=============================================
-# Launch a web browser of choice and go to our [page on
- Sourceforge](http://sourceforge.net/projects/opencvlibrary/files/opencv-win/).
+ Sourceforge](http://sourceforge.net/projects/opencvlibrary/files/).
-# Choose a build you want to use and download it.
-# Make sure you have admin rights. Unpack the self-extracting archive.
-# You can check the installation at the chosen path as you can see below.
{
CV_INSTRUMENT_REGION();
- if (method >= 32 && method <= 38)
+ if (method >= USAC_DEFAULT && method <= USAC_MAGSAC)
return usac::findEssentialMat(_points1, _points2, _cameraMatrix,
- method, prob, threshold, _mask);
+ method, prob, threshold, _mask, maxIters);
Mat points1, points2, cameraMatrix;
_points1.getMat().convertTo(points1, CV_64F);
Mat findEssentialMat( InputArray points1, InputArray points2,
InputArray cameraMatrix1,
int method, double prob,
- double threshold, OutputArray mask);
+ double threshold, OutputArray mask,
+ int maxIters);
Mat estimateAffine2D(InputArray from, InputArray to, OutputArray inliers,
int method, double ransacReprojThreshold, int maxIters,
}
Mat findEssentialMat (InputArray points1, InputArray points2, InputArray cameraMatrix1,
- int method, double prob, double thr, OutputArray mask) {
+ int method, double prob, double thr, OutputArray mask, int maxIters) {
Ptr<Model> params;
- setParameters(method, params, EstimationMethod::Essential, thr, 1000, prob, mask.needed());
+ setParameters(method, params, EstimationMethod::Essential, thr, maxIters, prob, mask.needed());
Ptr<RansacOutput> ransac_output;
if (run(params, points1, points2, params->getRandomGeneratorState(),
ransac_output, cameraMatrix1, cameraMatrix1, noArray(), noArray())) {
EXPECT_TRUE(m.empty());
}
+CV_ENUM(UsacMethod, USAC_DEFAULT, USAC_ACCURATE, USAC_PROSAC, USAC_FAST, USAC_MAGSAC)
+typedef TestWithParam<UsacMethod> usac_Essential;
-TEST(usac_Essential, accuracy) {
+TEST_P(usac_Essential, accuracy) {
+ int method = GetParam();
std::vector<int> gt_inliers;
const int pts_size = 1500;
cv::RNG &rng = cv::theRNG();
int inl_size = generatePoints(rng, pts1, pts2, K1, K2, false /*two calib*/,
pts_size, TestSolver ::Fundam, inl_ratio, 0.01 /*noise std, works bad with high noise*/, gt_inliers);
const double conf = 0.99, thr = 1.;
- for (auto flag : flags) {
- cv::Mat mask, E;
- try {
- E = cv::findEssentialMat(pts1, pts2, K1, flag, conf, thr, mask);
- } catch (cv::Exception &e) {
- if (e.code != cv::Error::StsNotImplemented)
- FAIL() << "Essential matrix estimation failed!\n";
- else continue;
- }
- // calibrate points
- cv::Mat cpts1_3d, cpts2_3d;
- cv::vconcat(pts1, cv::Mat::ones(1, pts1.cols, pts1.type()), cpts1_3d);
- cv::vconcat(pts2, cv::Mat::ones(1, pts2.cols, pts2.type()), cpts2_3d);
- cpts1_3d = K1.inv() * cpts1_3d; cpts2_3d = K1.inv() * cpts2_3d;
- checkInliersMask(TestSolver::Essen, inl_size, thr / ((K1.at<double>(0,0) + K1.at<double>(1,1)) / 2),
- cpts1_3d.rowRange(0,2), cpts2_3d.rowRange(0,2), E, mask);
+ cv::Mat mask, E;
+ try {
+ E = cv::findEssentialMat(pts1, pts2, K1, method, conf, thr, mask);
+ } catch (cv::Exception &e) {
+ if (e.code != cv::Error::StsNotImplemented)
+ FAIL() << "Essential matrix estimation failed!\n";
+ else continue;
+ }
+ // calibrate points
+ cv::Mat cpts1_3d, cpts2_3d;
+ cv::vconcat(pts1, cv::Mat::ones(1, pts1.cols, pts1.type()), cpts1_3d);
+ cv::vconcat(pts2, cv::Mat::ones(1, pts2.cols, pts2.type()), cpts2_3d);
+ cpts1_3d = K1.inv() * cpts1_3d; cpts2_3d = K1.inv() * cpts2_3d;
+ checkInliersMask(TestSolver::Essen, inl_size, thr / ((K1.at<double>(0,0) + K1.at<double>(1,1)) / 2),
+ cpts1_3d.rowRange(0,2), cpts2_3d.rowRange(0,2), E, mask);
+ }
+}
+
+TEST_P(usac_Essential, maxiters) {
+ int method = GetParam();
+ cv::RNG &rng = cv::theRNG();
+ cv::Mat mask;
+ cv::Mat K1 = cv::Mat(cv::Matx33d(1, 0, 0,
+ 0, 1, 0,
+ 0, 0, 1.));
+ const double conf = 0.99, thr = 0.5;
+ int roll_results_sum = 0;
+
+ for (int iters = 0; iters < 10; iters++) {
+ cv::Mat E1, E2;
+ try {
+ cv::Mat pts1 = cv::Mat(2, 50, CV_64F);
+ cv::Mat pts2 = cv::Mat(2, 50, CV_64F);
+ rng.fill(pts1, cv::RNG::UNIFORM, 0.0, 1.0);
+ rng.fill(pts2, cv::RNG::UNIFORM, 0.0, 1.0);
+
+ E1 = cv::findEssentialMat(pts1, pts2, K1, method, conf, thr, 1, mask);
+ E2 = cv::findEssentialMat(pts1, pts2, K1, method, conf, thr, 1000, mask);
+
+ if (E1.dims != E2.dims) { continue; }
+ roll_results_sum += cv::norm(E1, E2, NORM_L1) != 0;
+ } catch (cv::Exception &e) {
+ if (e.code != cv::Error::StsNotImplemented)
+ FAIL() << "Essential matrix estimation failed!\n";
+ else continue;
}
+ EXPECT_NE(roll_results_sum, 0);
}
}
+INSTANTIATE_TEST_CASE_P(Calib3d, usac_Essential, UsacMethod::all());
+
TEST(usac_P3P, accuracy) {
std::vector<int> gt_inliers;
const int pts_size = 3000;
CV_TRACE_FUNCTION();
CV_TRACE_ARG_VALUE(name, "name", name.c_str());
+ // FP16 fallback is not needed as we handle FP16 below
+
std::vector<Mat> inputs, outputs;
inputs_arr.getMatVector(inputs);
outputs_arr.getMatVector(outputs);
+ CV_CheckEQ(inputs.size(), (size_t)2, "");
+ CV_CheckEQ(outputs.size(), (size_t)1, "");
+
const Mat& inp = inputs[0];
- const Mat& indices = inputs[1];
+
+ int indicesType = inputs[1].type();
+ CV_CheckType(indicesType, indicesType == CV_32FC1 || indicesType == CV_16SC1, "");
+ Mat indices32S;
+ if (indicesType == CV_16S/*FP16*/)
+ {
+ Mat indicesF32;
+ convertFp16(inputs[1], indicesF32);
+ indicesF32.convertTo(indices32S, CV_32S);
+ }
+ else
+ {
+ inputs[1].convertTo(indices32S, CV_32S);
+ }
+ const size_t indices_total = indices32S.total();
+ indices32S = indices32S.reshape(1, indices_total);
+
Mat& out = outputs[0];
+ CV_CheckTypeEQ(inp.type(), out.type(), "");
+ CV_CheckTypeEQ(indices32S.type(), CV_32SC1, "");
+
const int axis = normalize_axis(m_axis, shape(inp));
+ // FIXIT: why should we work with non-normalized input? it should be handled in importer or layers's output generator
+ const int axis_size = (int)inp.size[axis];
+ for (size_t j = 0 ; j < indices_total; ++j)
+ {
+ int& idx = indices32S.at<int>(j);
+ idx = normalize_axis(idx, axis_size); // validate and normalize indices
+ }
+
const size_t outer_size = axis == 0 ? inp.total() : inp.step1(axis - 1);
const size_t outer_dims = inp.total() / outer_size;
const size_t inner_size = inp.step1(axis);
- const float* idx = indices.ptr<const float>(); // TODO: change type to integer in the future.
+ const int* idx = indices32S.ptr<int>();
const char* src = inp.ptr<const char>();
char* dst = out.ptr<char>();
+ CV_CheckEQ(out.total(), outer_dims * indices_total * inner_size, "");
const size_t es = inp.elemSize1();
+ // TODO: optimize through switch (inner_size * es)
+ const size_t inner_bytes = inner_size * es;
for (size_t i = 0; i < outer_dims; ++i)
{
const size_t src_offset = i * outer_size;
- for (size_t j = 0 ; j < indices.total(); ++j)
+ for (size_t j = 0 ; j < indices_total; ++j)
{
- const size_t index = (static_cast<int>(idx[j]) + inp.size[axis]) % inp.size[axis];
- const size_t new_offset = src_offset + index * inp.step1(axis);
- std::memcpy(dst, src + new_offset * es, inner_size * es);
- dst += inner_size * es;
+ const int index = idx[j];
+ CV_DbgCheck(index, index >= 0 && index < axis_size, "");
+ const size_t new_offset = src_offset + index * inner_size;
+ std::memcpy(dst, src + new_offset * es, inner_bytes);
+ dst += inner_bytes;
}
}
}
TEST_P(Test_ONNX_layers, Gather)
{
- if (backend == DNN_BACKEND_INFERENCE_ENGINE_NN_BUILDER_2019 && target == DNN_TARGET_MYRIAD)
- applyTestTag(CV_TEST_TAG_DNN_SKIP_IE_MYRIAD, CV_TEST_TAG_DNN_SKIP_IE_NN_BUILDER);
testONNXModels("gather", npy, 0, 0, false, false);
+}
+
+TEST_P(Test_ONNX_layers, Gather_Scalar)
+{
testONNXModels("gather_scalar", npy, 0, 0, false, false);
}
double sigma, int descriptorType);
CV_WRAP virtual String getDefaultName() const CV_OVERRIDE;
+
+ CV_WRAP virtual void setNFeatures(int maxFeatures) = 0;
+ CV_WRAP virtual int getNFeatures() const = 0;
+
+ CV_WRAP virtual void setNOctaveLayers(int nOctaveLayers) = 0;
+ CV_WRAP virtual int getNOctaveLayers() const = 0;
+
+ CV_WRAP virtual void setContrastThreshold(double contrastThreshold) = 0;
+ CV_WRAP virtual double getContrastThreshold() const = 0;
+
+ CV_WRAP virtual void setEdgeThreshold(double edgeThreshold) = 0;
+ CV_WRAP virtual double getEdgeThreshold() const = 0;
+
+ CV_WRAP virtual void setSigma(double sigma) = 0;
+ CV_WRAP virtual double getSigma() const = 0;
};
typedef SIFT SiftFeatureDetector;
/** @brief Set detection threshold.
@param threshold AGAST detection threshold score.
*/
- CV_WRAP virtual void setThreshold(int threshold) { CV_UNUSED(threshold); return; }
- CV_WRAP virtual int getThreshold() const { return -1; }
+ CV_WRAP virtual void setThreshold(int threshold) = 0;
+ CV_WRAP virtual int getThreshold() const = 0;
/** @brief Set detection octaves.
@param octaves detection octaves. Use 0 to do single scale.
*/
- CV_WRAP virtual void setOctaves(int octaves) { CV_UNUSED(octaves); return; }
- CV_WRAP virtual int getOctaves() const { return -1; }
+ CV_WRAP virtual void setOctaves(int octaves) = 0;
+ CV_WRAP virtual int getOctaves() const = 0;
+ /** @brief Set detection patternScale.
+ @param patternScale apply this scale to the pattern used for sampling the neighbourhood of a
+ keypoint.
