalid + src.cols + vlen - 1) / vlen;
oclMat t_sum , t_sqsum;
int w = src.cols + 1, h = src.rows + 1;
char build_option[250];
if(Context::getContext()->supportsFeature(ocl::FEATURE_CL_DOUBLE))
{
t_sqsum.create(src.cols, src.rows, CV_64FC1);
sqsum.create(h, w, CV_64FC1);
sprintf(build_option, "-D TYPE=double -D TYPE4=double4 -D convert_TYPE4=convert_double4");
}
else
{
t_sqsum.create(src.cols, src.rows, CV_32FC1);
sqsum.create(h, w, CV_32FC1);
sprintf(build_option, "-D TYPE=float -D TYPE4=float4 -D convert_TYPE4=convert_float4");
}
t_sum.create(src.cols, src.rows, type);
sum.create(h, w, type);
int sum_offset = sum.offset / sum.elemSize();
int sqsum_offset = sqsum.offset / sqsum.elemSize();
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&src.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&t_sum.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&t_sqsum.data ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&pre_invalid ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.rows ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.cols ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.step ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sum.step));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sqsum.step));
size_t gt[3] = {((vcols + 1) / 2) * 256, 1, 1}, lt[3] = {256, 1, 1};
openCLExecuteKernel(src.clCxt, &imgproc_integral, "integral_cols", gt, lt, args, -1, sdepth, build_option);
args.clear();
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&t_sum.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&t_sqsum.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&sum.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&sqsum.data ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sum.rows ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sum.cols ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sum.step ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sqsum.step));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sum.step));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sqsum.step));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sum_offset));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sqsum_offset));
size_t gt2[3] = {t_sum.cols * 32, 1, 1}, lt2[3] = {256, 1, 1};
openCLExecuteKernel(src.clCxt, &imgproc_integral, "integral_rows", gt2, lt2, args, -1, sdepth, build_option);
}
void integral(const oclMat &src, oclMat &sum, int sdepth)
{
CV_Assert(src.type() == CV_8UC1);
int vlen = 4;
int offset = src.offset / vlen;
int pre_invalid = src.offset % vlen;
int vcols = (pre_invalid + src.cols + vlen - 1) / vlen;
if( sdepth <= 0 )
sdepth = CV_32S;
sdepth = CV_MAT_DEPTH(sdepth);
int type = CV_MAKE_TYPE(sdepth, 1);
oclMat t_sum;
int w = src.cols + 1, h = src.rows + 1;
t_sum.create(src.cols, src.rows, type);
sum.create(h, w, type);
int sum_offset = sum.offset / vlen;
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&src.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&t_sum.data ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&pre_invalid ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.rows ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.cols ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.step ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sum.step));
size_t gt[3] = {((vcols + 1) / 2) * 256, 1, 1}, lt[3] = {256, 1, 1};
openCLExecuteKernel(src.clCxt, &imgproc_integral_sum, "integral_sum_cols", gt, lt, args, -1, sdepth);
args.clear();
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&t_sum.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&sum.data ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sum.rows ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sum.cols ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&t_sum.step ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sum.step));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sum_offset));
size_t gt2[3] = {t_sum.cols * 32, 1, 1}, lt2[3] = {256, 1, 1};
openCLExecuteKernel(src.clCxt, &imgproc_integral_sum, "integral_sum_rows", gt2, lt2, args, -1, sdepth);
}
/////////////////////// corner //////////////////////////////
static void extractCovData(const oclMat &src, oclMat &Dx, oclMat &Dy,
int blockSize, int ksize, int borderType)
{
CV_Assert(src.type() == CV_8UC1 || src.type() == CV_32FC1);
double scale = static_cast<double>(1 << ((ksize > 0 ? ksize : 3) - 1)) * blockSize;
if (ksize < 0)
scale *= 2.;
if (src.depth() == CV_8U)
{
scale *= 255.;
scale = 1. / scale;
}
else
scale = 1. / scale;
if (ksize > 0)
{
Context* clCxt = Context::getContext();
