1/*M///////////////////////////////////////////////////////////////////////////////////////
2//
3//  IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
4//
5//  By downloading, copying, installing or using the software you agree to this license.
6//  If you do not agree to this license, do not download, install,
7//  copy or use the software.
8//
9//
10//                           License Agreement
11//                For Open Source Computer Vision Library
12//
13// Copyright (C) 2000-2008, Intel Corporation, all rights reserved.
14// Copyright (C) 2009, Willow Garage Inc., all rights reserved.
15// Third party copyrights are property of their respective owners.
16//
17// Redistribution and use in source and binary forms, with or without modification,
18// are permitted provided that the following conditions are met:
19//
20//   * Redistribution's of source code must retain the above copyright notice,
21//     this list of conditions and the following disclaimer.
22//
23//   * Redistribution's in binary form must reproduce the above copyright notice,
24//     this list of conditions and the following disclaimer in the documentation
25//     and/or other materials provided with the distribution.
26//
27//   * The name of the copyright holders may not be used to endorse or promote products
28//     derived from this software without specific prior written permission.
29//
30// This software is provided by the copyright holders and contributors "as is" and
31// any express or implied warranties, including, but not limited to, the implied
32// warranties of merchantability and fitness for a particular purpose are disclaimed.
33// In no event shall the Intel Corporation or contributors be liable for any direct,
34// indirect, incidental, special, exemplary, or consequential damages
35// (including, but not limited to, procurement of substitute goods or services;
36// loss of use, data, or profits; or business interruption) however caused
37// and on any theory of liability, whether in contract, strict liability,
38// or tort (including negligence or otherwise) arising in any way out of
39// the use of this software, even if advised of the possibility of such damage.
40//
41//M*/
42
43#if !defined CUDA_DISABLER
44
45#include "opencv2/core/cuda/common.hpp"
46
47namespace cv { namespace cuda { namespace device
48{
49    namespace blend
50    {
51        template <typename T>
52        __global__ void blendLinearKernel(int rows, int cols, int cn, const PtrStep<T> img1, const PtrStep<T> img2,
53                                          const PtrStepf weights1, const PtrStepf weights2, PtrStep<T> result)
54        {
55            int x = blockIdx.x * blockDim.x + threadIdx.x;
56            int y = blockIdx.y * blockDim.y + threadIdx.y;
57
58            if (y < rows && x < cols)
59            {
60                int x_ = x / cn;
61                float w1 = weights1.ptr(y)[x_];
62                float w2 = weights2.ptr(y)[x_];
63                T p1 = img1.ptr(y)[x];
64                T p2 = img2.ptr(y)[x];
65                result.ptr(y)[x] = (p1 * w1 + p2 * w2) / (w1 + w2 + 1e-5f);
66            }
67        }
68
69        template <typename T>
70        void blendLinearCaller(int rows, int cols, int cn, PtrStep<T> img1, PtrStep<T> img2, PtrStepf weights1, PtrStepf weights2, PtrStep<T> result, cudaStream_t stream)
71        {
72            dim3 threads(16, 16);
73            dim3 grid(divUp(cols * cn, threads.x), divUp(rows, threads.y));
74
75            blendLinearKernel<<<grid, threads, 0, stream>>>(rows, cols * cn, cn, img1, img2, weights1, weights2, result);
76            cudaSafeCall( cudaGetLastError() );
77
78            if (stream == 0)
79                cudaSafeCall(cudaDeviceSynchronize());
80        }
81
82        template void blendLinearCaller<uchar>(int, int, int, PtrStep<uchar>, PtrStep<uchar>, PtrStepf, PtrStepf, PtrStep<uchar>, cudaStream_t stream);
83        template void blendLinearCaller<float>(int, int, int, PtrStep<float>, PtrStep<float>, PtrStepf, PtrStepf, PtrStep<float>, cudaStream_t stream);
84
85
86        __global__ void blendLinearKernel8UC4(int rows, int cols, const PtrStepb img1, const PtrStepb img2,
87                                              const PtrStepf weights1, const PtrStepf weights2, PtrStepb result)
88        {
89            int x = blockIdx.x * blockDim.x + threadIdx.x;
90            int y = blockIdx.y * blockDim.y + threadIdx.y;
91
92            if (y < rows && x < cols)
93            {
94                float w1 = weights1.ptr(y)[x];
95                float w2 = weights2.ptr(y)[x];
96                float sum_inv = 1.f / (w1 + w2 + 1e-5f);
97                w1 *= sum_inv;
98                w2 *= sum_inv;
99                uchar4 p1 = ((const uchar4*)img1.ptr(y))[x];
100                uchar4 p2 = ((const uchar4*)img2.ptr(y))[x];
101                ((uchar4*)result.ptr(y))[x] = make_uchar4(p1.x * w1 + p2.x * w2, p1.y * w1 + p2.y * w2,
102                                                          p1.z * w1 + p2.z * w2, p1.w * w1 + p2.w * w2);
103            }
104        }
105
106        void blendLinearCaller8UC4(int rows, int cols, PtrStepb img1, PtrStepb img2, PtrStepf weights1, PtrStepf weights2, PtrStepb result, cudaStream_t stream)
107        {
108            dim3 threads(16, 16);
109            dim3 grid(divUp(cols, threads.x), divUp(rows, threads.y));
110
111            blendLinearKernel8UC4<<<grid, threads, 0, stream>>>(rows, cols, img1, img2, weights1, weights2, result);
112            cudaSafeCall( cudaGetLastError() );
113
114            if (stream == 0)
115                cudaSafeCall(cudaDeviceSynchronize());
116        }
117    } // namespace blend
118}}} // namespace cv { namespace cuda { namespace cudev
119
120
121#endif /* CUDA_DISABLER */
122