+ */
+ CV_WRAP virtual void setPatternScale(float patternScale) = 0;
+ CV_WRAP virtual float getPatternScale() const = 0;
};
/** @brief Class implementing the ORB (*oriented BRIEF*) keypoint detector and descriptor extractor
CV_WRAP virtual void setMaxArea(int maxArea) = 0;
CV_WRAP virtual int getMaxArea() const = 0;
+ CV_WRAP virtual void setMaxVariation(double maxVariation) = 0;
+ CV_WRAP virtual double getMaxVariation() const = 0;
+
+ CV_WRAP virtual void setMinDiversity(double minDiversity) = 0;
+ CV_WRAP virtual double getMinDiversity() const = 0;
+
+ CV_WRAP virtual void setMaxEvolution(int maxEvolution) = 0;
+ CV_WRAP virtual int getMaxEvolution() const = 0;
+
+ CV_WRAP virtual void setAreaThreshold(double areaThreshold) = 0;
+ CV_WRAP virtual double getAreaThreshold() const = 0;
+
+ CV_WRAP virtual void setMinMargin(double min_margin) = 0;
+ CV_WRAP virtual double getMinMargin() const = 0;
+
+ CV_WRAP virtual void setEdgeBlurSize(int edge_blur_size) = 0;
+ CV_WRAP virtual int getEdgeBlurSize() const = 0;
+
CV_WRAP virtual void setPass2Only(bool f) = 0;
CV_WRAP virtual bool getPass2Only() const = 0;
+
CV_WRAP virtual String getDefaultName() const CV_OVERRIDE;
};
CV_WRAP virtual void setBlockSize(int blockSize) = 0;
CV_WRAP virtual int getBlockSize() const = 0;
+ CV_WRAP virtual void setGradientSize(int gradientSize_) = 0;
+ CV_WRAP virtual int getGradientSize() = 0;
+
CV_WRAP virtual void setHarrisDetector(bool val) = 0;
CV_WRAP virtual bool getHarrisDetector() const = 0;
CV_WRAP static Ptr<SimpleBlobDetector>
create(const SimpleBlobDetector::Params ¶meters = SimpleBlobDetector::Params());
+
+ CV_WRAP virtual void setParams(const SimpleBlobDetector::Params& params ) = 0;
+ CV_WRAP virtual SimpleBlobDetector::Params getParams() const = 0;
+
CV_WRAP virtual String getDefaultName() const CV_OVERRIDE;
CV_WRAP virtual const std::vector<std::vector<cv::Point> >& getBlobContours() const;
};
--- /dev/null
+package org.opencv.test.features2d;
+
+import org.opencv.test.OpenCVTestCase;
+import org.opencv.test.OpenCVTestRunner;
+import org.opencv.features2d.AgastFeatureDetector;
+
+public class AGASTFeatureDetectorTest extends OpenCVTestCase {
+
+ AgastFeatureDetector detector;
+
+ @Override
+ protected void setUp() throws Exception {
+ super.setUp();
+ detector = AgastFeatureDetector.create(); // default (10,true,3)
+ }
+
+ public void testCreate() {
+ assertNotNull(detector);
+ }
+
+ public void testDetectListOfMatListOfListOfKeyPoint() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectListOfMatListOfListOfKeyPointListOfMat() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectMatListOfKeyPoint() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectMatListOfKeyPointMat() {
+ fail("Not yet implemented");
+ }
+
+ public void testEmpty() {
+ fail("Not yet implemented");
+ }
+
+ public void testRead() {
+ String filename = OpenCVTestRunner.getTempFileName("xml");
+ writeFile(filename, "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.AgastFeatureDetector</name>\n<threshold>11</threshold>\n<nonmaxSuppression>0</nonmaxSuppression>\n<type>2</type>\n</opencv_storage>\n");
+
+ detector.read(filename);
+
+ assertEquals(11, detector.getThreshold());
+ assertEquals(false, detector.getNonmaxSuppression());
+ assertEquals(2, detector.getType());
+ }
+
+ public void testReadYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+ writeFile(filename, "%YAML:1.0\n---\nname: \"Feature2D.AgastFeatureDetector\"\nthreshold: 11\nnonmaxSuppression: 0\ntype: 2\n");
+
+ detector.read(filename);
+
+ assertEquals(11, detector.getThreshold());
+ assertEquals(false, detector.getNonmaxSuppression());
+ assertEquals(2, detector.getType());
+ }
+
+ public void testWrite() {
+ String filename = OpenCVTestRunner.getTempFileName("xml");
+
+ detector.write(filename);
+
+ String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.AgastFeatureDetector</name>\n<threshold>10</threshold>\n<nonmaxSuppression>1</nonmaxSuppression>\n<type>3</type>\n</opencv_storage>\n";
+ String actual = readFile(filename);
+ actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
+ assertEquals(truth, actual);
+ }
+
+ public void testWriteYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+
+ detector.write(filename);
+
+ String truth = "%YAML:1.0\n---\nname: \"Feature2D.AgastFeatureDetector\"\nthreshold: 10\nnonmaxSuppression: 1\ntype: 3\n";
+ String actual = readFile(filename);
+ actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
+ assertEquals(truth, actual);
+ }
+
+}
--- /dev/null
+package org.opencv.test.features2d;
+
+import org.opencv.test.OpenCVTestCase;
+import org.opencv.test.OpenCVTestRunner;
+import org.opencv.features2d.AKAZE;
+
+public class AKAZEDescriptorExtractorTest extends OpenCVTestCase {
+
+ AKAZE extractor;
+
+ @Override
+ protected void setUp() throws Exception {
+ super.setUp();
+ extractor = AKAZE.create(); // default (5,0,3,0.001f,4,4,1)
+ }
+
+ public void testCreate() {
+ assertNotNull(extractor);
+ }
+
+ public void testDetectListOfMatListOfListOfKeyPoint() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectListOfMatListOfListOfKeyPointListOfMat() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectMatListOfKeyPoint() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectMatListOfKeyPointMat() {
+ fail("Not yet implemented");
+ }
+
+ public void testEmpty() {
+ fail("Not yet implemented");
+ }
+
+ public void testReadYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+ writeFile(filename, "%YAML:1.0\n---\nformat: 3\nname: \"Feature2D.AKAZE\"\ndescriptor: 4\ndescriptor_channels: 2\ndescriptor_size: 32\nthreshold: 0.125\noctaves: 3\nsublevels: 5\ndiffusivity: 2\n");
+
+ extractor.read(filename);
+
+ assertEquals(4, extractor.getDescriptorType());
+ assertEquals(2, extractor.getDescriptorChannels());
+ assertEquals(32, extractor.getDescriptorSize());
+ assertEquals(0.125, extractor.getThreshold());
+ assertEquals(3, extractor.getNOctaves());
+ assertEquals(5, extractor.getNOctaveLayers());
+ assertEquals(2, extractor.getDiffusivity());
+ }
+
+ public void testWriteYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+
+ extractor.write(filename);
+
+ String truth = "%YAML:1.0\n---\nformat: 3\nname: \"Feature2D.AKAZE\"\ndescriptor: 5\ndescriptor_channels: 3\ndescriptor_size: 0\nthreshold: 1.0000000474974513e-03\noctaves: 4\nsublevels: 4\ndiffusivity: 1\n";
+ String actual = readFile(filename);
+ actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
+ assertEquals(truth, actual);
+ }
+
+}
extractor.write(filename);
- String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<descriptorSize>32</descriptorSize>\n</opencv_storage>\n";
+ String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.BRIEF</name>\n<descriptorSize>32</descriptorSize>\n<use_orientation>0</use_orientation>\n</opencv_storage>\n";
assertEquals(truth, readFile(filename));
}
extractor.write(filename);
- String truth = "%YAML:1.0\n---\ndescriptorSize: 32\n";
+ String truth = "%YAML:1.0\n---\nname: \"Feature2D.BRIEF\"\ndescriptorSize: 32\nuse_orientation: 0\n";
assertEquals(truth, readFile(filename));
}
--- /dev/null
+package org.opencv.test.features2d;
+
+import org.opencv.test.OpenCVTestCase;
+import org.opencv.test.OpenCVTestRunner;
+import org.opencv.features2d.BRISK;
+
+public class BRISKDescriptorExtractorTest extends OpenCVTestCase {
+
+ BRISK extractor;
+
+ @Override
+ protected void setUp() throws Exception {
+ super.setUp();
+ extractor = BRISK.create(); // default (30,3,1)
+ }
+
+ public void testCreate() {
+ assertNotNull(extractor);
+ }
+
+ public void testDetectListOfMatListOfListOfKeyPoint() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectListOfMatListOfListOfKeyPointListOfMat() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectMatListOfKeyPoint() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectMatListOfKeyPointMat() {
+ fail("Not yet implemented");
+ }
+
+ public void testEmpty() {
+ fail("Not yet implemented");
+ }
+
+ public void testReadYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+ writeFile(filename, "%YAML:1.0\n---\nname: \"Feature2D.BRISK\"\nthreshold: 31\noctaves: 4\npatternScale: 1.1\n");
+
+ extractor.read(filename);
+
+ assertEquals(31, extractor.getThreshold());
+ assertEquals(4, extractor.getOctaves());
+ assertEquals(1.1f, extractor.getPatternScale());
+ }
+
+ public void testWriteYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+
+ extractor.write(filename);
+
+ String truth = "%YAML:1.0\n---\nname: \"Feature2D.BRISK\"\nthreshold: 30\noctaves: 3\npatternScale: 1.\n";
+ String actual = readFile(filename);
+ actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
+ assertEquals(truth, actual);
+ }
+
+}
import org.opencv.core.MatOfKeyPoint;
import org.opencv.core.Point;
import org.opencv.core.Scalar;
-import org.opencv.features2d.Feature2D;
import org.opencv.features2d.FastFeatureDetector;
import org.opencv.core.KeyPoint;
import org.opencv.test.OpenCVTestCase;
public class FASTFeatureDetectorTest extends OpenCVTestCase {
- Feature2D detector;
+ FastFeatureDetector detector;
KeyPoint[] truth;
private Mat getMaskImg() {
public void testEmpty() {
// assertFalse(detector.empty());
- fail("Not yet implemented"); //FAST does not override empty() method
+ fail("Not yet implemented"); // FAST does not override empty() method
}
public void testRead() {
- String filename = OpenCVTestRunner.getTempFileName("yml");
+ String filename = OpenCVTestRunner.getTempFileName("xml");
- writeFile(filename, "%YAML:1.0\n---\nthreshold: 130\nnonmaxSuppression: 1\n");
+ writeFile(filename, "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.FastFeatureDetector</name>\n<threshold>10</threshold>\n<nonmaxSuppression>1</nonmaxSuppression>\n<type>2</type>\n</opencv_storage>\n");
detector.read(filename);
+ assertEquals(10, detector.getThreshold());
+ assertEquals(true, detector.getNonmaxSuppression());
+ assertEquals(2, detector.getType());
+
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
detector.detect(grayChess, keypoints1);
- writeFile(filename, "%YAML:1.0\n---\nthreshold: 150\nnonmaxSuppression: 1\n");
+ writeFile(filename, "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.FastFeatureDetector</name>\n<threshold>150</threshold>\n<nonmaxSuppression>1</nonmaxSuppression>\n<type>2</type>\n</opencv_storage>\n");
detector.read(filename);
MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
public void testReadYml() {
String filename = OpenCVTestRunner.getTempFileName("yml");
- writeFile(filename,
- "<?xml version=\"1.0\"?>\n<opencv_storage>\n<threshold>130</threshold>\n<nonmaxSuppression>1</nonmaxSuppression>\n</opencv_storage>\n");
+ writeFile(filename, "%YAML:1.0\n---\nthreshold: 130\nnonmaxSuppression: 1\ntype: 2\n");
detector.read(filename);
+ assertEquals(130, detector.getThreshold());
+ assertEquals(true, detector.getNonmaxSuppression());
+ assertEquals(2, detector.getType());
+
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
detector.detect(grayChess, keypoints1);
- writeFile(filename,
- "<?xml version=\"1.0\"?>\n<opencv_storage>\n<threshold>150</threshold>\n<nonmaxSuppression>1</nonmaxSuppression>\n</opencv_storage>\n");
+ writeFile(filename, "%YAML:1.0\n---\nthreshold: 150\nnonmaxSuppression: 1\ntype: 2\n");
detector.read(filename);
MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
assertTrue(keypoints2.total() <= keypoints1.total());
}
- public void testWrite() {
- String filename = OpenCVTestRunner.getTempFileName("xml");
-
- detector.write(filename);
-
-// String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.FAST</name>\n<nonmaxSuppression>1</nonmaxSuppression>\n<threshold>10</threshold>\n<type>2</type>\n</opencv_storage>\n";
- String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n</opencv_storage>\n";
- String data = readFile(filename);
- //Log.d("qqq", "\"" + data + "\"");
- assertEquals(truth, data);
- }
-
public void testWriteYml() {
String filename = OpenCVTestRunner.getTempFileName("yml");
detector.write(filename);
