generic-library/opencv
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Use hypot to do vector calculation.

Browse Source 
Signed-off-by: Yan Wang <yan.wang@linux.intel.com>
This commit is contained in:
Yan Wang 2014-12-22 18:41:50 +08:00
parent 700a388173
commit 9699fde8d2
1 changed files with 16 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
modules/objdetect/src/opencl/objdetect_hog.cl
View File

@ -48,7 +48,7 @@
#define CELLS_PER_BLOCK_X 2 #define CELLS_PER_BLOCK_X 2
#define CELLS_PER_BLOCK_Y 2 #define CELLS_PER_BLOCK_Y 2
#define NTHREADS 256 #define NTHREADS 256
#define CV_PI_F 3.1415926535897932384626433832795f #define CV_PI_F M_PI_F
#ifdef INTEL_DEVICE #ifdef INTEL_DEVICE
#define QANGLE_TYPE int #define QANGLE_TYPE int
@ -606,23 +606,23 @@ __kernel void compute_gradients_8UC4_kernel(
barrier(CLK_LOCAL_MEM_FENCE); barrier(CLK_LOCAL_MEM_FENCE);
if (x < width) if (x < width)
{ {
float3 a = (float3) (sh_row[tid], sh_row[tid + (NTHREADS + 2)], float4 a = (float4) (sh_row[tid], sh_row[tid + (NTHREADS + 2)],
sh_row[tid + 2 * (NTHREADS + 2)]); sh_row[tid + 2 * (NTHREADS + 2)], 0);
float3 b = (float3) (sh_row[tid + 2], sh_row[tid + 2 + (NTHREADS + 2)], float4 b = (float4) (sh_row[tid + 2], sh_row[tid + 2 + (NTHREADS + 2)],
sh_row[tid + 2 + 2 * (NTHREADS + 2)]); sh_row[tid + 2 + 2 * (NTHREADS + 2)], 0);
float3 dx; float4 dx;
if (correct_gamma == 1) if (correct_gamma == 1)
dx = sqrt(b) - sqrt(a); dx = sqrt(b) - sqrt(a);
else else
dx = b - a; dx = b - a;
float3 dy = (float3) 0.f; float4 dy = (float4) 0.f;
if (gidY > 0 && gidY < height - 1) if (gidY > 0 && gidY < height - 1)
{ {
a = convert_float3(img[(gidY - 1) * img_step + x].xyz); a = convert_float4(img[(gidY - 1) * img_step + x].xyzw);
b = convert_float3(img[(gidY + 1) * img_step + x].xyz); b = convert_float4(img[(gidY + 1) * img_step + x].xyzw);
if (correct_gamma == 1) if (correct_gamma == 1)
dy = sqrt(b) - sqrt(a); dy = sqrt(b) - sqrt(a);
@ -630,28 +630,25 @@ __kernel void compute_gradients_8UC4_kernel(
dy = b - a; dy = b - a;
} }
float4 mag = hypot(dx, dy);
float best_dx = dx.x; float best_dx = dx.x;
float best_dy = dy.x; float best_dy = dy.x;
float mag0 = dx.x * dx.x + dy.x * dy.x; float mag0 = mag.x;
float mag1 = dx.y * dx.y + dy.y * dy.y; if (mag0 < mag.y)
if (mag0 < mag1)
{ {
best_dx = dx.y; best_dx = dx.y;
best_dy = dy.y; best_dy = dy.y;
mag0 = mag1; mag0 = mag.y;
} }
mag1 = dx.z * dx.z + dy.z * dy.z; if (mag0 < mag.z)
if (mag0 < mag1)
{ {
best_dx = dx.z; best_dx = dx.z;
best_dy = dy.z; best_dy = dy.z;
mag0 = mag1; mag0 = mag.z;
} }
mag0 = sqrt(mag0);
float ang = (atan2(best_dy, best_dx) + CV_PI_F) * angle_scale - 0.5f; float ang = (atan2(best_dy, best_dx) + CV_PI_F) * angle_scale - 0.5f;
int hidx = (int)floor(ang); int hidx = (int)floor(ang);
ang -= hidx; ang -= hidx;
@ -710,7 +707,7 @@ __kernel void compute_gradients_8UC1_kernel(
else else
dy = a - b; dy = a - b;
} }
float mag = sqrt(dx * dx + dy * dy); float mag = hypot(dx, dy);
float ang = (atan2(dy, dx) + CV_PI_F) * angle_scale - 0.5f; float ang = (atan2(dy, dx) + CV_PI_F) * angle_scale - 0.5f;
int hidx = (int)floor(ang); int hidx = (int)floor(ang);