if(clCxt->supportsFeature(FEATURE_CL_INTEL_DEVICE) && src.type() == CV_8UC1 &&
src.cols % 8 == 0 && src.rows % 8 == 0 &&
ksize==3 &&
(borderType ==cv::BORDER_REFLECT ||
borderType == cv::BORDER_REPLICATE ||
borderType ==cv::BORDER_REFLECT101 ||
borderType ==cv::BORDER_WRAP))
{
Dx.create(src.size(), CV_32FC1);
Dy.create(src.size(), CV_32FC1);
const unsigned int block_x = 8;
const unsigned int block_y = 8;
unsigned int src_pitch = src.step;
unsigned int dst_pitch = Dx.cols;
float _scale = scale;
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&src.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&Dx.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&Dy.data ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.cols ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.rows ));
args.push_back( std::make_pair( sizeof(cl_uint) , (void *)&src_pitch ));
args.push_back( std::make_pair( sizeof(cl_uint) , (void *)&dst_pitch ));
args.push_back( std::make_pair( sizeof(cl_float) , (void *)&_scale ));
size_t gt2[3] = {src.cols, src.rows, 1}, lt2[3] = {block_x, block_y, 1};
String option = "-D BLK_X=8 -D BLK_Y=8";
switch(borderType)
{
case cv::BORDER_REPLICATE:
option += " -D BORDER_REPLICATE";
break;
case cv::BORDER_REFLECT:
option += " -D BORDER_REFLECT";
break;
case cv::BORDER_REFLECT101:
option += " -D BORDER_REFLECT101";
break;
case cv::BORDER_WRAP:
option += " -D BORDER_WRAP";
break;
}
openCLExecuteKernel(src.clCxt, &imgproc_sobel3, "sobel3", gt2, lt2, args, -1, -1, option.c_str() );
}
else
{
Sobel(src, Dx, CV_32F, 1, 0, ksize, scale, 0, borderType);
Sobel(src, Dy, CV_32F, 0, 1, ksize, scale, 0, borderType);
}
}
else
{
Scharr(src, Dx, CV_32F, 1, 0, scale, 0, borderType);
Scharr(src, Dy, CV_32F, 0, 1, scale, 0, borderType);
}
CV_Assert(Dx.offset == 0 && Dy.offset == 0);
}
static void corner_ocl(const cv::ocl::ProgramEntry* source, String kernelName, int block_size, float k, oclMat &Dx, oclMat &Dy,
oclMat &dst, int border_type)
{
char borderType[30];
switch (border_type)
{
case cv::BORDER_CONSTANT:
sprintf(borderType, "BORDER_CONSTANT");
break;
case cv::BORDER_REFLECT101:
sprintf(borderType, "BORDER_REFLECT101");
break;
case cv::BORDER_REFLECT:
sprintf(borderType, "BORDER_REFLECT");
break;
case cv::BORDER_REPLICATE:
sprintf(borderType, "BORDER_REPLICATE");
break;
default:
CV_Error(Error::StsBadFlag, "BORDER type is not supported!");
}
std::string buildOptions = format("-D anX=%d -D anY=%d -D ksX=%d -D ksY=%d -D %s",
block_size / 2, block_size / 2, block_size, block_size, borderType);
size_t blockSizeX = 256, blockSizeY = 1;
size_t gSize = blockSizeX - block_size / 2 * 2;
size_t globalSizeX = (Dx.cols) % gSize == 0 ? Dx.cols / gSize * blockSizeX : (Dx.cols / gSize + 1) * blockSizeX;
size_t rows_per_thread = 2;
size_t globalSizeY = ((Dx.rows + rows_per_thread - 1) / rows_per_thread) % blockSizeY == 0 ?
((Dx.rows + rows_per_thread - 1) / rows_per_thread) :
(((Dx.rows + rows_per_thread - 1) / rows_per_thread) / blockSizeY + 1) * blockSizeY;
size_t gt[3] = { globalSizeX, globalSizeY, 1 };
size_t lt[3] = { blockSizeX, blockSizeY, 1 };
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&Dx.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&Dy.data));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&dst.data));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&Dx.offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&Dx.wholerows ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&Dx.wholecols ));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&Dx.step));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&Dy.offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&Dy.wholerows ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&Dy.wholecols ));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&Dy.step));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.offset));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.rows));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.cols));
args.push_back( std::make_pair(sizeof(cl_int), (void *)&dst.step));
args.push_back( std::make_pair( sizeof(cl_float) , (void *)&k));
openCLExecuteKernel(dst.clCxt, source, kernelName, gt, lt, args, -1, -1, buildOptions.c_str());
}
void cornerHarris(const oclMat &src, oclMat &dst, int blockSize, int ksize,
double k, int borderType)
{
oclMat dx, dy;
cornerHarris_dxdy(src, dst, dx, dy, blockSize, ksize, k, borderType);
}
void cornerHarris_dxdy(const oclMat &src, oclMat &dst, oclMat &dx, oclMat &dy, int blockSize, int ksize,
double k, int borderType)
{
if (!src.clCxt->supportsFeature(FEATURE_CL_DOUBLE) && src.depth() == CV_64F)