-// String truth = "%YAML:1.0\n---\nname: \"Feature2D.FAST\"\nnonmaxSuppression: 1\nthreshold: 10\ntype: 2\n";
- String truth = "%YAML:1.0\n---\n";
+ String truth = "%YAML:1.0\n---\nname: \"Feature2D.FastFeatureDetector\"\nthreshold: 10\nnonmaxSuppression: 1\ntype: 2\n";
String data = readFile(filename);
- //Log.d("qqq", "\"" + data + "\"");
assertEquals(truth, data);
}
}
public void testPTOD()
{
- String detectorCfg = "%YAML:1.0\n---\nhessianThreshold: 4000.\noctaves: 3\noctaveLayers: 4\nupright: 0\n";
- String extractorCfg = "%YAML:1.0\n---\nnOctaves: 4\nnOctaveLayers: 2\nextended: 0\nupright: 0\n";
+ String detectorCfg = "%YAML:1.0\n---\nhessianThreshold: 4000.\nextended: 0\nupright: 0\nOctaves: 4\nOctaveLayers: 3\n";
+ String extractorCfg = "%YAML:1.0\n---\nhessianThreshold: 4000.\nextended: 0\nupright: 0\nOctaves: 4\nOctaveLayers: 3\n";
Feature2D detector = createClassInstance(XFEATURES2D+"SURF", DEFAULT_FACTORY, null, null);
Feature2D extractor = createClassInstance(XFEATURES2D+"SURF", DEFAULT_FACTORY, null, null);
package org.opencv.test.features2d;
import org.opencv.test.OpenCVTestCase;
+import org.opencv.test.OpenCVTestRunner;
+import org.opencv.features2d.GFTTDetector;
public class GFTTFeatureDetectorTest extends OpenCVTestCase {
+ GFTTDetector detector;
+
+ @Override
+ protected void setUp() throws Exception {
+ super.setUp();
+ detector = GFTTDetector.create(); // default constructor have (1000, 0.01, 1, 3, 3, false, 0.04)
+ }
+
public void testCreate() {
- fail("Not yet implemented");
+ assertNotNull(detector);
}
public void testDetectListOfMatListOfListOfKeyPoint() {
fail("Not yet implemented");
}
- public void testRead() {
- fail("Not yet implemented");
+ public void testReadYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+
+ writeFile(filename, "%YAML:1.0\n---\nname: \"Feature2D.GFTTDetector\"\nnfeatures: 500\nqualityLevel: 2.0000000000000000e-02\nminDistance: 2.\nblockSize: 4\ngradSize: 5\nuseHarrisDetector: 1\nk: 5.0000000000000000e-02\n");
+ detector.read(filename);
+
+ assertEquals(500, detector.getMaxFeatures());
+ assertEquals(0.02, detector.getQualityLevel());
+ assertEquals(2.0, detector.getMinDistance());
+ assertEquals(4, detector.getBlockSize());
+ assertEquals(5, detector.getGradientSize());
+ assertEquals(true, detector.getHarrisDetector());
+ assertEquals(0.05, detector.getK());
}
- public void testWrite() {
- fail("Not yet implemented");
+ public void testWriteYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+
+ detector.write(filename);
+
+ String truth = "%YAML:1.0\n---\nname: \"Feature2D.GFTTDetector\"\nnfeatures: 1000\nqualityLevel: 1.0000000000000000e-02\nminDistance: 1.\nblockSize: 3\ngradSize: 3\nuseHarrisDetector: 0\nk: 4.0000000000000001e-02\n";
+ String actual = readFile(filename);
+ actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
+ assertEquals(truth, actual);
}
}
--- /dev/null
+package org.opencv.test.features2d;
+
+import org.opencv.test.OpenCVTestCase;
+import org.opencv.test.OpenCVTestRunner;
+import org.opencv.features2d.KAZE;
+
+public class KAZEDescriptorExtractorTest extends OpenCVTestCase {
+
+ KAZE extractor;
+
+ @Override
+ protected void setUp() throws Exception {
+ super.setUp();
+ extractor = KAZE.create(); // default (false,false,0.001f,4,4,1)
+ }
+
+ public void testCreate() {
+ assertNotNull(extractor);
+ }
+
+ public void testDetectListOfMatListOfListOfKeyPoint() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectListOfMatListOfListOfKeyPointListOfMat() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectMatListOfKeyPoint() {
+ fail("Not yet implemented");
+ }
+
+ public void testDetectMatListOfKeyPointMat() {
+ fail("Not yet implemented");
+ }
+
+ public void testEmpty() {
+ fail("Not yet implemented");
+ }
+
+ public void testReadYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+ writeFile(filename, "%YAML:1.0\n---\nformat: 3\nname: \"Feature2D.KAZE\"\nextended: 1\nupright: 1\nthreshold: 0.125\noctaves: 3\nsublevels: 5\ndiffusivity: 2\n");
+
+ extractor.read(filename);
+
+ assertEquals(true, extractor.getExtended());
+ assertEquals(true, extractor.getUpright());
+ assertEquals(0.125, extractor.getThreshold());
+ assertEquals(3, extractor.getNOctaves());
+ assertEquals(5, extractor.getNOctaveLayers());
+ assertEquals(2, extractor.getDiffusivity());
+ }
+
+ public void testWriteYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+
+ extractor.write(filename);
+
+ String truth = "%YAML:1.0\n---\nformat: 3\nname: \"Feature2D.KAZE\"\nextended: 0\nupright: 0\nthreshold: 1.0000000474974513e-03\noctaves: 4\nsublevels: 4\ndiffusivity: 1\n";
+ String actual = readFile(filename);
+ actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
+ assertEquals(truth, actual);
+ }
+
+}
package org.opencv.test.features2d;
import org.opencv.test.OpenCVTestCase;
+import org.opencv.test.OpenCVTestRunner;
+import org.opencv.features2d.MSER;
public class MSERFeatureDetectorTest extends OpenCVTestCase {
+ MSER detector;
+
+ @Override
+ protected void setUp() throws Exception {
+ super.setUp();
+ detector = MSER.create(); // default constructor have (5, 60, 14400, .25, .2, 200, 1.01, .003, 5)
+ }
+
public void testCreate() {
- fail("Not yet implemented");
+ assertNotNull(detector);
}
public void testDetectListOfMatListOfListOfKeyPoint() {
fail("Not yet implemented");
}
- public void testRead() {
- fail("Not yet implemented");
+ public void testReadYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+
+ writeFile(filename, "%YAML:1.0\n---\nname: \"Feature2D.MSER\"\ndelta: 6\nminArea: 62\nmaxArea: 14402\nmaxVariation: .26\nminDiversity: .3\nmaxEvolution: 201\nareaThreshold: 1.02\nminMargin: 3.0e-3\nedgeBlurSize: 3\npass2Only: 1\n");
+ detector.read(filename);
+
+ assertEquals(6, detector.getDelta());
+ assertEquals(62, detector.getMinArea());
+ assertEquals(14402, detector.getMaxArea());
+ assertEquals(.26, detector.getMaxVariation());
+ assertEquals(.3, detector.getMinDiversity());
+ assertEquals(201, detector.getMaxEvolution());
+ assertEquals(1.02, detector.getAreaThreshold());
+ assertEquals(0.003, detector.getMinMargin());
+ assertEquals(3, detector.getEdgeBlurSize());
+ assertEquals(true, detector.getPass2Only());
}
- public void testWrite() {
- fail("Not yet implemented");
+ public void testWriteYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+
+ detector.write(filename);
+
+ String truth = "%YAML:1.0\n---\nname: \"Feature2D.MSER\"\ndelta: 5\nminArea: 60\nmaxArea: 14400\nmaxVariation: 2.5000000000000000e-01\nminDiversity: 2.0000000000000001e-01\nmaxEvolution: 200\nareaThreshold: 1.0100000000000000e+00\nminMargin: 3.0000000000000001e-03\nedgeBlurSize: 5\npass2Only: 0\n";
+ String actual = readFile(filename);
+ actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
+ assertEquals(truth, actual);
}
}
fail("Not yet implemented"); // ORB does not override empty() method
}
- public void testRead() {
+ public void testReadYml() {
KeyPoint point = new KeyPoint(55.775577545166016f, 44.224422454833984f, 16, 9.754629f, 8617.863f, 1, -1);
MatOfKeyPoint keypoints = new MatOfKeyPoint(point);
Mat img = getTestImg();
Mat descriptors = new Mat();
-// String filename = OpenCVTestRunner.getTempFileName("yml");
-// writeFile(filename, "%YAML:1.0\n---\nscaleFactor: 1.1\nnLevels: 3\nfirstLevel: 0\nedgeThreshold: 31\npatchSize: 31\n");
-// extractor.read(filename);
- extractor = ORB.create(500, 1.1f, 3, 31, 0, 2, ORB.HARRIS_SCORE, 31, 20);
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+ writeFile(filename, "%YAML:1.0\n---\nnfeatures: 500\nscaleFactor: 1.1\nnlevels: 3\nedgeThreshold: 31\nfirstLevel: 0\nwta_k: 2\nscoreType: 0\npatchSize: 31\nfastThreshold: 20\n");
+ extractor.read(filename);
+
+ assertEquals(500, extractor.getMaxFeatures());
+ assertEquals(1.1, extractor.getScaleFactor());
+ assertEquals(3, extractor.getNLevels());
+ assertEquals(31, extractor.getEdgeThreshold());
+ assertEquals(0, extractor.getFirstLevel());
+ assertEquals(2, extractor.getWTA_K());
+ assertEquals(0, extractor.getScoreType());
+ assertEquals(31, extractor.getPatchSize());
+ assertEquals(20, extractor.getFastThreshold());
extractor.compute(img, keypoints, descriptors);
assertDescriptorsClose(truth, descriptors, 1);
}
- public void testWrite() {
- String filename = OpenCVTestRunner.getTempFileName("xml");
-
- extractor.write(filename);
-
-// String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.ORB</name>\n<WTA_K>2</WTA_K>\n<edgeThreshold>31</edgeThreshold>\n<firstLevel>0</firstLevel>\n<nFeatures>500</nFeatures>\n<nLevels>8</nLevels>\n<patchSize>31</patchSize>\n<scaleFactor>1.2000000476837158e+00</scaleFactor>\n<scoreType>0</scoreType>\n</opencv_storage>\n";
- String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n</opencv_storage>\n";
- String actual = readFile(filename);
- actual = actual.replaceAll("e\\+000", "e+00"); // NOTE: workaround for different platforms double representation
- assertEquals(truth, actual);
- }
-
public void testWriteYml() {
String filename = OpenCVTestRunner.getTempFileName("yml");
extractor.write(filename);
-// String truth = "%YAML:1.0\n---\nname: \"Feature2D.ORB\"\nWTA_K: 2\nedgeThreshold: 31\nfirstLevel: 0\nnFeatures: 500\nnLevels: 8\npatchSize: 31\nscaleFactor: 1.2000000476837158e+00\nscoreType: 0\n";
- String truth = "%YAML:1.0\n---\n";
+ String truth = "%YAML:1.0\n---\nname: \"Feature2D.ORB\"\nnfeatures: 500\nscaleFactor: 1.2000000476837158e+00\nnlevels: 8\nedgeThreshold: 31\nfirstLevel: 0\nwta_k: 2\nscoreType: 0\npatchSize: 31\nfastThreshold: 20\n";
+// String truth = "%YAML:1.0\n---\n";
String actual = readFile(filename);
actual = actual.replaceAll("e\\+000", "e+00"); // NOTE: workaround for different platforms double representation
assertEquals(truth, actual);
import org.opencv.test.OpenCVTestCase;
import org.opencv.test.OpenCVTestRunner;
import org.opencv.imgproc.Imgproc;
-import org.opencv.features2d.Feature2D;
+import org.opencv.features2d.SIFT;
public class SIFTDescriptorExtractorTest extends OpenCVTestCase {
- Feature2D extractor;
+ SIFT extractor;
KeyPoint keypoint;
int matSize;
Mat truth;
117, 112, 117, 76, 117, 54, 117, 25, 29, 22, 117, 117, 16, 11, 14,
1, 0, 0, 22, 26, 0, 0, 0, 0, 1, 4, 15, 2, 47, 8, 0, 0, 82, 56, 31,
17, 81, 12, 0, 0, 26, 23, 18, 23, 0, 0, 0, 0, 0, 0, 0, 0
- );
+ );
}
};
}
public void testEmpty() {
// assertFalse(extractor.empty());
- fail("Not yet implemented"); //SIFT does not override empty() method
+ fail("Not yet implemented"); // SIFT does not override empty() method
}
- public void testRead() {
- fail("Not yet implemented");
- }
+ public void testReadYml() {
+ String filename = OpenCVTestRunner.getTempFileName("yml");
+ writeFile(filename, "%YAML:1.0\n---\nname: \"Feature2D.SIFT\"\nnfeatures: 100\nnOctaveLayers: 4\ncontrastThreshold: 5.0000000000000001e-02\nedgeThreshold: 11\nsigma: 1.7\ndescriptorType: 5\n");
- public void testWrite() {
- String filename = OpenCVTestRunner.getTempFileName("xml");
+ extractor.read(filename);
- extractor.write(filename);
+ assertEquals(128, extractor.descriptorSize());
-// String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.SIFT</name>\n<contrastThreshold>4.0000000000000001e-02</contrastThreshold>\n<edgeThreshold>10.</edgeThreshold>\n<nFeatures>0</nFeatures>\n<nOctaveLayers>3</nOctaveLayers>\n<sigma>1.6000000000000001e+00</sigma>\n</opencv_storage>\n";
- String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n</opencv_storage>\n";
- String actual = readFile(filename);
- actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
- assertEquals(truth, actual);
+ assertEquals(100, extractor.getNFeatures());
+ assertEquals(4, extractor.getNOctaveLayers());
+ assertEquals(0.05, extractor.getContrastThreshold());
+ assertEquals(11., extractor.getEdgeThreshold());
+ assertEquals(1.7, extractor.getSigma());
+ assertEquals(5, extractor.descriptorType());
}
public void testWriteYml() {
extractor.write(filename);
-// String truth = "%YAML:1.0\n---\nname: \"Feature2D.SIFT\"\ncontrastThreshold: 4.0000000000000001e-02\nedgeThreshold: 10.\nnFeatures: 0\nnOctaveLayers: 3\nsigma: 1.6000000000000001e+00\n";
