1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
|
/*=========================================================================
*
* Copyright NumFOCUS
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* https://www.apache.org/licenses/LICENSE-2.0.txt
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
*=========================================================================*/
#ifndef itkGPUMeanImageFilter_hxx
#define itkGPUMeanImageFilter_hxx
namespace itk
{
template <typename TInputImage, typename TOutputImage>
GPUMeanImageFilter<TInputImage, TOutputImage>::GPUMeanImageFilter()
{
std::ostringstream defines;
if (TInputImage::ImageDimension > 3)
{
itkExceptionMacro("GPUMeanImageFilter supports 1/2/3D image.");
}
defines << "#define DIM_" << TInputImage::ImageDimension << '\n' << "#define PIXELTYPE ";
GetTypenameInString(typeid(typename TInputImage::PixelType), defines);
const char * GPUSource = GPUMeanImageFilter::GetOpenCLSource();
// load and build program
this->m_GPUKernelManager->LoadProgramFromString(GPUSource, defines.str().c_str());
// create kernel
m_MeanFilterGPUKernelHandle = this->m_GPUKernelManager->CreateKernel("MeanFilter");
}
template <typename TInputImage, typename TOutputImage>
GPUMeanImageFilter<TInputImage, TOutputImage>::~GPUMeanImageFilter() = default;
template <typename TInputImage, typename TOutputImage>
void
GPUMeanImageFilter<TInputImage, TOutputImage>::PrintSelf(std::ostream & os, Indent indent) const
{
Superclass::PrintSelf(os, indent);
}
template <typename TInputImage, typename TOutputImage>
void
GPUMeanImageFilter<TInputImage, TOutputImage>::GPUGenerateData()
{
using GPUInputImage = typename itk::GPUTraits<TInputImage>::Type;
using GPUOutputImage = typename itk::GPUTraits<TOutputImage>::Type;
typename GPUInputImage::Pointer inPtr = dynamic_cast<GPUInputImage *>(this->ProcessObject::GetInput(0));
typename GPUOutputImage::Pointer otPtr = dynamic_cast<GPUOutputImage *>(this->ProcessObject::GetOutput(0));
typename GPUOutputImage::SizeType outSize = otPtr->GetLargestPossibleRegion().GetSize();
int radius[3];
int imgSize[3];
radius[0] = radius[1] = radius[2] = 0;
imgSize[0] = imgSize[1] = imgSize[2] = 1;
int ImageDim = static_cast<int>(TInputImage::ImageDimension);
for (int i = 0; i < ImageDim; ++i)
{
radius[i] = (this->GetRadius())[i];
imgSize[i] = outSize[i];
}
size_t localSize[3], globalSize[3];
localSize[0] = localSize[1] = localSize[2] = OpenCLGetLocalBlockSize(ImageDim);
for (int i = 0; i < ImageDim; ++i)
{
// total # of threads
globalSize[i] =
localSize[i] * static_cast<unsigned int>(ceil(static_cast<float>(outSize[i]) / static_cast<float>(localSize[i])));
}
// arguments set up
int argidx = 0;
this->m_GPUKernelManager->SetKernelArgWithImage(m_MeanFilterGPUKernelHandle, argidx++, inPtr->GetGPUDataManager());
this->m_GPUKernelManager->SetKernelArgWithImage(m_MeanFilterGPUKernelHandle, argidx++, otPtr->GetGPUDataManager());
for (int i = 0; i < ImageDim; ++i)
{
this->m_GPUKernelManager->SetKernelArg(m_MeanFilterGPUKernelHandle, argidx++, sizeof(int), &(radius[i]));
}
for (int i = 0; i < ImageDim; ++i)
{
this->m_GPUKernelManager->SetKernelArg(m_MeanFilterGPUKernelHandle, argidx++, sizeof(int), &(imgSize[i]));
}
// launch kernel
this->m_GPUKernelManager->LaunchKernel(
m_MeanFilterGPUKernelHandle, static_cast<int>(TInputImage::ImageDimension), globalSize, localSize);
}
} // end namespace itk
#endif
|