{
CV_Error(Error::OpenCLDoubleNotSupported, "Selected device doesn't support double");
return;
}
CV_Assert(borderType == cv::BORDER_CONSTANT || borderType == cv::BORDER_REFLECT101 || borderType == cv::BORDER_REPLICATE
|| borderType == cv::BORDER_REFLECT);
extractCovData(src, dx, dy, blockSize, ksize, borderType);
dst.create(src.size(), CV_32FC1);
corner_ocl(&imgproc_calcHarris, "calcHarris", blockSize, static_cast<float>(k), dx, dy, dst, borderType);
}
void cornerMinEigenVal(const oclMat &src, oclMat &dst, int blockSize, int ksize, int borderType)
{
oclMat dx, dy;
cornerMinEigenVal_dxdy(src, dst, dx, dy, blockSize, ksize, borderType);
}
void cornerMinEigenVal_dxdy(const oclMat &src, oclMat &dst, oclMat &dx, oclMat &dy, int blockSize, int ksize, int borderType)
{
if (!src.clCxt->supportsFeature(FEATURE_CL_DOUBLE) && src.depth() == CV_64F)
{
CV_Error(Error::OpenCLDoubleNotSupported, "Selected device doesn't support double");
return;
}
CV_Assert(borderType == cv::BORDER_CONSTANT || borderType == cv::BORDER_REFLECT101 ||
borderType == cv::BORDER_REPLICATE || borderType == cv::BORDER_REFLECT);
extractCovData(src, dx, dy, blockSize, ksize, borderType);
dst.create(src.size(), CV_32F);
corner_ocl(&imgproc_calcMinEigenVal, "calcMinEigenVal", blockSize, 0, dx, dy, dst, borderType);
}
/////////////////////////////////// MeanShiftfiltering ///////////////////////////////////////////////
static void meanShiftFiltering_gpu(const oclMat &src, oclMat dst, int sp, int sr, int maxIter, float eps)
{
CV_Assert( (src.cols == dst.cols) && (src.rows == dst.rows) );
CV_Assert( !(dst.step & 0x3) );
//Arrange the NDRange
int col = src.cols, row = src.rows;
int ltx = 16, lty = 8;
if (src.cols % ltx != 0)
col = (col / ltx + 1) * ltx;
if (src.rows % lty != 0)
row = (row / lty + 1) * lty;
size_t globalThreads[3] = {col, row, 1};
size_t localThreads[3] = {ltx, lty, 1};
//set args
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&dst.data ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dst.step ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&src.data ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.step ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dst.offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dst.cols ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dst.rows ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sp ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sr ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&maxIter ));
args.push_back( std::make_pair( sizeof(cl_float) , (void *)&eps ));
openCLExecuteKernel(src.clCxt, &meanShift, "meanshift_kernel", globalThreads, localThreads, args, -1, -1);
}
void meanShiftFiltering(const oclMat &src, oclMat &dst, int sp, int sr, TermCriteria criteria)
{
if (src.empty())
CV_Error(Error::StsBadArg, "The input image is empty");
if ( src.depth() != CV_8U || src.oclchannels() != 4 )
CV_Error(Error::StsUnsupportedFormat, "Only 8-bit, 4-channel images are supported");
dst.create( src.size(), CV_8UC4 );
if ( !(criteria.type & TermCriteria::MAX_ITER) )
criteria.maxCount = 5;
int maxIter = std::min(std::max(criteria.maxCount, 1), 100);
float eps;
if ( !(criteria.type & TermCriteria::EPS) )
eps = 1.f;
eps = (float)std::max(criteria.epsilon, 0.0);
meanShiftFiltering_gpu(src, dst, sp, sr, maxIter, eps);
}
static void meanShiftProc_gpu(const oclMat &src, oclMat dstr, oclMat dstsp, int sp, int sr, int maxIter, float eps)
{
//sanity checks
CV_Assert( (src.cols == dstr.cols) && (src.rows == dstr.rows) &&
(src.rows == dstsp.rows) && (src.cols == dstsp.cols));
CV_Assert( !(dstsp.step & 0x3) );
//Arrange the NDRange
int col = src.cols, row = src.rows;
int ltx = 16, lty = 8;
if (src.cols % ltx != 0)
col = (col / ltx + 1) * ltx;
if (src.rows % lty != 0)
row = (row / lty + 1) * lty;
size_t globalThreads[3] = {col, row, 1};
size_t localThreads[3] = {ltx, lty, 1};
//set args
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&src.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&dstr.data ));
args.push_back( std::make_pair( sizeof(cl_mem) , (void *)&dstsp.data ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.step ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dstr.step ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dstsp.step ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&src.offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dstr.offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dstsp.offset ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dstr.cols ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&dstr.rows ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sp ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&sr ));