- String truth = "%YAML:1.0\n---\n";
+ String truth = "%YAML:1.0\n---\nname: \"Feature2D.SIFT\"\nnfeatures: 0\nnOctaveLayers: 3\ncontrastThreshold: 4.0000000000000001e-02\nedgeThreshold: 10.\nsigma: 1.6000000000000001e+00\ndescriptorType: 5\n";
String actual = readFile(filename);
actual = actual.replaceAll("e([+-])0(\\d\\d)", "e$1$2"); // NOTE: workaround for different platforms double representation
assertEquals(truth, actual);
import org.opencv.test.OpenCVTestCase;
import org.opencv.test.OpenCVTestRunner;
import org.opencv.imgproc.Imgproc;
-import org.opencv.features2d.Feature2D;
import org.opencv.features2d.SimpleBlobDetector;
+import org.opencv.features2d.SimpleBlobDetector_Params;
public class SIMPLEBLOBFeatureDetectorTest extends OpenCVTestCase {
- Feature2D detector;
+ SimpleBlobDetector detector;
int matSize;
KeyPoint[] truth;
detector = SimpleBlobDetector.create();
matSize = 200;
truth = new KeyPoint[] {
- new KeyPoint( 140, 100, 41.036568f, -1, 0, 0, -1),
- new KeyPoint( 60, 100, 48.538486f, -1, 0, 0, -1),
+ new KeyPoint(140, 100, 41.036568f, -1, 0, 0, -1),
+ new KeyPoint(60, 100, 48.538486f, -1, 0, 0, -1),
new KeyPoint(100, 60, 36.769554f, -1, 0, 0, -1),
new KeyPoint(100, 140, 28.635643f, -1, 0, 0, -1),
new KeyPoint(100, 100, 20.880613f, -1, 0, 0, -1)
fail("Not yet implemented");
}
- public void testRead() {
+ public void testReadYml() {
Mat img = getTestImg();
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
detector.detect(img, keypoints1);
String filename = OpenCVTestRunner.getTempFileName("yml");
- writeFile(filename, "%YAML:1.0\nthresholdStep: 10\nminThreshold: 50\nmaxThreshold: 220\nminRepeatability: 2\nfilterByArea: true\nminArea: 800\nmaxArea: 5000\n");
+ writeFile(filename, "%YAML:1.0\nthresholdStep: 10.0\nminThreshold: 50\nmaxThreshold: 220\nminRepeatability: 2\nminDistBetweenBlobs: 10.\nfilterByColor: 1\nblobColor: 0\nfilterByArea: 1\nminArea: 800\nmaxArea: 6000\nfilterByCircularity: 0\nminCircularity: 0.7\nmaxCircularity: 10.\nfilterByInertia: 1\nminInertiaRatio: 0.2\nmaxInertiaRatio: 11.\nfilterByConvexity: true\nminConvexity: 0.9\nmaxConvexity: 12.\n");
detector.read(filename);
+ SimpleBlobDetector_Params params = detector.getParams();
+ assertEquals(10.0f, params.get_thresholdStep());
+ assertEquals(50f, params.get_minThreshold());
+ assertEquals(220f, params.get_maxThreshold());
+ assertEquals(2, params.get_minRepeatability());
+ assertEquals(10.0f, params.get_minDistBetweenBlobs());
+ assertEquals(true, params.get_filterByColor());
+ // FIXME: blobColor field has uchar type in C++ and cannot be automatically wrapped to Java as it does not support unsigned types
+ //assertEquals(0, params.get_blobColor());
+ assertEquals(true, params.get_filterByArea());
+ assertEquals(800f, params.get_minArea());
+ assertEquals(6000f, params.get_maxArea());
+ assertEquals(false, params.get_filterByCircularity());
+ assertEquals(0.7f, params.get_minCircularity());
+ assertEquals(10.0f, params.get_maxCircularity());
+ assertEquals(true, params.get_filterByInertia());
+ assertEquals(0.2f, params.get_minInertiaRatio());
+ assertEquals(11.0f, params.get_maxInertiaRatio());
+ assertEquals(true, params.get_filterByConvexity());
+ assertEquals(0.9f, params.get_minConvexity());
+ assertEquals(12.0f, params.get_maxConvexity());
+
MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
detector.detect(img, keypoints2);
+++ /dev/null
-package org.opencv.test.features2d;
-
-import java.util.Arrays;
-
-import org.opencv.core.CvType;
-import org.opencv.core.Mat;
-import org.opencv.core.MatOfKeyPoint;
-import org.opencv.core.Point;
-import org.opencv.core.Scalar;
-import org.opencv.core.KeyPoint;
-import org.opencv.test.OpenCVTestCase;
-import org.opencv.test.OpenCVTestRunner;
-import org.opencv.imgproc.Imgproc;
-import org.opencv.features2d.Feature2D;
-
-public class STARFeatureDetectorTest extends OpenCVTestCase {
-
- Feature2D detector;
- int matSize;
- KeyPoint[] truth;
-
- private Mat getMaskImg() {
- Mat mask = new Mat(matSize, matSize, CvType.CV_8U, new Scalar(255));
- Mat right = mask.submat(0, matSize, matSize / 2, matSize);
- right.setTo(new Scalar(0));
- return mask;
- }
-
- private Mat getTestImg() {
- Scalar color = new Scalar(0);
- int center = matSize / 2;
- int radius = 6;
- int offset = 40;
-
- Mat img = new Mat(matSize, matSize, CvType.CV_8U, new Scalar(255));
- Imgproc.circle(img, new Point(center - offset, center), radius, color, -1);
- Imgproc.circle(img, new Point(center + offset, center), radius, color, -1);
- Imgproc.circle(img, new Point(center, center - offset), radius, color, -1);
- Imgproc.circle(img, new Point(center, center + offset), radius, color, -1);
- Imgproc.circle(img, new Point(center, center), radius, color, -1);
- return img;
- }
-
- protected void setUp() throws Exception {
- super.setUp();
- detector = createClassInstance(XFEATURES2D+"StarDetector", DEFAULT_FACTORY, null, null);
- matSize = 200;
- truth = new KeyPoint[] {
- new KeyPoint( 95, 80, 22, -1, 31.5957f, 0, -1),
- new KeyPoint(105, 80, 22, -1, 31.5957f, 0, -1),
- new KeyPoint( 80, 95, 22, -1, 31.5957f, 0, -1),
- new KeyPoint(120, 95, 22, -1, 31.5957f, 0, -1),
- new KeyPoint(100, 100, 8, -1, 30.f, 0, -1),
- new KeyPoint( 80, 105, 22, -1, 31.5957f, 0, -1),
- new KeyPoint(120, 105, 22, -1, 31.5957f, 0, -1),
- new KeyPoint( 95, 120, 22, -1, 31.5957f, 0, -1),
- new KeyPoint(105, 120, 22, -1, 31.5957f, 0, -1)
- };
- }
-
- public void testCreate() {
- assertNotNull(detector);
- }
-
- public void testDetectListOfMatListOfListOfKeyPoint() {
- fail("Not yet implemented");
- }
-
- public void testDetectListOfMatListOfListOfKeyPointListOfMat() {
- fail("Not yet implemented");
- }
-
- public void testDetectMatListOfKeyPoint() {
- Mat img = getTestImg();
- MatOfKeyPoint keypoints = new MatOfKeyPoint();
-
- detector.detect(img, keypoints);
-
- assertListKeyPointEquals(Arrays.asList(truth), keypoints.toList(), EPS);
- }
-
- public void testDetectMatListOfKeyPointMat() {
- Mat img = getTestImg();
- Mat mask = getMaskImg();
- MatOfKeyPoint keypoints = new MatOfKeyPoint();
-
- detector.detect(img, keypoints, mask);
-
- assertListKeyPointEquals(Arrays.asList(truth[0], truth[2], truth[5], truth[7]), keypoints.toList(), EPS);
- }
-
- public void testEmpty() {
-// assertFalse(detector.empty());
- fail("Not yet implemented");
- }
-
- public void testRead() {
- Mat img = getTestImg();
-
- MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
- detector.detect(img, keypoints1);
-
- String filename = OpenCVTestRunner.getTempFileName("yml");
- writeFile(filename, "%YAML:1.0\n---\nmaxSize: 45\nresponseThreshold: 150\nlineThresholdProjected: 10\nlineThresholdBinarized: 8\nsuppressNonmaxSize: 5\n");
- detector.read(filename);
-
- MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
- detector.detect(img, keypoints2);
-
- assertTrue(keypoints2.total() <= keypoints1.total());
- }
-
- public void testWrite() {
- String filename = OpenCVTestRunner.getTempFileName("xml");
-
- detector.write(filename);
-
-// String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.STAR</name>\n<lineThresholdBinarized>8</lineThresholdBinarized>\n<lineThresholdProjected>10</lineThresholdProjected>\n<maxSize>45</maxSize>\n<responseThreshold>30</responseThreshold>\n<suppressNonmaxSize>5</suppressNonmaxSize>\n</opencv_storage>\n";
- String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n</opencv_storage>\n";
- assertEquals(truth, readFile(filename));
- }
-
- public void testWriteYml() {
- String filename = OpenCVTestRunner.getTempFileName("yml");
-
- detector.write(filename);
-
-// String truth = "%YAML:1.0\n---\nname: \"Feature2D.STAR\"\nlineThresholdBinarized: 8\nlineThresholdProjected: 10\nmaxSize: 45\nresponseThreshold: 30\nsuppressNonmaxSize: 5\n";
- String truth = "%YAML:1.0\n---\n";
- assertEquals(truth, readFile(filename));
- }
-
-}
+++ /dev/null
-package org.opencv.test.features2d;
-
-import org.opencv.core.CvType;
-import org.opencv.core.Mat;
-import org.opencv.core.MatOfKeyPoint;
-import org.opencv.core.Point;
-import org.opencv.core.Scalar;
-import org.opencv.core.KeyPoint;
-import org.opencv.test.OpenCVTestCase;
-import org.opencv.test.OpenCVTestRunner;
-import org.opencv.imgproc.Imgproc;
-import org.opencv.features2d.Feature2D;
-
-public class SURFDescriptorExtractorTest extends OpenCVTestCase {
-
- Feature2D extractor;
- int matSize;
-
- private Mat getTestImg() {
- Mat cross = new Mat(matSize, matSize, CvType.CV_8U, new Scalar(255));
- Imgproc.line(cross, new Point(20, matSize / 2), new Point(matSize - 21, matSize / 2), new Scalar(100), 2);
- Imgproc.line(cross, new Point(matSize / 2, 20), new Point(matSize / 2, matSize - 21), new Scalar(100), 2);
-
- return cross;
- }
-
- @Override
- protected void setUp() throws Exception {
- super.setUp();
-
- Class[] cParams = {double.class, int.class, int.class, boolean.class, boolean.class};
- Object[] oValues = {100, 2, 4, true, false};
- extractor = createClassInstance(XFEATURES2D+"SURF", DEFAULT_FACTORY, cParams, oValues);
-
- matSize = 100;
- }
-
- public void testComputeListOfMatListOfListOfKeyPointListOfMat() {
- fail("Not yet implemented");
- }
-
- public void testComputeMatListOfKeyPointMat() {
- KeyPoint point = new KeyPoint(55.775577545166016f, 44.224422454833984f, 16, 9.754629f, 8617.863f, 1, -1);
- MatOfKeyPoint keypoints = new MatOfKeyPoint(point);
- Mat img = getTestImg();
- Mat descriptors = new Mat();
-
- extractor.compute(img, keypoints, descriptors);
-
- Mat truth = new Mat(1, 128, CvType.CV_32FC1) {
- {
- put(0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0.058821894, 0.058821894, -0.045962855, 0.046261817, 0.0085156476,
- 0.0085754395, -0.0064509804, 0.0064509804, 0.00044069235, 0.00044069235, 0, 0, 0.00025723741,
- 0.00025723741, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.00025723741, 0.00025723741, -0.00044069235,
- 0.00044069235, 0, 0, 0.36278215, 0.36278215, -0.24688604, 0.26173124, 0.052068226, 0.052662034,
- -0.032815345, 0.032815345, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, -0.0064523756,
- 0.0064523756, 0.0082002236, 0.0088908644, -0.059001274, 0.059001274, 0.045789491, 0.04648013,
- 0.11961588, 0.22789426, -0.01322381, 0.18291828, -0.14042182, 0.23973691, 0.073782086, 0.23769434,
- -0.027880307, 0.027880307, 0.049587864, 0.049587864, -0.33991757, 0.33991757, 0.21437603, 0.21437603,
- -0.0020763327, 0.0020763327, 0.006245892, 0.006245892, -0.04067041, 0.04067041, 0.019361559,
- 0.019361559, 0, 0, -0.0035977389, 0.0035977389, 0, 0, -0.00099993451, 0.00099993451, 0.040670406,
- 0.040670406, -0.019361559, 0.019361559, 0.006245892, 0.006245892, -0.0020763327, 0.0020763327,
- -0.00034532088, 0.00034532088, 0, 0, 0, 0, 0.00034532088, 0.00034532088, -0.00099993451,
- 0.00099993451, 0, 0, 0, 0, 0.0035977389, 0.0035977389
- );
- }
- };
-
- assertMatEqual(truth, descriptors, EPS);
- }
-
- public void testCreate() {
- assertNotNull(extractor);
- }
-
- public void testDescriptorSize() {
- assertEquals(128, extractor.descriptorSize());
- }
-
- public void testDescriptorType() {
- assertEquals(CvType.CV_32F, extractor.descriptorType());
- }
-
- public void testEmpty() {
-// assertFalse(extractor.empty());
- fail("Not yet implemented");
- }
-
- public void testRead() {
- String filename = OpenCVTestRunner.getTempFileName("yml");
- writeFile(filename, "%YAML:1.0\n---\nnOctaves: 4\nnOctaveLayers: 2\nextended: 1\nupright: 0\n");
-
- extractor.read(filename);
-
- assertEquals(128, extractor.descriptorSize());
- }
-
- public void testWrite() {
- String filename = OpenCVTestRunner.getTempFileName("xml");
-
- extractor.write(filename);
-
-// String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.SURF</name>\n<extended>1</extended>\n<hessianThreshold>100.</hessianThreshold>\n<nOctaveLayers>2</nOctaveLayers>\n<nOctaves>4</nOctaves>\n<upright>0</upright>\n</opencv_storage>\n";
- String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n</opencv_storage>\n";
- assertEquals(truth, readFile(filename));
