1 // This file is part of OpenCV project.
2 // It is subject to the license terms in the LICENSE file found in the top-level directory
3 // of this distribution and at http://opencv.org/license.html.
5 // Copyright (C) 2014, Itseez, Inc., all rights reserved.
6 // Third party copyrights are property of their respective owners.
11 #ifdef ONLY_SUM_CONVERT
13 __kernel void sumConvert(__global const uchar * src1ptr, int src1_step, int src1_offset,
14 __global const uchar * src2ptr, int src2_step, int src2_offset,
15 __global uchar * dstptr, int dst_step, int dst_offset, int dst_rows, int dst_cols,
16 coeffT scale, coeffT delta)
18 int x = get_global_id(0);
19 int y = get_global_id(1);
21 if (y < dst_rows && x < dst_cols)
23 int src1_index = mad24(y, src1_step, mad24(x, (int)sizeof(srcT), src1_offset));
24 int src2_index = mad24(y, src2_step, mad24(x, (int)sizeof(srcT), src2_offset));
25 int dst_index = mad24(y, dst_step, mad24(x, (int)sizeof(dstT), dst_offset));
27 __global const srcT * src1 = (__global const srcT *)(src1ptr + src1_index);
28 __global const srcT * src2 = (__global const srcT *)(src2ptr + src2_index);
29 __global dstT * dst = (__global dstT *)(dstptr + dst_index);
32 dst[0] = convertToDT( mad24((WT)(scale), convertToWT(src1[0]) + convertToWT(src2[0]), (WT)(delta)) );
34 dst[0] = convertToDT( mad((WT)(scale), convertToWT(src1[0]) + convertToWT(src2[0]), (WT)(delta)) );
41 ///////////////////////////////////////////////////////////////////////////////////////////////////
42 /////////////////////////////////Macro for border type////////////////////////////////////////////
43 /////////////////////////////////////////////////////////////////////////////////////////////////
45 #ifdef BORDER_CONSTANT
46 // CCCCCC|abcdefgh|CCCCCCC
47 #define EXTRAPOLATE(x, maxV)
48 #elif defined BORDER_REPLICATE
49 // aaaaaa|abcdefgh|hhhhhhh
50 #define EXTRAPOLATE(x, maxV) \
52 (x) = max(min((x), (maxV) - 1), 0); \
54 #elif defined BORDER_WRAP
55 // cdefgh|abcdefgh|abcdefg
56 #define EXTRAPOLATE(x, maxV) \
58 (x) = ( (x) + (maxV) ) % (maxV); \
60 #elif defined BORDER_REFLECT
61 // fedcba|abcdefgh|hgfedcb
62 #define EXTRAPOLATE(x, maxV) \
64 (x) = min(((maxV)-1)*2-(x)+1, max((x),-(x)-1) ); \
66 #elif defined BORDER_REFLECT_101
67 // gfedcb|abcdefgh|gfedcba
68 #define EXTRAPOLATE(x, maxV) \
70 (x) = min(((maxV)-1)*2-(x), max((x),-(x)) ); \
73 #error No extrapolation method
77 #define loadpix(addr) *(__global const srcT *)(addr)
78 #define storepix(val, addr) *(__global dstT *)(addr) = val
79 #define SRCSIZE (int)sizeof(srcT)
80 #define DSTSIZE (int)sizeof(dstT)
82 #define loadpix(addr) vload3(0, (__global const srcT1 *)(addr))
83 #define storepix(val, addr) vstore3(val, 0, (__global dstT1 *)(addr))