args.push_back( std::make_pair( sizeof(cl_int) , (void *)&maxIter ));
args.push_back( std::make_pair( sizeof(cl_float) , (void *)&eps ));
openCLExecuteKernel(src.clCxt, &meanShift, "meanshiftproc_kernel", globalThreads, localThreads, args, -1, -1);
}
void meanShiftProc(const oclMat &src, oclMat &dstr, oclMat &dstsp, int sp, int sr, TermCriteria criteria)
{
if (src.empty())
CV_Error(Error::StsBadArg, "The input image is empty");
if ( src.depth() != CV_8U || src.oclchannels() != 4 )
CV_Error(Error::StsUnsupportedFormat, "Only 8-bit, 4-channel images are supported");
// if (!src.clCxt->supportsFeature(FEATURE_CL_DOUBLE))
// {
// CV_Error(Error::OpenCLDoubleNotSupportedNotSupported, "Selected device doesn't support double, so a deviation exists.\nIf the accuracy is acceptable, the error can be ignored.\n");
// return;
// }
dstr.create( src.size(), CV_8UC4 );
dstsp.create( src.size(), CV_16SC2 );
if ( !(criteria.type & TermCriteria::MAX_ITER) )
criteria.maxCount = 5;
int maxIter = std::min(std::max(criteria.maxCount, 1), 100);
float eps;
if ( !(criteria.type & TermCriteria::EPS) )
eps = 1.f;
eps = (float)std::max(criteria.epsilon, 0.0);
meanShiftProc_gpu(src, dstr, dstsp, sp, sr, maxIter, eps);
}
///////////////////////////////////////////////////////////////////////////////////////////////////
////////////////////////////////////////////////////hist///////////////////////////////////////////////
/////////////////////////////////////////////////////////////////////////////////////////////////////
namespace histograms
{
const int PARTIAL_HISTOGRAM256_COUNT = 256;
const int HISTOGRAM256_BIN_COUNT = 256;
}
///////////////////////////////calcHist/////////////////////////////////////////////////////////////////
static void calc_sub_hist(const oclMat &mat_src, const oclMat &mat_sub_hist)
{
using namespace histograms;
int depth = mat_src.depth();
size_t localThreads[3] = { HISTOGRAM256_BIN_COUNT, 1, 1 };
size_t globalThreads[3] = { PARTIAL_HISTOGRAM256_COUNT *localThreads[0], 1, 1};
int dataWidth = 16;
int dataWidth_bits = 4;
int mask = dataWidth - 1;
int cols = mat_src.cols * mat_src.oclchannels();
int src_offset = mat_src.offset;
int hist_step = mat_sub_hist.step >> 2;
int left_col = 0, right_col = 0;
if (cols >= dataWidth * 2 - 1)
{
left_col = dataWidth - (src_offset & mask);
left_col &= mask;
src_offset += left_col;
cols -= left_col;
right_col = cols & mask;
cols -= right_col;
}
else
{
left_col = cols;
right_col = 0;
cols = 0;
globalThreads[0] = 0;
}
std::vector<std::pair<size_t , const void *> > args;
if (globalThreads[0] != 0)
{
int tempcols = cols >> dataWidth_bits;
int inc_x = globalThreads[0] % tempcols;
int inc_y = globalThreads[0] / tempcols;
src_offset >>= dataWidth_bits;
int src_step = mat_src.step >> dataWidth_bits;
int datacount = tempcols * mat_src.rows;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&mat_src.data));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src_step));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src_offset));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&mat_sub_hist.data));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&datacount));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&tempcols));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&inc_x));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&inc_y));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&hist_step));
openCLExecuteKernel(mat_src.clCxt, &imgproc_histogram, "calc_sub_hist", globalThreads, localThreads, args, -1, depth);
}
if (left_col != 0 || right_col != 0)
{
src_offset = mat_src.offset;
localThreads[0] = 1;
localThreads[1] = 256;
globalThreads[0] = left_col + right_col;
globalThreads[1] = mat_src.rows;
args.clear();
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&mat_src.data));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&mat_src.step));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src_offset));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&mat_sub_hist.data));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&left_col));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&cols));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&mat_src.rows));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&hist_step));
openCLExecuteKernel(mat_src.clCxt, &imgproc_histogram, "calc_sub_hist_border", globalThreads, localThreads, args, -1, depth);
}
}
static void merge_sub_hist(const oclMat &sub_hist, oclMat &mat_hist)