- }
-
- public void testWriteYml() {
- String filename = OpenCVTestRunner.getTempFileName("yml");
-
- extractor.write(filename);
-
-// String truth = "%YAML:1.0\n---\nname: \"Feature2D.SURF\"\nextended: 1\nhessianThreshold: 100.\nnOctaveLayers: 2\nnOctaves: 4\nupright: 0\n";
- String truth = "%YAML:1.0\n---\n";
- assertEquals(truth, readFile(filename));
- }
-
-}
+++ /dev/null
-package org.opencv.test.features2d;
-
-import java.util.ArrayList;
-import java.util.Arrays;
-import java.util.Collections;
-import java.util.Comparator;
-import java.util.List;
-
-import org.opencv.core.CvType;
-import org.opencv.core.Mat;
-import org.opencv.core.MatOfKeyPoint;
-import org.opencv.core.Point;
-import org.opencv.core.Scalar;
-import org.opencv.core.KeyPoint;
-import org.opencv.test.OpenCVTestCase;
-import org.opencv.test.OpenCVTestRunner;
-import org.opencv.imgproc.Imgproc;
-import org.opencv.features2d.Feature2D;
-
-public class SURFFeatureDetectorTest extends OpenCVTestCase {
-
- Feature2D detector;
- int matSize;
- KeyPoint[] truth;
-
- private Mat getMaskImg() {
- Mat mask = new Mat(matSize, matSize, CvType.CV_8U, new Scalar(255));
- Mat right = mask.submat(0, matSize, matSize / 2, matSize);
- right.setTo(new Scalar(0));
- return mask;
- }
-
- private Mat getTestImg() {
- Mat cross = new Mat(matSize, matSize, CvType.CV_8U, new Scalar(255));
- Imgproc.line(cross, new Point(20, matSize / 2), new Point(matSize - 21, matSize / 2), new Scalar(100), 2);
- Imgproc.line(cross, new Point(matSize / 2, 20), new Point(matSize / 2, matSize - 21), new Scalar(100), 2);
-
- return cross;
- }
-
- private void order(List<KeyPoint> points) {
- Collections.sort(points, new Comparator<KeyPoint>() {
- public int compare(KeyPoint p1, KeyPoint p2) {
- if (p1.angle < p2.angle)
- return -1;
- if (p1.angle > p2.angle)
- return 1;
- return 0;
- }
- });
- }
-
- @Override
- protected void setUp() throws Exception {
- super.setUp();
- detector = createClassInstance(XFEATURES2D+"SURF", DEFAULT_FACTORY, null, null);
- matSize = 100;
- truth = new KeyPoint[] {
- new KeyPoint(55.775578f, 55.775578f, 16, 80.245735f, 8617.8633f, 0, -1),
- new KeyPoint(44.224422f, 55.775578f, 16, 170.24574f, 8617.8633f, 0, -1),
- new KeyPoint(44.224422f, 44.224422f, 16, 260.24573f, 8617.8633f, 0, -1),
- new KeyPoint(55.775578f, 44.224422f, 16, 350.24573f, 8617.8633f, 0, -1)
- };
- }
-
- public void testCreate() {
- assertNotNull(detector);
- }
-
- public void testDetectListOfMatListOfListOfKeyPoint() {
-
- setProperty(detector, "hessianThreshold", "double", 8000);
- setProperty(detector, "nOctaves", "int", 3);
- setProperty(detector, "nOctaveLayers", "int", 4);
- setProperty(detector, "upright", "boolean", false);
-
- List<MatOfKeyPoint> keypoints = new ArrayList<MatOfKeyPoint>();
- Mat cross = getTestImg();
- List<Mat> crosses = new ArrayList<Mat>(3);
- crosses.add(cross);
- crosses.add(cross);
- crosses.add(cross);
-
- detector.detect(crosses, keypoints);
-
- assertEquals(3, keypoints.size());
-
- for (MatOfKeyPoint mkp : keypoints) {
- List<KeyPoint> lkp = mkp.toList();
- order(lkp);
- assertListKeyPointEquals(Arrays.asList(truth), lkp, EPS);
- }
- }
-
- public void testDetectListOfMatListOfListOfKeyPointListOfMat() {
- fail("Not yet implemented");
- }
-
- public void testDetectMatListOfKeyPoint() {
-
- setProperty(detector, "hessianThreshold", "double", 8000);
- setProperty(detector, "nOctaves", "int", 3);
- setProperty(detector, "nOctaveLayers", "int", 4);
- setProperty(detector, "upright", "boolean", false);
-
- MatOfKeyPoint keypoints = new MatOfKeyPoint();
- Mat cross = getTestImg();
-
- detector.detect(cross, keypoints);
-
- List<KeyPoint> lkp = keypoints.toList();
- order(lkp);
- assertListKeyPointEquals(Arrays.asList(truth), lkp, EPS);
- }
-
- public void testDetectMatListOfKeyPointMat() {
-
- setProperty(detector, "hessianThreshold", "double", 8000);
- setProperty(detector, "nOctaves", "int", 3);
- setProperty(detector, "nOctaveLayers", "int", 4);
- setProperty(detector, "upright", "boolean", false);
-
- Mat img = getTestImg();
- Mat mask = getMaskImg();
- MatOfKeyPoint keypoints = new MatOfKeyPoint();
-
- detector.detect(img, keypoints, mask);
-
- List<KeyPoint> lkp = keypoints.toList();
- order(lkp);
- assertListKeyPointEquals(Arrays.asList(truth[1], truth[2]), lkp, EPS);
- }
-
- public void testEmpty() {
-// assertFalse(detector.empty());
- fail("Not yet implemented");
- }
-
- public void testRead() {
- Mat cross = getTestImg();
-
- MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
- detector.detect(cross, keypoints1);
-
- String filename = OpenCVTestRunner.getTempFileName("yml");
- writeFile(filename, "%YAML:1.0\n---\nhessianThreshold: 8000.\noctaves: 3\noctaveLayers: 4\nupright: 0\n");
- detector.read(filename);
-
- MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
- detector.detect(cross, keypoints2);
-
- assertTrue(keypoints2.total() <= keypoints1.total());
- }
-
- public void testWrite() {
- String filename = OpenCVTestRunner.getTempFileName("xml");
-
- detector.write(filename);
-
-// String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n<name>Feature2D.SURF</name>\n<extended>0</extended>\n<hessianThreshold>100.</hessianThreshold>\n<nOctaveLayers>3</nOctaveLayers>\n<nOctaves>4</nOctaves>\n<upright>0</upright>\n</opencv_storage>\n";
- String truth = "<?xml version=\"1.0\"?>\n<opencv_storage>\n</opencv_storage>\n";
- assertEquals(truth, readFile(filename));
- }
-
- public void testWriteYml() {
- String filename = OpenCVTestRunner.getTempFileName("yml");
-
- detector.write(filename);
-
-// String truth = "%YAML:1.0\n---\nname: \"Feature2D.SURF\"\nextended: 0\nhessianThreshold: 100.\nnOctaveLayers: 3\nnOctaves: 4\nupright: 0\n";
- String truth = "%YAML:1.0\n---\n";
- assertEquals(truth, readFile(filename));
- }
-
-}
{
+ "ManualFuncs" : {
+ "SimpleBlobDetector": {
+ "setParams": { "declaration" : [""], "implementation" : [""] },
+ "getParams": { "declaration" : [""], "implementation" : [""] }
+ }
+ },
"enum_fix" : {
"FastFeatureDetector" : { "DetectorType": "FastDetectorType" },
"AgastFeatureDetector" : { "DetectorType": "AgastDetectorType" }
: threshold(_threshold), nonmaxSuppression(_nonmaxSuppression), type(_type)
{}
+ void read( const FileNode& fn) CV_OVERRIDE
+ {
+ // if node is empty, keep previous value
+ if (!fn["threshold"].empty())
+ fn["threshold"] >> threshold;
+ if (!fn["nonmaxSuppression"].empty())
+ fn["nonmaxSuppression"] >> nonmaxSuppression;
+ if (!fn["type"].empty())
+ fn["type"] >> type;
+ }
+ void write( FileStorage& fs) const CV_OVERRIDE
+ {
+ if(fs.isOpened())
+ {
+ fs << "name" << getDefaultName();
+ fs << "threshold" << threshold;
+ fs << "nonmaxSuppression" << nonmaxSuppression;
+ fs << "type" << type;
+ }
+ }
+
void detect( InputArray _image, std::vector<KeyPoint>& keypoints, InputArray _mask ) CV_OVERRIDE
{
CV_INSTRUMENT_REGION();
void write(FileStorage& fs) const CV_OVERRIDE
{
writeFormat(fs);
+ fs << "name" << getDefaultName();
fs << "descriptor" << descriptor;
fs << "descriptor_channels" << descriptor_channels;
fs << "descriptor_size" << descriptor_size;
void read(const FileNode& fn) CV_OVERRIDE
{
- descriptor = static_cast<DescriptorType>((int)fn["descriptor"]);
- descriptor_channels = (int)fn["descriptor_channels"];
- descriptor_size = (int)fn["descriptor_size"];
- threshold = (float)fn["threshold"];
- octaves = (int)fn["octaves"];
- sublevels = (int)fn["sublevels"];
- diffusivity = static_cast<KAZE::DiffusivityType>((int)fn["diffusivity"]);
+ // if node is empty, keep previous value
+ if (!fn["descriptor"].empty())
+ descriptor = static_cast<DescriptorType>((int)fn["descriptor"]);
+ if (!fn["descriptor_channels"].empty())
+ descriptor_channels = (int)fn["descriptor_channels"];
+ if (!fn["descriptor_size"].empty())
+ descriptor_size = (int)fn["descriptor_size"];
+ if (!fn["threshold"].empty())
+ threshold = (float)fn["threshold"];
+ if (!fn["octaves"].empty())
+ octaves = (int)fn["octaves"];
+ if (!fn["sublevels"].empty())
+ sublevels = (int)fn["sublevels"];
+ if (!fn["diffusivity"].empty())
+ diffusivity = static_cast<KAZE::DiffusivityType>((int)fn["diffusivity"]);
}
DescriptorType descriptor;
virtual void read( const FileNode& fn ) CV_OVERRIDE;
virtual void write( FileStorage& fs ) const CV_OVERRIDE;
+ void setParams(const SimpleBlobDetector::Params& _params ) CV_OVERRIDE {
+ SimpleBlobDetectorImpl::validateParameters(_params);
+ params = _params;
+ }
+
+ SimpleBlobDetector::Params getParams() const CV_OVERRIDE { return params; }
+
+ static void validateParameters(const SimpleBlobDetector::Params& p)
+ {
+ if (p.thresholdStep <= 0)
+ CV_Error(Error::StsBadArg, "thresholdStep>0");
+
+ if (p.minThreshold > p.maxThreshold || p.minThreshold <= 0)
+ CV_Error(Error::StsBadArg, "0<minThreshold<=maxThreshold");
+
+ if (p.minDistBetweenBlobs <=0 )
+ CV_Error(Error::StsBadArg, "minDistBetweenBlobs>0");
+
+ if (p.minArea > p.maxArea || p.minArea <=0)
+ CV_Error(Error::StsBadArg, "0<minArea<=maxArea");
+
+ if (p.minCircularity > p.maxCircularity || p.minCircularity <= 0)
+ CV_Error(Error::StsBadArg, "0<minCircularity<=maxCircularity");
+
+ if (p.minInertiaRatio > p.maxInertiaRatio || p.minInertiaRatio <= 0)
+ CV_Error(Error::StsBadArg, "0<minInertiaRatio<=maxInertiaRatio");
+
+ if (p.minConvexity > p.maxConvexity || p.minConvexity <= 0)
+ CV_Error(Error::StsBadArg, "0<minConvexity<=maxConvexity");
+ }
+
protected:
struct CV_EXPORTS Center
{
void SimpleBlobDetectorImpl::read( const cv::FileNode& fn )
{
- params.read(fn);
+ SimpleBlobDetector::Params rp;
+ rp.read(fn);
+ SimpleBlobDetectorImpl::validateParameters(rp);
+ params = rp;
}
void SimpleBlobDetectorImpl::write( cv::FileStorage& fs ) const
Ptr<SimpleBlobDetector> SimpleBlobDetector::create(const SimpleBlobDetector::Params& params)
{
+ SimpleBlobDetectorImpl::validateParameters(params);
return makePtr<SimpleBlobDetectorImpl>(params);
}
class BRISK_Impl CV_FINAL : public BRISK
{
public:
- explicit BRISK_Impl(int thresh=30, int octaves=3, float patternScale=1.0f);
+ explicit BRISK_Impl(int _threshold=30, int _octaves=3, float _patternScale=1.0f);
// custom setup
explicit BRISK_Impl(const std::vector<float> &radiusList, const std::vector<int> &numberList,
float dMax=5.85f, float dMin=8.2f, const std::vector<int> indexChange=std::vector<int>());
virtual ~BRISK_Impl();
+ void read( const FileNode& fn) CV_OVERRIDE;
+ void write( FileStorage& fs) const CV_OVERRIDE;
+
int descriptorSize() const CV_OVERRIDE
{
return strings_;
{
return octaves;
}
+ virtual void setPatternScale(float _patternScale) CV_OVERRIDE
+ {
+ patternScale = _patternScale;
+ std::vector<float> rList;
+ std::vector<int> nList;
+
+ // this is the standard pattern found to be suitable also
+ rList.resize(5);
+ nList.resize(5);
+ const double f = 0.85 * patternScale;
+
+ rList[0] = (float)(f * 0.);
+ rList[1] = (float)(f * 2.9);
+ rList[2] = (float)(f * 4.9);
+ rList[3] = (float)(f * 7.4);
+ rList[4] = (float)(f * 10.8);
+
+ nList[0] = 1;
+ nList[1] = 10;
+ nList[2] = 14;
+ nList[3] = 15;
+ nList[4] = 20;
+
+ generateKernel(rList, nList, (float)(5.85 * patternScale), (float)(8.2 * patternScale));
+ }
+ virtual float getPatternScale() const CV_OVERRIDE
+ {
+ return patternScale;
+ }
// call this to generate the kernel:
// circle of radius r (pixels), with n points;
// Feature parameters
CV_PROP_RW int threshold;
CV_PROP_RW int octaves;
+ CV_PROP_RW float patternScale;
// some helper structures for the Brisk pattern representation
struct BriskPatternPoint{
const float BriskScaleSpace::basicSize_ = 12.0f;
// constructors
-BRISK_Impl::BRISK_Impl(int thresh, int octaves_in, float patternScale)
+BRISK_Impl::BRISK_Impl(int _threshold, int _octaves, float _patternScale)
{
- threshold = thresh;
- octaves = octaves_in;
-
- std::vector<float> rList;
- std::vector<int> nList;
-
- // this is the standard pattern found to be suitable also
- rList.resize(5);
- nList.resize(5);
- const double f = 0.85 * patternScale;
-
- rList[0] = (float)(f * 0.);
- rList[1] = (float)(f * 2.9);
