-
-
Notifications
You must be signed in to change notification settings - Fork 667
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
BUG: Add tests demonstration current behavior of histogram based median
- Loading branch information
Showing
3 changed files
with
255 additions
and
2 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,4 +1,5 @@ | ||
test/*\.cxx Namespace Disable | ||
test/itkAttributePositionLabelMapFilterTest1.cxx Namespace Disable | ||
test/itkShapeLabelMapFilterGTest.cxx Namespace Disable | ||
test/itkStatisticsLabelMapFilterGTest.cxx Namespace Disable | ||
test/itkShapeLabelObjectAccessorsTest1.cxx SemicolonSpace Disable |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
251 changes: 251 additions & 0 deletions
251
Modules/Filtering/LabelMap/test/itkStatisticsLabelMapFilterGTest.cxx
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,251 @@ | ||
/*========================================================================= | ||
* | ||
* 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 | ||
* | ||
* http://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. | ||
* | ||
*=========================================================================*/ | ||
|
||
#include "itkGTest.h" | ||
|
||
#include "itkImage.h" | ||
#include "itkLabelImageToStatisticsLabelMapFilter.h" | ||
#include "itkImageRegionIterator.h" | ||
#include <algorithm> | ||
|
||
|
||
namespace | ||
{ | ||
|
||
class StatisticsLabelMapFixture : public ::testing::Test | ||
{ | ||
public: | ||
StatisticsLabelMapFixture() = default; | ||
~StatisticsLabelMapFixture() override = default; | ||
|
||
protected: | ||
void | ||
SetUp() override | ||
{} | ||
void | ||
TearDown() override | ||
{} | ||
|
||
template <unsigned int D, typename TPixelType = unsigned short> | ||
struct FixtureUtilities | ||
{ | ||
static const unsigned int Dimension = D; | ||
|
||
using PixelType = TPixelType; | ||
using ImageType = itk::Image<PixelType, Dimension>; | ||
|
||
using LabelPixelType = unsigned char; | ||
using LabelImageType = itk::Image<LabelPixelType, Dimension>; | ||
|
||
using LabelObjectType = itk::StatisticsLabelObject<LabelPixelType, Dimension>; | ||
using StatisticsLabelMapType = itk::LabelMap<LabelObjectType>; | ||
|
||
|
||
static typename ImageType::Pointer | ||
CreateImage() | ||
{ | ||
typename ImageType::Pointer image = ImageType::New(); | ||
|
||
typename ImageType::SizeType imageSize; | ||
imageSize.Fill(25); | ||
image->SetRegions(typename ImageType::RegionType(imageSize)); | ||
image->Allocate(); | ||
image->FillBuffer(0); | ||
|
||
return image; | ||
} | ||
|
||
static typename ImageType::Pointer | ||
CreateImageRandom(PixelType randMax = 500, unsigned int randSeed = 0) | ||
{ | ||
typename ImageType::Pointer image = ImageType::New(); | ||
|
||
typename ImageType::SizeType imageSize; | ||
imageSize.Fill(25); | ||
image->SetRegions(typename ImageType::RegionType(imageSize)); | ||
image->Allocate(); | ||
image->FillBuffer(0); | ||
|
||
srand(randSeed); | ||
|
||
itk::ImageRegionIterator<ImageType> it(image, image->GetLargestPossibleRegion()); | ||
while (!it.IsAtEnd()) | ||
{ | ||
it.Set(rand() % randMax); | ||
++it; | ||
} | ||
return image; | ||
} | ||
|
||
|
||
static typename LabelImageType::Pointer | ||
CreateLabelImage() | ||
{ | ||
auto image = LabelImageType::New(); | ||
|
||
typename LabelImageType::SizeType imageSize; | ||
imageSize.Fill(25); | ||
image->SetRegions(typename ImageType::RegionType(imageSize)); | ||
image->Allocate(); | ||
image->FillBuffer(0); | ||
|
||
return image; | ||
} | ||
|
||
|
||
static typename LabelObjectType::ConstPointer | ||
ComputeLabelObject(const LabelImageType * labelImage, | ||
const ImageType * image, | ||
const PixelType label = 1, | ||
const size_t numberOfBins = 0) | ||
{ | ||
|
||
auto l2s = itk::LabelImageToStatisticsLabelMapFilter<LabelImageType, ImageType>::New(); | ||
l2s->SetInput1(labelImage); | ||
l2s->SetFeatureImage(image); | ||
l2s->ComputeFeretDiameterOn(); | ||
l2s->ComputePerimeterOn(); | ||
// l2s->ComputeOrientedBoundingBoxOn(); | ||