{
using namespace histograms;
size_t localThreads[3] = { 256, 1, 1 };
size_t globalThreads[3] = { HISTOGRAM256_BIN_COUNT *localThreads[0], 1, 1};
int src_step = sub_hist.step >> 2;
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&sub_hist.data));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&mat_hist.data));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src_step));
openCLExecuteKernel(sub_hist.clCxt, &imgproc_histogram, "merge_hist", globalThreads, localThreads, args, -1, -1);
}
void calcHist(const oclMat &mat_src, oclMat &mat_hist)
{
using namespace histograms;
CV_Assert(mat_src.type() == CV_8UC1);
mat_hist.create(1, 256, CV_32SC1);
oclMat buf(PARTIAL_HISTOGRAM256_COUNT, HISTOGRAM256_BIN_COUNT, CV_32SC1);
buf.setTo(0);
calc_sub_hist(mat_src, buf);
merge_sub_hist(buf, mat_hist);
}
///////////////////////////////////equalizeHist/////////////////////////////////////////////////////
void equalizeHist(const oclMat &mat_src, oclMat &mat_dst)
{
mat_dst.create(mat_src.rows, mat_src.cols, CV_8UC1);
oclMat mat_hist(1, 256, CV_32SC1);
calcHist(mat_src, mat_hist);
size_t localThreads[3] = { 256, 1, 1};
size_t globalThreads[3] = { 256, 1, 1};
oclMat lut(1, 256, CV_8UC1);
int total = mat_src.rows * mat_src.cols;
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&lut.data));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&mat_hist.data));
args.push_back( std::make_pair( sizeof(int), (void *)&total));
openCLExecuteKernel(mat_src.clCxt, &imgproc_histogram, "calLUT", globalThreads, localThreads, args, -1, -1);
LUT(mat_src, lut, mat_dst);
}
////////////////////////////////////////////////////////////////////////
// CLAHE
namespace clahe
{
static void calcLut(const oclMat &src, oclMat &dst,
const int tilesX, const int tilesY, const cv::Size tileSize,
const int clipLimit, const float lutScale)
{
cl_int2 tile_size;
tile_size.s[0] = tileSize.width;
tile_size.s[1] = tileSize.height;
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&src.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&dst.data ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.step ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst.step ));
args.push_back( std::make_pair( sizeof(cl_int2), (void *)&tile_size ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&tilesX ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&clipLimit ));
args.push_back( std::make_pair( sizeof(cl_float), (void *)&lutScale ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.offset ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst.offset ));
String kernelName = "calcLut";
size_t localThreads[3] = { 32, 8, 1 };
size_t globalThreads[3] = { tilesX * localThreads[0], tilesY * localThreads[1], 1 };
bool is_cpu = isCpuDevice();
if (is_cpu)
openCLExecuteKernel(Context::getContext(), &imgproc_clahe, kernelName, globalThreads, localThreads, args, -1, -1, (char*)"-D CPU");
else
{
cl_kernel kernel = openCLGetKernelFromSource(Context::getContext(), &imgproc_clahe, kernelName);
int wave_size = (int)queryWaveFrontSize(kernel);
openCLSafeCall(clReleaseKernel(kernel));
std::string opt = format("-D WAVE_SIZE=%d", wave_size);
openCLExecuteKernel(Context::getContext(), &imgproc_clahe, kernelName, globalThreads, localThreads, args, -1, -1, opt.c_str());
}
}
static void transform(const oclMat &src, oclMat &dst, const oclMat &lut,
const int tilesX, const int tilesY, const Size & tileSize)
{
cl_int2 tile_size;
tile_size.s[0] = tileSize.width;
tile_size.s[1] = tileSize.height;
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&src.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&dst.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&lut.data ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.step ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst.step ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&lut.step ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.cols ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.rows ));
args.push_back( std::make_pair( sizeof(cl_int2), (void *)&tile_size ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&tilesX ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&tilesY ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.offset ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst.offset ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&lut.offset ));
size_t localThreads[3] = { 32, 8, 1 };
size_t globalThreads[3] = { src.cols, src.rows, 1 };
openCLExecuteKernel(Context::getContext(), &imgproc_clahe, "transform", globalThreads, localThreads, args, -1, -1);