- rList[2] = (float)(f * 4.9);
- rList[3] = (float)(f * 7.4);
- rList[4] = (float)(f * 10.8);
+ threshold = _threshold;
+ octaves = _octaves;
- nList[0] = 1;
- nList[1] = 10;
- nList[2] = 14;
- nList[3] = 15;
- nList[4] = 20;
-
- generateKernel(rList, nList, (float)(5.85 * patternScale), (float)(8.2 * patternScale));
+ setPatternScale(_patternScale);
}
BRISK_Impl::BRISK_Impl(const std::vector<float> &radiusList,
octaves = octaves_in;
}
+void BRISK_Impl::read( const FileNode& fn)
+{
+ // if node is empty, keep previous value
+ if (!fn["threshold"].empty())
+ fn["threshold"] >> threshold;
+ if (!fn["octaves"].empty())
+ fn["octaves"] >> octaves;
+ if (!fn["patternScale"].empty())
+ {
+ float _patternScale;
+ fn["patternScale"] >> _patternScale;
+ setPatternScale(_patternScale);
+ }
+}
+void BRISK_Impl::write( FileStorage& fs) const
+{
+ if(fs.isOpened())
+ {
+ fs << "name" << getDefaultName();
+ fs << "threshold" << threshold;
+ fs << "octaves" << octaves;
+ fs << "patternScale" << patternScale;
+ }
+}
+
void
BRISK_Impl::generateKernel(const std::vector<float> &radiusList,
const std::vector<int> &numberList,
: threshold(_threshold), nonmaxSuppression(_nonmaxSuppression), type(_type)
{}
+ void read( const FileNode& fn) CV_OVERRIDE
+ {
+ // if node is empty, keep previous value
+ if (!fn["threshold"].empty())
+ fn["threshold"] >> threshold;
+ if (!fn["nonmaxSuppression"].empty())
+ fn["nonmaxSuppression"] >> nonmaxSuppression;
+ if (!fn["type"].empty())
+ fn["type"] >> type;
+ }
+ void write( FileStorage& fs) const CV_OVERRIDE
+ {
+ if(fs.isOpened())
+ {
+ fs << "name" << getDefaultName();
+ fs << "threshold" << threshold;
+ fs << "nonmaxSuppression" << nonmaxSuppression;
+ fs << "type" << type;
+ }
+ }
+
void detect( InputArray _image, std::vector<KeyPoint>& keypoints, InputArray _mask ) CV_OVERRIDE
{
CV_INSTRUMENT_REGION();
{
}
+ void read( const FileNode& fn) CV_OVERRIDE
+ {
+ // if node is empty, keep previous value
+ if (!fn["nfeatures"].empty())
+ fn["nfeatures"] >> nfeatures;
+ if (!fn["qualityLevel"].empty())
+ fn["qualityLevel"] >> qualityLevel;
+ if (!fn["minDistance"].empty())
+ fn["minDistance"] >> minDistance;
+ if (!fn["blockSize"].empty())
+ fn["blockSize"] >> blockSize;
+ if (!fn["gradSize"].empty())
+ fn["gradSize"] >> gradSize;
+ if (!fn["useHarrisDetector"].empty())
+ fn["useHarrisDetector"] >> useHarrisDetector;
+ if (!fn["k"].empty())
+ fn["k"] >> k;
+ }
+ void write( FileStorage& fs) const CV_OVERRIDE
+ {
+ if(fs.isOpened())
+ {
+ fs << "name" << getDefaultName();
+ fs << "nfeatures" << nfeatures;
+ fs << "qualityLevel" << qualityLevel;
+ fs << "minDistance" << minDistance;
+ fs << "blockSize" << blockSize;
+ fs << "gradSize" << gradSize;
+ fs << "useHarrisDetector" << useHarrisDetector;
+ fs << "k" << k;
+ }
+ }
+
void setMaxFeatures(int maxFeatures) CV_OVERRIDE { nfeatures = maxFeatures; }
int getMaxFeatures() const CV_OVERRIDE { return nfeatures; }
void setBlockSize(int blockSize_) CV_OVERRIDE { blockSize = blockSize_; }
int getBlockSize() const CV_OVERRIDE { return blockSize; }
- //void setGradientSize(int gradientSize_) { gradSize = gradientSize_; }
- //int getGradientSize() { return gradSize; }
+ void setGradientSize(int gradientSize_) CV_OVERRIDE { gradSize = gradientSize_; }
+ int getGradientSize() CV_OVERRIDE { return gradSize; }
void setHarrisDetector(bool val) CV_OVERRIDE { useHarrisDetector = val; }
bool getHarrisDetector() const CV_OVERRIDE { return useHarrisDetector; }
void write(FileStorage& fs) const CV_OVERRIDE
{
writeFormat(fs);
+ fs << "name" << getDefaultName();
fs << "extended" << (int)extended;
fs << "upright" << (int)upright;
fs << "threshold" << threshold;
void read(const FileNode& fn) CV_OVERRIDE
{
- extended = (int)fn["extended"] != 0;
- upright = (int)fn["upright"] != 0;
- threshold = (float)fn["threshold"];
- octaves = (int)fn["octaves"];
- sublevels = (int)fn["sublevels"];
- diffusivity = static_cast<KAZE::DiffusivityType>((int)fn["diffusivity"]);
+ // if node is empty, keep previous value
+ if (!fn["extended"].empty())
+ extended = (int)fn["extended"] != 0;
+ if (!fn["upright"].empty())
+ upright = (int)fn["upright"] != 0;
+ if (!fn["threshold"].empty())
+ threshold = (float)fn["threshold"];
+ if (!fn["octaves"].empty())
+ octaves = (int)fn["octaves"];
+ if (!fn["sublevels"].empty())
+ sublevels = (int)fn["sublevels"];
+ if (!fn["diffusivity"].empty())
+ diffusivity = static_cast<KAZE::DiffusivityType>((int)fn["diffusivity"]);
}
bool extended;
virtual ~MSER_Impl() CV_OVERRIDE {}
+ void read( const FileNode& fn) CV_OVERRIDE
+ {
+ // if node is empty, keep previous value
+ if (!fn["delta"].empty())
+ fn["delta"] >> params.delta;
+ if (!fn["minArea"].empty())
+ fn["minArea"] >> params.minArea;
+ if (!fn["maxArea"].empty())
+ fn["maxArea"] >> params.maxArea;
+ if (!fn["maxVariation"].empty())
+ fn["maxVariation"] >> params.maxVariation;
+ if (!fn["minDiversity"].empty())
+ fn["minDiversity"] >> params.minDiversity;
+ if (!fn["maxEvolution"].empty())
+ fn["maxEvolution"] >> params.maxEvolution;
+ if (!fn["areaThreshold"].empty())
+ fn["areaThreshold"] >> params.areaThreshold;
+ if (!fn["minMargin"].empty())
+ fn["minMargin"] >> params.minMargin;
+ if (!fn["edgeBlurSize"].empty())
+ fn["edgeBlurSize"] >> params.edgeBlurSize;
+ if (!fn["pass2Only"].empty())
+ fn["pass2Only"] >> params.pass2Only;
+ }
+ void write( FileStorage& fs) const CV_OVERRIDE
+ {
+ if(fs.isOpened())
+ {
+ fs << "name" << getDefaultName();
+ fs << "delta" << params.delta;
+ fs << "minArea" << params.minArea;
+ fs << "maxArea" << params.maxArea;
+ fs << "maxVariation" << params.maxVariation;
+ fs << "minDiversity" << params.minDiversity;
+ fs << "maxEvolution" << params.maxEvolution;
+ fs << "areaThreshold" << params.areaThreshold;
+ fs << "minMargin" << params.minMargin;
+ fs << "edgeBlurSize" << params.edgeBlurSize;
+ fs << "pass2Only" << params.pass2Only;
+ }
+ }
+
void setDelta(int delta) CV_OVERRIDE { params.delta = delta; }
int getDelta() const CV_OVERRIDE { return params.delta; }
void setMaxArea(int maxArea) CV_OVERRIDE { params.maxArea = maxArea; }
int getMaxArea() const CV_OVERRIDE { return params.maxArea; }
+ void setMaxVariation(double maxVariation) CV_OVERRIDE { params.maxVariation = maxVariation; }
+ double getMaxVariation() const CV_OVERRIDE { return params.maxVariation; }
+
+ void setMinDiversity(double minDiversity) CV_OVERRIDE { params.minDiversity = minDiversity; }
+ double getMinDiversity() const CV_OVERRIDE { return params.minDiversity; }
+
+ void setMaxEvolution(int maxEvolution) CV_OVERRIDE { params.maxEvolution = maxEvolution; }
+ int getMaxEvolution() const CV_OVERRIDE { return params.maxEvolution; }
+
+ void setAreaThreshold(double areaThreshold) CV_OVERRIDE { params.areaThreshold = areaThreshold; }
+ double getAreaThreshold() const CV_OVERRIDE { return params.areaThreshold; }
+
+ void setMinMargin(double min_margin) CV_OVERRIDE { params.minMargin = min_margin; }
+ double getMinMargin() const CV_OVERRIDE { return params.minMargin; }
+
+ void setEdgeBlurSize(int edge_blur_size) CV_OVERRIDE { params.edgeBlurSize = edge_blur_size; }
+ int getEdgeBlurSize() const CV_OVERRIDE { return params.edgeBlurSize; }
+
void setPass2Only(bool f) CV_OVERRIDE { params.pass2Only = f; }
bool getPass2Only() const CV_OVERRIDE { return params.pass2Only; }
scoreType(_scoreType), patchSize(_patchSize), fastThreshold(_fastThreshold)
{}
+ void read( const FileNode& fn) CV_OVERRIDE;
+ void write( FileStorage& fs) const CV_OVERRIDE;
+
void setMaxFeatures(int maxFeatures) CV_OVERRIDE { nfeatures = maxFeatures; }
int getMaxFeatures() const CV_OVERRIDE { return nfeatures; }
int fastThreshold;
};
+void ORB_Impl::read( const FileNode& fn)
+{
+ // if node is empty, keep previous value
+ if (!fn["nfeatures"].empty())
+ fn["nfeatures"] >> nfeatures;
+ if (!fn["scaleFactor"].empty())
+ fn["scaleFactor"] >> scaleFactor;
+ if (!fn["nlevels"].empty())
+ fn["nlevels"] >> nlevels;
+ if (!fn["edgeThreshold"].empty())
+ fn["edgeThreshold"] >> edgeThreshold;
+ if (!fn["firstLevel"].empty())
+ fn["firstLevel"] >> firstLevel;
+ if (!fn["wta_k"].empty())
+ fn["wta_k"] >> wta_k;
+ if (!fn["scoreType"].empty())
+ fn["scoreType"] >> scoreType;
+ if (!fn["patchSize"].empty())
+ fn["patchSize"] >> patchSize;
+ if (!fn["fastThreshold"].empty())
+ fn["fastThreshold"] >> fastThreshold;
+}
+void ORB_Impl::write( FileStorage& fs) const
+{
+ if(fs.isOpened())
+ {
+ fs << "name" << getDefaultName();
+ fs << "nfeatures" << nfeatures;
+ fs << "scaleFactor" << scaleFactor;
+ fs << "nlevels" << nlevels;
+ fs << "edgeThreshold" << edgeThreshold;
+ fs << "firstLevel" << firstLevel;
+ fs << "wta_k" << wta_k;
+ fs << "scoreType" << scoreType;
+ fs << "patchSize" << patchSize;
+ fs << "fastThreshold" << fastThreshold;
+ }
+}
+
int ORB_Impl::descriptorSize() const
{
return kBytes;
void findScaleSpaceExtrema( const std::vector<Mat>& gauss_pyr, const std::vector<Mat>& dog_pyr,
std::vector<KeyPoint>& keypoints ) const;
+ void read( const FileNode& fn) CV_OVERRIDE;
+ void write( FileStorage& fs) const CV_OVERRIDE;
+
+ void setNFeatures(int maxFeatures) CV_OVERRIDE { nfeatures = maxFeatures; }
+ int getNFeatures() const CV_OVERRIDE { return nfeatures; }
+
+ void setNOctaveLayers(int nOctaveLayers_) CV_OVERRIDE { nOctaveLayers = nOctaveLayers_; }
+ int getNOctaveLayers() const CV_OVERRIDE { return nOctaveLayers; }
+
+ void setContrastThreshold(double contrastThreshold_) CV_OVERRIDE { contrastThreshold = contrastThreshold_; }
+ double getContrastThreshold() const CV_OVERRIDE { return contrastThreshold; }
+
+ void setEdgeThreshold(double edgeThreshold_) CV_OVERRIDE { edgeThreshold = edgeThreshold_; }
+ double getEdgeThreshold() const CV_OVERRIDE { return edgeThreshold; }
+
+ void setSigma(double sigma_) CV_OVERRIDE { sigma = sigma_; }
+ double getSigma() const CV_OVERRIDE { return sigma; }
+
protected:
CV_PROP_RW int nfeatures;
CV_PROP_RW int nOctaveLayers;
}
}
+void SIFT_Impl::read( const FileNode& fn)
+{
+ // if node is empty, keep previous value
+ if (!fn["nfeatures"].empty())
+ fn["nfeatures"] >> nfeatures;
+ if (!fn["nOctaveLayers"].empty())
+ fn["nOctaveLayers"] >> nOctaveLayers;
+ if (!fn["contrastThreshold"].empty())
+ fn["contrastThreshold"] >> contrastThreshold;
+ if (!fn["edgeThreshold"].empty())
+ fn["edgeThreshold"] >> edgeThreshold;
+ if (!fn["sigma"].empty())
+ fn["sigma"] >> sigma;
+ if (!fn["descriptorType"].empty())
+ fn["descriptorType"] >> descriptor_type;
+}
+void SIFT_Impl::write( FileStorage& fs) const
+{
+ if(fs.isOpened())
+ {
+ fs << "name" << getDefaultName();
+ fs << "nfeatures" << nfeatures;
+ fs << "nOctaveLayers" << nOctaveLayers;
+ fs << "contrastThreshold" << contrastThreshold;
+ fs << "edgeThreshold" << edgeThreshold;
+ fs << "sigma" << sigma;
+ fs << "descriptorType" << descriptor_type;
+ }
+}
+
}
if os.path.isfile(path):
if path.endswith("FeatureDetector.java"):
for prefix1 in ("", "Grid", "Pyramid", "Dynamic"):
- for prefix2 in ("FAST", "STAR", "MSER", "ORB", "SIFT", "SURF", "GFTT", "HARRIS", "SIMPLEBLOB", "DENSE"):
+ for prefix2 in ("FAST", "STAR", "MSER", "ORB", "SIFT", "SURF", "GFTT", "HARRIS", "SIMPLEBLOB", "DENSE", "AKAZE", "KAZE", "BRISK", "AGAST"):
parser.parse_file(path,prefix1+prefix2)
elif path.endswith("DescriptorExtractor.java"):
for prefix1 in ("", "Opponent"):
- for prefix2 in ("BRIEF", "ORB", "SIFT", "SURF"):
+ for prefix2 in ("BRIEF", "ORB", "SIFT", "SURF", "AKAZE", "KAZE", "BEBLID", "DAISY", "FREAK", "LUCID", "LATCH"):
parser.parse_file(path,prefix1+prefix2)
elif path.endswith("GenericDescriptorMatcher.java"):
for prefix in ("OneWay", "Fern"):
/** @brief Write a set of DetectorParameters to FileStorage
*/
- bool writeDetectorParameters(FileStorage& fs);
-
- /** @brief simplified API for language bindings
- */
- CV_WRAP bool writeDetectorParameters(const Ptr<FileStorage>& fs, const String& name = String());
+ CV_WRAP bool writeDetectorParameters(FileStorage& fs, const String& name = String());
/// minimum window size for adaptive thresholding before finding contours (default 3).