l2s->ComputeHistogramOn(); | ||
if (numberOfBins != 0) | ||
{ | ||
l2s->SetNumberOfBins(numberOfBins); | ||
} | ||
l2s->Update(); | ||
return l2s->GetOutput()->GetLabelObject(label); | ||
} | ||
|
||
static double | ||
ComputeExactMedian(const LabelObjectType * labelObject, const ImageType * image) | ||
{ | ||
std::vector<PixelType> values; | ||
typename LabelObjectType::ConstIndexIterator it(labelObject); | ||
|
||
while (!it.IsAtEnd()) | ||
{ | ||
const typename ImageType::IndexType & idx = it.GetIndex(); | ||
values.push_back(image->GetPixel(idx)); | ||
++it; | ||
} | ||
|
||
std::sort(values.begin(), values.end()); | ||
|
||
auto n1 = values.size() / 2; | ||
if (values.size() % 2 == 0) | ||
{ | ||
return 0.5 * (double(values[n1]) + double(values[n1 - 1])); | ||
} | ||
return values[n1]; | ||
} | ||
}; | ||
}; | ||
} // namespace | ||
|
||
|
||
TEST_F(StatisticsLabelMapFixture, 2D_zero) | ||
{ | ||
using Utils = FixtureUtilities<2, unsigned char>; | ||
|
||
auto image = Utils::CreateImage(); | ||
auto labelImage = Utils ::CreateLabelImage(); | ||
|
||
Utils::LabelPixelType label = 1; | ||
labelImage->FillBuffer(label); | ||
|
||
|
||
Utils::LabelObjectType::ConstPointer labelObject = Utils::ComputeLabelObject(labelImage, image, 1, 1 << 8); | ||
|
||
EXPECT_NEAR(0.0, labelObject->GetMinimum(), 1e-12); | ||
EXPECT_NEAR(0.0, labelObject->GetMaximum(), 1e-12); | ||
EXPECT_NEAR(Utils::ComputeExactMedian(labelObject, image), labelObject->GetMedian(), 1e-12); | ||
EXPECT_NEAR(0.0, labelObject->GetSum(), 1e-12); | ||
EXPECT_NEAR(0.0, labelObject->GetVariance(), 1e-12); | ||
EXPECT_NEAR(0.0, labelObject->GetStandardDeviation(), 1e-12); | ||
|
||
if (::testing::Test::HasFailure()) | ||
{ | ||
labelObject->Print(std::cout); | ||
} | ||
} | ||
|
||
|
||
TEST_F(StatisticsLabelMapFixture, 2D_ones_with_outliers) | ||
{ | ||
using Utils = FixtureUtilities<2, signed short>; | ||
|
||
auto image = Utils::CreateImage(); | ||
Utils::PixelType value = 1; | ||
image->FillBuffer(value); | ||
|
||
// Test with outliers outside the label. | ||
image->SetPixel({ 0, 0 }, 32000); | ||
image->SetPixel({ 0, 1 }, -32000); | ||
|
||
|
||
auto labelImage = Utils ::CreateLabelImage(); | ||
Utils::LabelPixelType label = 1; | ||
labelImage->FillBuffer(label); | ||
labelImage->SetPixel({ 0, 0 }, 0); | ||
labelImage->SetPixel({ 0, 1 }, 0); | ||
|
||
Utils::LabelObjectType::ConstPointer labelObject = Utils::ComputeLabelObject(labelImage, image, label, 1 << 16); | ||
|
||
EXPECT_NEAR(value, labelObject->GetMinimum(), 1e-12); | ||
EXPECT_NEAR(value, labelObject->GetMaximum(), 1e-12); | ||
EXPECT_NEAR(Utils::ComputeExactMedian(labelObject, image), labelObject->GetMedian(), 0.5); | ||
EXPECT_NEAR(25 * 25 - 2, labelObject->GetSum(), 1e-12); | ||
EXPECT_NEAR(0.0, labelObject->GetVariance(), 1e-12); | ||
EXPECT_NEAR(0.0, labelObject->GetStandardDeviation(), 1e-12); | ||
|
||
if (::testing::Test::HasFailure()) | ||
{ | ||
labelObject->Print(std::cout); | ||
} | ||
} | ||
|
||
TEST_F(StatisticsLabelMapFixture, 2D_rand_with_outliers) | ||
{ | ||
using Utils = FixtureUtilities<2, signed short>; | ||
|
||
auto image = Utils::CreateImageRandom(500, 0); | ||
auto labelImage = Utils ::CreateLabelImage(); | ||
|
||
// Test with outliers outside the label. | ||
image->SetPixel({ 0, 0 }, 32000); | ||
image->SetPixel({ 0, 1 }, -2000); | ||
// Set min/max in label | ||
image->SetPixel({ 0, 2 }, 0); | ||
image->SetPixel({ 0, 3 }, 500); | ||
|
||
Utils::LabelPixelType label = 1; | ||
labelImage->FillBuffer(label); | ||
labelImage->SetPixel({ 0, 0 }, 0); | ||
labelImage->SetPixel({ 0, 1 }, 0); | ||
|
||
|
||
Utils::LabelObjectType::ConstPointer labelObject = Utils::ComputeLabelObject(labelImage, image, label, 1 << 16); | ||
|
||
EXPECT_NEAR(0.0, labelObject->GetMinimum(), 1e-12); | ||
EXPECT_NEAR(500.0, labelObject->GetMaximum(), 1e-12); | ||
EXPECT_NEAR(Utils::ComputeExactMedian(labelObject, image), labelObject->GetMedian(), 0.5); | ||
|
||
if (::testing::Test::HasFailure()) | ||
{ | ||
labelObject->Print(std::cout); | ||
} | ||
} |