84 #define SRCSIZE (int)sizeof(srcT1)*3
85 #define DSTSIZE (int)sizeof(dstT1)*3
88 #define SRC(_x,_y) convertToWT(loadpix(Src + mad24(_y, src_step, SRCSIZE * _x)))
90 #ifdef BORDER_CONSTANT
91 // CCCCCC|abcdefgh|CCCCCCC
92 #define ELEM(_x,_y,r_edge,t_edge,const_v) (_x)<0 | (_x) >= (r_edge) | (_y)<0 | (_y) >= (t_edge) ? (const_v) : SRC((_x),(_y))
94 #define ELEM(_x,_y,r_edge,t_edge,const_v) SRC((_x),(_y))
97 // horizontal and vertical filter kernels
98 // should be defined on host during compile time to avoid overhead
100 __constant WT1 mat_kernelX[] = { KERNEL_MATRIX_X };
101 __constant WT1 mat_kernelY[] = { KERNEL_MATRIX_Y };
103 __kernel void laplacian(__global uchar* Src, int src_step, int srcOffsetX, int srcOffsetY, int height, int width,
104 __global uchar* Dst, int dst_step, int dst_offset, int dst_rows, int dst_cols,
105 WT1 scale, WT1 delta)
107 __local WT lsmem[BLK_Y + 2 * RADIUS][BLK_X + 2 * RADIUS];
108 __local WT lsmemDy1[BLK_Y][BLK_X + 2 * RADIUS];
109 __local WT lsmemDy2[BLK_Y][BLK_X + 2 * RADIUS];
111 int lix = get_local_id(0);
112 int liy = get_local_id(1);
114 int x = get_global_id(0);
116 int srcX = x + srcOffsetX - RADIUS;
121 int yb = clocY + srcOffsetY - RADIUS;
122 EXTRAPOLATE(yb, (height));
129 EXTRAPOLATE(xb,(width));
130 lsmem[clocY][clocX] = ELEM(xb, yb, (width), (height), 0 );
135 while(clocX < BLK_X+(RADIUS*2));
139 while (clocY < BLK_Y+(RADIUS*2));
140 barrier(CLK_LOCAL_MEM_FENCE);
142 WT scale_v = (WT)scale;
143 WT delta_v = (WT)delta;
144 for (int y = 0; y < dst_rows; y+=BLK_Y)
153 for (i=0; i<=2*RADIUS; i++)
155 sum1 = mad(lsmem[liy + i][clocX], mat_kernelY[i], sum1);
156 sum2 = mad(lsmem[liy + i][clocX], mat_kernelX[i], sum2);
158 lsmemDy1[liy][clocX] = sum1;
159 lsmemDy2[liy][clocX] = sum2;
162 while(clocX < BLK_X+(RADIUS*2));
163 barrier(CLK_LOCAL_MEM_FENCE);
165 if ((x < dst_cols) && (y + liy < dst_rows))
169 for (i=0; i<=2*RADIUS; i++)
171 sum1 = mad(lsmemDy1[liy][lix+i], mat_kernelX[i], sum1);
172 sum2 = mad(lsmemDy2[liy][lix+i], mat_kernelY[i], sum2);
175 WT sum = mad(scale_v, (sum1 + sum2), delta_v);
176 storepix(convertToDT(sum), Dst + mad24(y + liy, dst_step, mad24(x, DSTSIZE, dst_offset)));
179 for (int i = liy * BLK_X + lix; i < (RADIUS*2) * (BLK_X+(RADIUS*2)); i += BLK_X * BLK_Y)
181 int clocX = i % (BLK_X+(RADIUS*2));
182 int clocY = i / (BLK_X+(RADIUS*2));
183 lsmem[clocY][clocX] = lsmem[clocY + BLK_Y][clocX];
185 barrier(CLK_LOCAL_MEM_FENCE);
187 int yb = y + liy + BLK_Y + srcOffsetY + RADIUS;
188 EXTRAPOLATE(yb, (height));
191 int cSrcX = x + srcOffsetX - RADIUS;
195 EXTRAPOLATE(xb,(width));
196 lsmem[liy + 2*RADIUS][clocX] = ELEM(xb, yb, (width), (height), 0 );
201 while(clocX < BLK_X+(RADIUS*2));
202 barrier(CLK_LOCAL_MEM_FENCE);