CV_PROP_RW int adaptiveThreshWinSizeMin;
/** @brief Write a set of RefineParameters to FileStorage
*/
- bool writeRefineParameters(FileStorage& fs);
-
- /** @brief simplified API for language bindings
- */
- CV_WRAP bool writeRefineParameters(const Ptr<FileStorage>& fs, const String& name = String());
+ CV_WRAP bool writeRefineParameters(FileStorage& fs, const String& name = String());
/** @brief minRepDistance minimum distance between the corners of the rejected candidate and the reprojected marker
in order to consider it as a correspondence.
/** @brief simplified API for language bindings
*/
- CV_WRAP inline void write(const Ptr<FileStorage>& fs, const String& name = String()) { Algorithm::write(fs, name); }
+ CV_WRAP inline void write(FileStorage& fs, const String& name) { Algorithm::write(fs, name); }
/** @brief Reads algorithm parameters from a file storage
*/
/** @brief Write a dictionary to FileStorage, format is the same as in readDictionary().
*/
- void writeDictionary(FileStorage& fs);
-
- /** @brief simplified API for language bindings
- */
- CV_WRAP void writeDictionary(Ptr<FileStorage>& fs, const String& name = String());
+ CV_WRAP void writeDictionary(FileStorage& fs, const String& name = String());
/** @brief Given a matrix of bits. Returns whether if marker is identified or not.
*
using namespace std;
-static inline bool readWrite(DetectorParameters ¶ms, const Ptr<FileNode>& readNode,
- const Ptr<FileStorage>& writeStorage = nullptr) {
- CV_Assert(!readNode.empty() || !writeStorage.empty());
+static inline bool readWrite(DetectorParameters ¶ms, const FileNode* readNode,
+ FileStorage* writeStorage = nullptr)
+{
+ CV_Assert(readNode || writeStorage);
bool check = false;
- check |= readWriteParameter("adaptiveThreshWinSizeMin", params.adaptiveThreshWinSizeMin, *readNode, *writeStorage);
- check |= readWriteParameter("adaptiveThreshWinSizeMax", params.adaptiveThreshWinSizeMax, *readNode, *writeStorage);
- check |= readWriteParameter("adaptiveThreshWinSizeStep", params.adaptiveThreshWinSizeStep, *readNode, *writeStorage);
- check |= readWriteParameter("adaptiveThreshConstant", params.adaptiveThreshConstant, *readNode, *writeStorage);
- check |= readWriteParameter("minMarkerPerimeterRate", params.minMarkerPerimeterRate, *readNode, *writeStorage);
- check |= readWriteParameter("maxMarkerPerimeterRate", params.maxMarkerPerimeterRate, *readNode, *writeStorage);
+ check |= readWriteParameter("adaptiveThreshWinSizeMin", params.adaptiveThreshWinSizeMin, readNode, writeStorage);
+ check |= readWriteParameter("adaptiveThreshWinSizeMax", params.adaptiveThreshWinSizeMax, readNode, writeStorage);
+ check |= readWriteParameter("adaptiveThreshWinSizeStep", params.adaptiveThreshWinSizeStep, readNode, writeStorage);
+ check |= readWriteParameter("adaptiveThreshConstant", params.adaptiveThreshConstant, readNode, writeStorage);
+ check |= readWriteParameter("minMarkerPerimeterRate", params.minMarkerPerimeterRate, readNode, writeStorage);
+ check |= readWriteParameter("maxMarkerPerimeterRate", params.maxMarkerPerimeterRate, readNode, writeStorage);
check |= readWriteParameter("polygonalApproxAccuracyRate", params.polygonalApproxAccuracyRate,
- *readNode, *writeStorage);
- check |= readWriteParameter("minCornerDistanceRate", params.minCornerDistanceRate, *readNode, *writeStorage);
- check |= readWriteParameter("minDistanceToBorder", params.minDistanceToBorder, *readNode, *writeStorage);
- check |= readWriteParameter("minMarkerDistanceRate", params.minMarkerDistanceRate, *readNode, *writeStorage);
- check |= readWriteParameter("cornerRefinementMethod", params.cornerRefinementMethod, *readNode, *writeStorage);
- check |= readWriteParameter("cornerRefinementWinSize", params.cornerRefinementWinSize, *readNode, *writeStorage);
+ readNode, writeStorage);
+ check |= readWriteParameter("minCornerDistanceRate", params.minCornerDistanceRate, readNode, writeStorage);
+ check |= readWriteParameter("minDistanceToBorder", params.minDistanceToBorder, readNode, writeStorage);
+ check |= readWriteParameter("minMarkerDistanceRate", params.minMarkerDistanceRate, readNode, writeStorage);
+ check |= readWriteParameter("cornerRefinementMethod", params.cornerRefinementMethod, readNode, writeStorage);
+ check |= readWriteParameter("cornerRefinementWinSize", params.cornerRefinementWinSize, readNode, writeStorage);
check |= readWriteParameter("cornerRefinementMaxIterations", params.cornerRefinementMaxIterations,
- *readNode, *writeStorage);
+ readNode, writeStorage);
check |= readWriteParameter("cornerRefinementMinAccuracy", params.cornerRefinementMinAccuracy,
- *readNode, *writeStorage);
- check |= readWriteParameter("markerBorderBits", params.markerBorderBits, *readNode, *writeStorage);
+ readNode, writeStorage);
+ check |= readWriteParameter("markerBorderBits", params.markerBorderBits, readNode, writeStorage);
check |= readWriteParameter("perspectiveRemovePixelPerCell", params.perspectiveRemovePixelPerCell,
- *readNode, *writeStorage);
+ readNode, writeStorage);
check |= readWriteParameter("perspectiveRemoveIgnoredMarginPerCell", params.perspectiveRemoveIgnoredMarginPerCell,
- *readNode, *writeStorage);
+ readNode, writeStorage);
check |= readWriteParameter("maxErroneousBitsInBorderRate", params.maxErroneousBitsInBorderRate,
- *readNode, *writeStorage);
- check |= readWriteParameter("minOtsuStdDev", params.minOtsuStdDev, *readNode, *writeStorage);
- check |= readWriteParameter("errorCorrectionRate", params.errorCorrectionRate, *readNode, *writeStorage);
+ readNode, writeStorage);
+ check |= readWriteParameter("minOtsuStdDev", params.minOtsuStdDev, readNode, writeStorage);
+ check |= readWriteParameter("errorCorrectionRate", params.errorCorrectionRate, readNode, writeStorage);
// new aruco 3 functionality
- check |= readWriteParameter("useAruco3Detection", params.useAruco3Detection, *readNode, *writeStorage);
- check |= readWriteParameter("minSideLengthCanonicalImg", params.minSideLengthCanonicalImg, *readNode, *writeStorage);
+ check |= readWriteParameter("useAruco3Detection", params.useAruco3Detection, readNode, writeStorage);
+ check |= readWriteParameter("minSideLengthCanonicalImg", params.minSideLengthCanonicalImg, readNode, writeStorage);
check |= readWriteParameter("minMarkerLengthRatioOriginalImg", params.minMarkerLengthRatioOriginalImg,
- *readNode, *writeStorage);
+ readNode, writeStorage);
return check;
}
-bool DetectorParameters::readDetectorParameters(const FileNode& fn) {
- if(fn.empty())
+bool DetectorParameters::readDetectorParameters(const FileNode& fn)
+{
+ if (fn.empty())
return false;
- Ptr<FileNode> pfn = makePtr<FileNode>(fn);
- return readWrite(*this, pfn);
+ return readWrite(*this, &fn);
}
-bool DetectorParameters::writeDetectorParameters(const Ptr<FileStorage>& fs, const String& name) {
- if (fs.empty())
- return false;
- if(name.empty())
- return writeDetectorParameters(*fs);
- *fs << name << "{";
- bool res = writeDetectorParameters(*fs);
- *fs << "}";
+bool DetectorParameters::writeDetectorParameters(FileStorage& fs, const String& name)
+{
+ CV_Assert(fs.isOpened());
+ if (!name.empty())
+ fs << name << "{";
+ bool res = readWrite(*this, nullptr, &fs);
+ if (!name.empty())
+ fs << "}";
return res;
}
-bool DetectorParameters::writeDetectorParameters(FileStorage &fs) {
- if (!fs.isOpened())
- return false;
- return readWrite(*this, nullptr, makePtr<FileStorage>(fs));
-}
-
-static inline bool readWrite(RefineParameters& refineParameters, const Ptr<FileNode>& readNode,
- const Ptr<FileStorage>& writeStorage = nullptr) {
- CV_Assert(!readNode.empty() || !writeStorage.empty());
+static inline bool readWrite(RefineParameters& refineParameters, const FileNode* readNode,
+ FileStorage* writeStorage = nullptr)
+{
+ CV_Assert(readNode || writeStorage);
bool check = false;
- check |= readWriteParameter("minRepDistance", refineParameters.minRepDistance, *readNode, *writeStorage);
- check |= readWriteParameter("errorCorrectionRate", refineParameters.errorCorrectionRate, *readNode, *writeStorage);
- check |= readWriteParameter("checkAllOrders", refineParameters.checkAllOrders, *readNode, *writeStorage);
+ check |= readWriteParameter("minRepDistance", refineParameters.minRepDistance, readNode, writeStorage);
+ check |= readWriteParameter("errorCorrectionRate", refineParameters.errorCorrectionRate, readNode, writeStorage);
+ check |= readWriteParameter("checkAllOrders", refineParameters.checkAllOrders, readNode, writeStorage);
return check;
}
minRepDistance(_minRepDistance), errorCorrectionRate(_errorCorrectionRate),
checkAllOrders(_checkAllOrders){}
-bool RefineParameters::readRefineParameters(const FileNode &fn) {
- if(fn.empty())
+bool RefineParameters::readRefineParameters(const FileNode &fn)
+{
+ if (fn.empty())
return false;
- Ptr<FileNode> pfn = makePtr<FileNode>(fn);
- return readWrite(*this, pfn);
+ return readWrite(*this, &fn);
}
-bool RefineParameters::writeRefineParameters(FileStorage &fs) {
- if(!fs.isOpened())
- return false;
- return readWrite(*this, nullptr, makePtr<FileStorage>(fs));
-}
-
-bool RefineParameters::writeRefineParameters(const Ptr<FileStorage>& fs, const String& name) {
- if(fs.empty())
- return false;
- if(name.empty())
- return writeRefineParameters(*fs);
- *fs << name << "{";
- bool res = writeRefineParameters(*fs);
- *fs << "}";
+bool RefineParameters::writeRefineParameters(FileStorage& fs, const String& name)
+{
+ CV_Assert(fs.isOpened());
+ if (!name.empty())
+ fs << name << "{";
+ bool res = readWrite(*this, nullptr, &fs);
+ if (!name.empty())
+ fs << "}";
return res;
}
}
}
-void ArucoDetector::write(FileStorage &fs) const {
- Ptr<FileStorage> pfs = makePtr<FileStorage>(fs);
- arucoDetectorImpl->dictionary.writeDictionary(pfs);
+void ArucoDetector::write(FileStorage &fs) const
+{
+ arucoDetectorImpl->dictionary.writeDictionary(fs);
arucoDetectorImpl->detectorParams.writeDetectorParameters(fs);
arucoDetectorImpl->refineParams.writeRefineParameters(fs);
}
return true;
}
-void Dictionary::writeDictionary(FileStorage &fs) {
+void Dictionary::writeDictionary(FileStorage& fs, const String &name)
+{
+ CV_Assert(fs.isOpened());
+
+ if (!name.empty())
+ fs << name << "{";
+
fs << "nmarkers" << bytesList.rows;
fs << "markersize" << markerSize;
fs << "maxCorrectionBits" << maxCorrectionBits;
marker.push_back(bitMarker.at<uint8_t>(j) + '0');
fs << markerName << marker;
}
-}
-void Dictionary::writeDictionary(Ptr<FileStorage>& fs, const String &name) {
- if(name.empty())
- return writeDictionary(*fs);
- *fs << name << "{";
- writeDictionary(*fs);
- *fs << "}";
+ if (!name.empty())
+ fs << "}";
}
}
template<typename T>
-inline bool readWriteParameter(const std::string& name, T& parameter, const FileNode& readNode, FileStorage& writeStorage) {
- if (!readNode.empty())
- return readParameter(name, parameter, readNode);
- writeStorage << name << parameter;
+inline bool readWriteParameter(const std::string& name, T& parameter, const FileNode* readNode, FileStorage* writeStorage)
+{
+ if (readNode)
+ return readParameter(name, parameter, *readNode);
+ CV_Assert(writeStorage);
+ *writeStorage << name << parameter;
return true;
}
--- /dev/null
+// This file is part of OpenCV project.