}
}
namespace
{
class CLAHE_Impl : public cv::CLAHE
{
public:
CLAHE_Impl(double clipLimit = 40.0, int tilesX = 8, int tilesY = 8);
cv::AlgorithmInfo* info() const;
void apply(cv::InputArray src, cv::OutputArray dst);
void setClipLimit(double clipLimit);
double getClipLimit() const;
void setTilesGridSize(cv::Size tileGridSize);
cv::Size getTilesGridSize() const;
void collectGarbage();
private:
double clipLimit_;
int tilesX_;
int tilesY_;
oclMat srcExt_;
oclMat lut_;
};
CLAHE_Impl::CLAHE_Impl(double clipLimit, int tilesX, int tilesY) :
clipLimit_(clipLimit), tilesX_(tilesX), tilesY_(tilesY)
{
}
CV_INIT_ALGORITHM(CLAHE_Impl, "CLAHE_OCL",
obj.info()->addParam(obj, "clipLimit", obj.clipLimit_);
obj.info()->addParam(obj, "tilesX", obj.tilesX_);
obj.info()->addParam(obj, "tilesY", obj.tilesY_))
void CLAHE_Impl::apply(cv::InputArray src_raw, cv::OutputArray dst_raw)
{
oclMat& src = getOclMatRef(src_raw);
oclMat& dst = getOclMatRef(dst_raw);
CV_Assert( src.type() == CV_8UC1 );
dst.create( src.size(), src.type() );
const int histSize = 256;
ensureSizeIsEnough(tilesX_ * tilesY_, histSize, CV_8UC1, lut_);
cv::Size tileSize;
oclMat srcForLut;
if (src.cols % tilesX_ == 0 && src.rows % tilesY_ == 0)
{
tileSize = cv::Size(src.cols / tilesX_, src.rows / tilesY_);
srcForLut = src;
}
else
{
ocl::copyMakeBorder(src, srcExt_, 0, tilesY_ - (src.rows % tilesY_), 0,
tilesX_ - (src.cols % tilesX_), BORDER_REFLECT_101, Scalar::all(0));
tileSize = Size(srcExt_.cols / tilesX_, srcExt_.rows / tilesY_);
srcForLut = srcExt_;
}
const int tileSizeTotal = tileSize.area();
const float lutScale = static_cast<float>(histSize - 1) / tileSizeTotal;
int clipLimit = 0;
if (clipLimit_ > 0.0)
{
clipLimit = static_cast<int>(clipLimit_ * tileSizeTotal / histSize);
clipLimit = std::max(clipLimit, 1);
}
clahe::calcLut(srcForLut, lut_, tilesX_, tilesY_, tileSize, clipLimit, lutScale);
clahe::transform(src, dst, lut_, tilesX_, tilesY_, tileSize);
}
void CLAHE_Impl::setClipLimit(double clipLimit)
{
clipLimit_ = clipLimit;
}
double CLAHE_Impl::getClipLimit() const
{
return clipLimit_;
}
void CLAHE_Impl::setTilesGridSize(cv::Size tileGridSize)
{
tilesX_ = tileGridSize.width;
tilesY_ = tileGridSize.height;
}
cv::Size CLAHE_Impl::getTilesGridSize() const
{
return cv::Size(tilesX_, tilesY_);
}
void CLAHE_Impl::collectGarbage()
{
srcExt_.release();
lut_.release();
}
}
cv::Ptr<cv::CLAHE> createCLAHE(double clipLimit, cv::Size tileGridSize)
{
return makePtr<CLAHE_Impl>(clipLimit, tileGridSize.width, tileGridSize.height);
}
//////////////////////////////////bilateralFilter////////////////////////////////////////////////////
static void oclbilateralFilter_8u( const oclMat &src, oclMat &dst, int d,
double sigma_color, double sigma_space,
int borderType )
{
int cn = src.channels();
int i, j, maxk, radius;
CV_Assert( (src.channels() == 1 || src.channels() == 3) &&
src.type() == dst.type() && src.size() == dst.size() &&
src.data != dst.data );
if ( sigma_color <= 0 )
sigma_color = 1;
if ( sigma_space <= 0 )
sigma_space = 1;
double gauss_color_coeff = -0.5 / (sigma_color * sigma_color);
double gauss_space_coeff = -0.5 / (sigma_space * sigma_space);
if ( d <= 0 )
radius = cvRound(sigma_space * 1.5);
else
radius = d / 2;
radius = MAX(radius, 1);
d = radius * 2 + 1;
oclMat temp;
copyMakeBorder( src, temp, radius, radius, radius, radius, borderType );
std::vector<float> _color_weight(cn * 256);
std::vector<float> _space_weight(d * d);
std::vector<int> _space_ofs(d * d);
float *color_weight = &_color_weight[0];
float *space_weight = &_space_weight[0];
int *space_ofs = &_space_ofs[0];
int dst_step_in_pixel = dst.step / dst.elemSize();
int dst_offset_in_pixel = dst.offset / dst.elemSize();
int temp_step_in_pixel = temp.step / temp.elemSize();
// initialize color-related bilateral filter coefficients
for( i = 0; i < 256 * cn; i++ )
color_weight[i] = (float)std::exp(i * i * gauss_color_coeff);
// initialize space-related bilateral filter coefficients
for( i = -radius, maxk = 0; i <= radius; i++ )
for( j = -radius; j <= radius; j++ )
{
double r = std::sqrt((double)i * i + (double)j * j);
if ( r > radius )
continue;
space_weight[maxk] = (float)std::exp(r * r * gauss_space_coeff);
space_ofs[maxk++] = (int)(i * temp_step_in_pixel + j);
}
oclMat oclcolor_weight(1, cn * 256, CV_32FC1, color_weight);
oclMat oclspace_weight(1, d * d, CV_32FC1, space_weight);
oclMat oclspace_ofs(1, d * d, CV_32SC1, space_ofs);
String kernelName = "bilateral";
#ifdef ANDROID
size_t localThreads[3] = { 16, 8, 1 };
#else
size_t localThreads[3] = { 16, 16, 1 };
#endif
size_t globalThreads[3] = { dst.cols, dst.rows, 1 };
if ((dst.type() == CV_8UC1) && ((dst.offset & 3) == 0) && ((dst.cols & 3) == 0))
{
kernelName = "bilateral2";
globalThreads[0] = dst.cols >> 2;
}
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&dst.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&temp.data ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst.rows ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst.cols ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&maxk ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&radius ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst_step_in_pixel ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst_offset_in_pixel ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&temp_step_in_pixel ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&temp.rows ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&temp.cols ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&oclcolor_weight.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&oclspace_weight.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&oclspace_ofs.data ));
openCLExecuteKernel(src.clCxt, &imgproc_bilateral, kernelName, globalThreads, localThreads, args, dst.oclchannels(), dst.depth());
}
void bilateralFilter(const oclMat &src, oclMat &dst, int radius, double sigmaclr, double sigmaspc, int borderType)
{
dst.create( src.size(), src.type() );
if ( src.depth() == CV_8U )
oclbilateralFilter_8u( src, dst, radius, sigmaclr, sigmaspc, borderType );
else
CV_Error(Error::StsUnsupportedFormat, "Bilateral filtering is only implemented for CV_8U images");
}
}
}
//////////////////////////////////mulSpectrums////////////////////////////////////////////////////
void cv::ocl::mulSpectrums(const oclMat &a, const oclMat &b, oclMat &c, int /*flags*/, float scale, bool conjB)
{
CV_Assert(a.type() == CV_32FC2);
CV_Assert(b.type() == CV_32FC2);
c.create(a.size(), CV_32FC2);
size_t lt[3] = { 16, 16, 1 };
size_t gt[3] = { a.cols, a.rows, 1 };
String kernelName = conjB ? "mulAndScaleSpectrumsKernel_CONJ":"mulAndScaleSpectrumsKernel";
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&a.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&b.data ));
args.push_back( std::make_pair( sizeof(cl_float), (void *)&scale));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&c.data ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&a.cols ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&a.rows));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&a.step ));
Context *clCxt = Context::getContext();
openCLExecuteKernel(clCxt, &imgproc_mulAndScaleSpectrums, kernelName, gt, lt, args, -1, -1);
}
//////////////////////////////////convolve////////////////////////////////////////////////////
// ported from CUDA module
void cv::ocl::ConvolveBuf::create(Size image_size, Size templ_size)
{
result_size = Size(image_size.width - templ_size.width + 1,
image_size.height - templ_size.height + 1);
block_size = user_block_size;
if (user_block_size.width == 0 || user_block_size.height == 0)
block_size = estimateBlockSize(result_size, templ_size);
dft_size.width = 1 << int(ceil(std::log(block_size.width + templ_size.width - 1.) / std::log(2.)));
dft_size.height = 1 << int(ceil(std::log(block_size.height + templ_size.height - 1.) / std::log(2.)));
// CUFFT has hard-coded kernels for power-of-2 sizes (up to 8192),
// see CUDA Toolkit 4.1 CUFFT Library Programming Guide
//if (dft_size.width > 8192)
dft_size.width = getOptimalDFTSize(block_size.width + templ_size.width - 1.);
//if (dft_size.height > 8192)
dft_size.height = getOptimalDFTSize(block_size.height + templ_size.height - 1.);
// To avoid wasting time doing small DFTs
dft_size.width = std::max(dft_size.width, 512);
dft_size.height = std::max(dft_size.height, 512);
image_block.create(dft_size, CV_32F);
templ_block.create(dft_size, CV_32F);
result_data.create(dft_size, CV_32F);
//spect_len = dft_size.height * (dft_size.width / 2 + 1);
image_spect.create(dft_size.height, dft_size.width / 2 + 1, CV_32FC2);
templ_spect.create(dft_size.height, dft_size.width / 2 + 1, CV_32FC2);
result_spect.create(dft_size.height, dft_size.width / 2 + 1, CV_32FC2);
// Use maximum result matrix block size for the estimated DFT block size
block_size.width = std::min(dft_size.width - templ_size.width + 1, result_size.width);
block_size.height = std::min(dft_size.height - templ_size.height + 1, result_size.height);
}
Size cv::ocl::ConvolveBuf::estimateBlockSize(Size result_size, Size /*templ_size*/)
{
int width = (result_size.width + 2) / 3;
int height = (result_size.height + 2) / 3;