+// It is subject to the license terms in the LICENSE file found in the top-level directory
+// of this distribution and at http://opencv.org/license.html.
+
+#include "perf_precomp.hpp"
+
+namespace opencv_test
+{
+namespace
+{
+struct ExposureSeq
+{
+ std::vector<Mat> images;
+ std::vector<float> times;
+};
+
+ExposureSeq loadExposureSeq(const std::string& list_filename)
+{
+ std::ifstream list_file(list_filename);
+ EXPECT_TRUE(list_file.is_open());
+ string name;
+ float val;
+ const String path(list_filename.substr(0, list_filename.find_last_of("\\/") + 1));
+ ExposureSeq seq;
+ while (list_file >> name >> val)
+ {
+ Mat img = imread(path + name);
+ EXPECT_FALSE(img.empty()) << "Could not load input image " << path + name;
+ seq.images.push_back(img);
+ seq.times.push_back(1 / val);
+ }
+ list_file.close();
+ return seq;
+}
+
+PERF_TEST(HDR, Mertens)
+{
+ const ExposureSeq seq = loadExposureSeq(getDataPath("cv/hdr/exposures/list.txt"));
+ Ptr<MergeMertens> merge = createMergeMertens();
+ Mat result(seq.images.front().size(), seq.images.front().type());
+ TEST_CYCLE() merge->process(seq.images, result);
+ SANITY_CHECK_NOTHING();
+}
+
+PERF_TEST(HDR, Debevec)
+{
+ const ExposureSeq seq = loadExposureSeq(getDataPath("cv/hdr/exposures/list.txt"));
+ Ptr<MergeDebevec> merge = createMergeDebevec();
+ Mat result(seq.images.front().size(), seq.images.front().type());
+ TEST_CYCLE() merge->process(seq.images, result, seq.times);
+ SANITY_CHECK_NOTHING();
+}
+
+PERF_TEST(HDR, Robertson)
+{
+ const ExposureSeq seq = loadExposureSeq(getDataPath("cv/hdr/exposures/list.txt"));
+ Ptr<MergeRobertson> merge = createMergeRobertson();
+ Mat result(seq.images.front().size(), seq.images.front().type());
+ TEST_CYCLE() merge->process(seq.images, result, seq.times);
+ SANITY_CHECK_NOTHING();
+}
+
+} // namespace
+} // namespace opencv_test
std::vector<Mat> weights(images.size());
Mat weight_sum = Mat::zeros(size, CV_32F);
+ Mutex weight_sum_mutex;
- for(size_t i = 0; i < images.size(); i++) {
- Mat img, gray, contrast, saturation, wellexp;
- std::vector<Mat> splitted(channels);
-
- images[i].convertTo(img, CV_32F, 1.0f/255.0f);
- if(channels == 3) {
- cvtColor(img, gray, COLOR_RGB2GRAY);
- } else {
- img.copyTo(gray);
- }
- split(img, splitted);
+ parallel_for_(Range(0, static_cast<int>(images.size())), [&](const Range& range) {
+ for(int i = range.start; i < range.end; i++) {
+ Mat img, gray, contrast, saturation, wellexp;
+ std::vector<Mat> splitted(channels);
- Laplacian(gray, contrast, CV_32F);
- contrast = abs(contrast);
+ images[i].convertTo(img, CV_32F, 1.0f/255.0f);
+ if(channels == 3) {
+ cvtColor(img, gray, COLOR_RGB2GRAY);
+ } else {
+ img.copyTo(gray);
+ }
+ images[i] = img;
+ split(img, splitted);
- Mat mean = Mat::zeros(size, CV_32F);
- for(int c = 0; c < channels; c++) {
- mean += splitted[c];
- }
- mean /= channels;
+ Laplacian(gray, contrast, CV_32F);
+ contrast = abs(contrast);
- saturation = Mat::zeros(size, CV_32F);
- for(int c = 0; c < channels; c++) {
- Mat deviation = splitted[c] - mean;
- pow(deviation, 2.0f, deviation);
- saturation += deviation;
- }
- sqrt(saturation, saturation);
+ Mat mean = Mat::zeros(size, CV_32F);
+ for(int c = 0; c < channels; c++) {
+ mean += splitted[c];
+ }
+ mean /= channels;
- wellexp = Mat::ones(size, CV_32F);
- for(int c = 0; c < channels; c++) {
- Mat expo = splitted[c] - 0.5f;
- pow(expo, 2.0f, expo);
- expo = -expo / 0.08f;
- exp(expo, expo);
- wellexp = wellexp.mul(expo);
- }
+ saturation = Mat::zeros(size, CV_32F);
+ for(int c = 0; c < channels; c++) {
+ Mat deviation = splitted[c] - mean;
+ pow(deviation, 2.0f, deviation);
+ saturation += deviation;
+ }
+ sqrt(saturation, saturation);
+
+ wellexp = Mat::ones(size, CV_32F);
+ for(int c = 0; c < channels; c++) {
+ Mat expo = splitted[c] - 0.5f;
+ pow(expo, 2.0f, expo);
+ expo = -expo / 0.08f;
+ exp(expo, expo);
+ wellexp = wellexp.mul(expo);
+ }
- pow(contrast, wcon, contrast);
- pow(saturation, wsat, saturation);
- pow(wellexp, wexp, wellexp);
+ pow(contrast, wcon, contrast);
+ pow(saturation, wsat, saturation);
+ pow(wellexp, wexp, wellexp);
+
+ weights[i] = contrast;
+ if(channels == 3) {
+ weights[i] = weights[i].mul(saturation);
+ }
+ weights[i] = weights[i].mul(wellexp) + 1e-12f;
- weights[i] = contrast;
- if(channels == 3) {
- weights[i] = weights[i].mul(saturation);
+ AutoLock lock(weight_sum_mutex);
+ weight_sum += weights[i];
}
- weights[i] = weights[i].mul(wellexp) + 1e-12f;
- weight_sum += weights[i];
- }
+ });
+
int maxlevel = static_cast<int>(logf(static_cast<float>(min(size.width, size.height))) / logf(2.0f));
std::vector<Mat> res_pyr(maxlevel + 1);
+ std::vector<Mutex> res_pyr_mutexes(maxlevel + 1);
- for(size_t i = 0; i < images.size(); i++) {
- weights[i] /= weight_sum;
- Mat img;
- images[i].convertTo(img, CV_32F, 1.0f/255.0f);
-
- std::vector<Mat> img_pyr, weight_pyr;
- buildPyramid(img, img_pyr, maxlevel);
- buildPyramid(weights[i], weight_pyr, maxlevel);
-
- for(int lvl = 0; lvl < maxlevel; lvl++) {
- Mat up;
- pyrUp(img_pyr[lvl + 1], up, img_pyr[lvl].size());
- img_pyr[lvl] -= up;
- }
- for(int lvl = 0; lvl <= maxlevel; lvl++) {
- std::vector<Mat> splitted(channels);
- split(img_pyr[lvl], splitted);
- for(int c = 0; c < channels; c++) {
- splitted[c] = splitted[c].mul(weight_pyr[lvl]);
+ parallel_for_(Range(0, static_cast<int>(images.size())), [&](const Range& range) {
+ for(int i = range.start; i < range.end; i++) {
+ weights[i] /= weight_sum;
+
+ std::vector<Mat> img_pyr, weight_pyr;
+ buildPyramid(images[i], img_pyr, maxlevel);
+ buildPyramid(weights[i], weight_pyr, maxlevel);
+
+ for(int lvl = 0; lvl < maxlevel; lvl++) {
+ Mat up;
+ pyrUp(img_pyr[lvl + 1], up, img_pyr[lvl].size());
+ img_pyr[lvl] -= up;
}
- merge(splitted, img_pyr[lvl]);
- if(res_pyr[lvl].empty()) {
- res_pyr[lvl] = img_pyr[lvl];
- } else {
- res_pyr[lvl] += img_pyr[lvl];
+ for(int lvl = 0; lvl <= maxlevel; lvl++) {
+ std::vector<Mat> splitted(channels);
+ split(img_pyr[lvl], splitted);
+ for(int c = 0; c < channels; c++) {
+ splitted[c] = splitted[c].mul(weight_pyr[lvl]);
+ }
+ merge(splitted, img_pyr[lvl]);
+
+ AutoLock lock(res_pyr_mutexes[lvl]);
+ if(res_pyr[lvl].empty()) {
+ res_pyr[lvl] = img_pyr[lvl];
+ } else {
+ res_pyr[lvl] += img_pyr[lvl];
+ }
}
}
- }
+ });
for(int lvl = maxlevel; lvl > 0; lvl--) {
Mat up;
pyrUp(res_pyr[lvl], up, res_pyr[lvl - 1].size());
images, _ = load_exposure_seq(os.path.join(test_data_path, 'exposures'))
+ # As we want to test mat vs. umat here, we temporarily set only one worker-thread to achieve
+ # deterministic summations inside mertens' parallelized process.
+ num_threads = cv.getNumThreads()
+ cv.setNumThreads(1)
+
merge = cv.createMergeMertens()
mat_result = merge.process(images)
umat_images = [cv.UMat(img) for img in images]
umat_result = merge.process(umat_images)
+ cv.setNumThreads(num_threads)
+
self.assertTrue(np.allclose(umat_result.get(), mat_result))
objdetect = {'': ['groupRectangles'],
'HOGDescriptor': ['load', 'HOGDescriptor', 'getDefaultPeopleDetector', 'getDaimlerPeopleDetector', 'setSVMDetector', 'detectMultiScale'],
'CascadeClassifier': ['load', 'detectMultiScale2', 'CascadeClassifier', 'detectMultiScale3', 'empty', 'detectMultiScale'],
- 'QRCodeDetector': ['QRCodeDetector', 'decode', 'decodeCurved', 'detect', 'detectAndDecode', 'detectMulti', 'setEpsX', 'setEpsY']}
+ 'QRCodeDetector': ['QRCodeDetector', 'decode', 'decodeCurved', 'detect', 'detectAndDecode', 'detectMulti', 'setEpsX', 'setEpsY'],
+ 'ArucoDetector': ['getPredefinedDictionary', 'detectMarkers', 'refineDetectedMarkers', 'getDictionary', 'stetDictionary', 'getDetectorParameters', 'setDetectorParameters', 'getRefineParameters', 'setRefineParameters'],
+ 'GridBoard': ['create','generateImage', 'getGridSize', 'getMarkerLength', 'getMarkerSeparation'],
+ 'CharucoBoard': ['create', 'generateImage', 'getChessboardCorners', 'getNearestMarkerCorners', 'checkCharucoCornersCollinear']
+}
video = {
'': [
'getColorAdaptation', 'setColorAdaptation']
}
-aruco = {'': ['detectMarkers', 'drawDetectedMarkers', 'drawAxis', 'estimatePoseSingleMarkers', 'estimatePoseBoard', 'estimatePoseCharucoBoard', 'interpolateCornersCharuco', 'drawDetectedCornersCharuco'],
- 'aruco_Dictionary': ['get', 'drawMarker'],
- 'aruco_Board': ['create'],
- 'aruco_GridBoard': ['create', 'draw'],
- 'aruco_CharucoBoard': ['create', 'draw'],
- 'aruco_DetectorParameters': ['create']
- }
-
calib3d = {
'': [
'findHomography',
],
}
-white_list = makeWhiteList([core, imgproc, objdetect, video, dnn, features2d, photo, aruco, calib3d])
+white_list = makeWhiteList([core, imgproc, objdetect, video, dnn, features2d, photo, calib3d])
# namespace_prefix_override['dnn'] = '' # compatibility stuff (enabled by default)
# namespace_prefix_override['aruco'] = '' # compatibility stuff (enabled by default)
fun:start_thread
fun:clone
}
+
+{
+ avcodec57-ffv1-16bit-ubuntu18
+ Memcheck:Cond
+ ...
+ obj:/usr/lib/x86_64-linux-gnu/libavcodec.so.57.107.100
+ fun:avcodec_default_execute
+ obj:/usr/lib/x86_64-linux-gnu/libavcodec.so.57.107.100
+ fun:avcodec_encode_video2
+ obj:/usr/lib/x86_64-linux-gnu/libavcodec.so.57.107.100
+ fun:avcodec_send_frame
+ ...
+}