width = std::min(width, result_size.width);
height = std::min(height, result_size.height);
return Size(width, height);
}
static void convolve_run_fft(const oclMat &image, const oclMat &templ, oclMat &result, bool ccorr, ConvolveBuf& buf)
{
#if defined HAVE_CLAMDFFT
CV_Assert(image.type() == CV_32F);
CV_Assert(templ.type() == CV_32F);
buf.create(image.size(), templ.size());
result.create(buf.result_size, CV_32F);
Size& block_size = buf.block_size;
Size& dft_size = buf.dft_size;
oclMat& image_block = buf.image_block;
oclMat& templ_block = buf.templ_block;
oclMat& result_data = buf.result_data;
oclMat& image_spect = buf.image_spect;
oclMat& templ_spect = buf.templ_spect;
oclMat& result_spect = buf.result_spect;
oclMat templ_roi = templ;
copyMakeBorder(templ_roi, templ_block, 0, templ_block.rows - templ_roi.rows, 0,
templ_block.cols - templ_roi.cols, 0, Scalar());
cv::ocl::dft(templ_block, templ_spect, dft_size);
// Process all blocks of the result matrix
for (int y = 0; y < result.rows; y += block_size.height)
{
for (int x = 0; x < result.cols; x += block_size.width)
{
Size image_roi_size(std::min(x + dft_size.width, image.cols) - x,
std::min(y + dft_size.height, image.rows) - y);
Rect roi0(x, y, image_roi_size.width, image_roi_size.height);
oclMat image_roi(image, roi0);
copyMakeBorder(image_roi, image_block, 0, image_block.rows - image_roi.rows,
0, image_block.cols - image_roi.cols, 0, Scalar());
cv::ocl::dft(image_block, image_spect, dft_size);
mulSpectrums(image_spect, templ_spect, result_spect, 0,
1.f / dft_size.area(), ccorr);
cv::ocl::dft(result_spect, result_data, dft_size, cv::DFT_INVERSE | cv::DFT_REAL_OUTPUT);
Size result_roi_size(std::min(x + block_size.width, result.cols) - x,
std::min(y + block_size.height, result.rows) - y);
Rect roi1(x, y, result_roi_size.width, result_roi_size.height);
Rect roi2(0, 0, result_roi_size.width, result_roi_size.height);
oclMat result_roi(result, roi1);
oclMat result_block(result_data, roi2);
result_block.copyTo(result_roi);
}
}
#else
CV_Error(Error::OpenCLNoAMDBlasFft, "OpenCL DFT is not implemented");
#define UNUSED(x) (void)(x);
UNUSED(image) UNUSED(templ) UNUSED(result) UNUSED(ccorr) UNUSED(buf)
#undef UNUSED
#endif
}
static void convolve_run(const oclMat &src, const oclMat &temp1, oclMat &dst, String kernelName, const cv::ocl::ProgramEntry* source)
{
CV_Assert(src.depth() == CV_32FC1);
CV_Assert(temp1.depth() == CV_32F);
CV_Assert(temp1.cols <= 17 && temp1.rows <= 17);
dst.create(src.size(), src.type());
CV_Assert(src.cols == dst.cols && src.rows == dst.rows);
CV_Assert(src.type() == dst.type());
size_t localThreads[3] = { 16, 16, 1 };
size_t globalThreads[3] = { dst.cols, dst.rows, 1 };
int src_step = src.step / src.elemSize(), src_offset = src.offset / src.elemSize();
int dst_step = dst.step / dst.elemSize(), dst_offset = dst.offset / dst.elemSize();
int temp1_step = temp1.step / temp1.elemSize(), temp1_offset = temp1.offset / temp1.elemSize();
std::vector<std::pair<size_t , const void *> > args;
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&src.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&temp1.data ));
args.push_back( std::make_pair( sizeof(cl_mem), (void *)&dst.data ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.rows ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src.cols ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src_step ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst_step ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&temp1_step ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&temp1.rows ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&temp1.cols ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&src_offset ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&dst_offset ));
args.push_back( std::make_pair( sizeof(cl_int), (void *)&temp1_offset ));
openCLExecuteKernel(src.clCxt, source, kernelName, globalThreads, localThreads, args, -1, dst.depth());
}
void cv::ocl::convolve(const oclMat &x, const oclMat &t, oclMat &y, bool ccorr)
{
CV_Assert(x.depth() == CV_32F);
CV_Assert(t.depth() == CV_32F);
y.create(x.size(), x.type());
String kernelName = "convolve";
if(t.cols > 17 || t.rows > 17)
{
ConvolveBuf buf;
convolve_run_fft(x, t, y, ccorr, buf);
}
else
{
CV_Assert(ccorr == false);
convolve_run(x, t, y, kernelName, &imgproc_convolve);
}
}
void cv::ocl::convolve(const oclMat &image, const oclMat &templ, oclMat &result, bool ccorr, ConvolveBuf& buf)
{
result.create(image.size(), image.type());
convolve_run_fft(image, templ, result, ccorr, buf);
}
Reference in New Issue View Git Blame Copy Permalink