transform.
@param image 8-bit, single-channel binary source image. The image may be modified by the function.
-@param lines Output vector of lines. Each line is represented by a two-element vector
-\f$(\rho, \theta)\f$ . \f$\rho\f$ is the distance from the coordinate origin \f$(0,0)\f$ (top-left corner of
+@param lines Output vector of lines. Each line is represented by a 2 or 3 element vector
+\f$(\rho, \theta)\f$ or \f$(\rho, \theta, \votes)\f$ . \f$\rho\f$ is the distance from the coordinate origin \f$(0,0)\f$ (top-left corner of
the image). \f$\theta\f$ is the line rotation angle in radians (
\f$0 \sim \textrm{vertical line}, \pi/2 \sim \textrm{horizontal line}\f$ ).
+\f$\votes\f$ is the value of accumulator.
@param rho Distance resolution of the accumulator in pixels.
@param theta Angle resolution of the accumulator in radians.
@param threshold Accumulator threshold parameter. Only those lines are returned that get enough
search, and find the correct radius using an additional procedure.
@param image 8-bit, single-channel, grayscale input image.
-@param circles Output vector of found circles. Each vector is encoded as a 3-element
-floating-point vector \f$(x, y, radius)\f$ .
+@param circles Output vector of found circles. Each vector is encoded as 3 or 4 element
+floating-point vector \f$(x, y, radius)\f$ or \f$(x, y, radius, votes)\f$ .
@param method Detection method, see #HoughModes. Currently, the only implemented method is #HOUGH_GRADIENT
@param dp Inverse ratio of the accumulator resolution to the image resolution. For example, if
dp=1 , the accumulator has the same resolution as the input image. If dp=2 , the accumulator has
SANITY_CHECK_NOTHING();
}
+PERF_TEST(PerfHoughCircles4f, Basic)
+{
+ string filename = getDataPath("cv/imgproc/stuff.jpg");
+ const double dp = 1.0;
+ double minDist = 20;
+ double edgeThreshold = 20;
+ double accumThreshold = 30;
+ int minRadius = 20;
+ int maxRadius = 200;
+
+ Mat img = imread(filename, IMREAD_GRAYSCALE);
+ ASSERT_FALSE(img.empty()) << "Unable to load source image " << filename;
+
+ GaussianBlur(img, img, Size(9, 9), 2, 2);
+
+ vector<Vec4f> circles;
+ declare.in(img);
+
+ TEST_CYCLE()
+ {
+ HoughCircles(img, circles, CV_HOUGH_GRADIENT, dp, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
+ }
+
+ SANITY_CHECK_NOTHING();
+}
+
} // namespace
SANITY_CHECK_NOTHING();
}
+PERF_TEST_P(Image_RhoStep_ThetaStep_Threshold, HoughLines3f,
+ testing::Combine(
+ testing::Values( "cv/shared/pic5.png", "stitching/a1.png" ),
+ testing::Values( 1, 10 ),
+ testing::Values( 0.01, 0.1 ),
+ testing::Values( 0.5, 1.1 )
+ )
+ )
+{
+ string filename = getDataPath(get<0>(GetParam()));
+ double rhoStep = get<1>(GetParam());
+ double thetaStep = get<2>(GetParam());
+ double threshold_ratio = get<3>(GetParam());
+
+ Mat image = imread(filename, IMREAD_GRAYSCALE);
+ if (image.empty())
+ FAIL() << "Unable to load source image" << filename;
+
+ Canny(image, image, 32, 128);
+
+ // add some syntetic lines:
+ line(image, Point(0, 0), Point(image.cols, image.rows), Scalar::all(255), 3);
+ line(image, Point(image.cols, 0), Point(image.cols/2, image.rows), Scalar::all(255), 3);
+
+ vector<Vec3f> lines;
+ declare.time(60);
+
+ int hough_threshold = (int)(std::min(image.cols, image.rows) * threshold_ratio);
+
+ TEST_CYCLE() HoughLines(image, lines, rhoStep, thetaStep, hough_threshold);
+
+ printf("%dx%d: %d lines\n", image.cols, image.rows, (int)lines.size());
+
+ if (threshold_ratio < 1.0)
+ {
+ EXPECT_GE(lines.size(), 2u);
+ }
+
+ EXPECT_LT(lines.size(), 3000u);
+
+ SANITY_CHECK_NOTHING();
+}
+
} // namespace
Functions return the actual number of found lines.
*/
static void
-HoughLinesStandard( const Mat& img, float rho, float theta,
- int threshold, std::vector<Vec2f>& lines, int linesMax,
+HoughLinesStandard( InputArray src, OutputArray lines, int type,
+ float rho, float theta,
+ int threshold, int linesMax,
double min_theta, double max_theta )
{
+ CV_CheckType(type, type == CV_32FC2 || type == CV_32FC3, "Internal error");
+
+ Mat img = src.getMat();
+
int i, j;
float irho = 1 / rho;
CV_Assert( img.type() == CV_8UC1 );
+ CV_Assert( linesMax > 0 );
const uchar* image = img.ptr();
int step = (int)img.step;
int width = img.cols;
int height = img.rows;
- if (max_theta < min_theta ) {
- CV_Error( CV_StsBadArg, "max_theta must be greater than min_theta" );
- }
+ int max_rho = width + height;
+ int min_rho = -max_rho;
+
+ CV_CheckGE(max_theta, min_theta, "max_theta must be greater than min_theta");
+
int numangle = cvRound((max_theta - min_theta) / theta);
- int numrho = cvRound(((width + height) * 2 + 1) / rho);
+ int numrho = cvRound(((max_rho - min_rho) + 1) / rho);
#if defined HAVE_IPP && IPP_VERSION_X100 >= 810 && !IPP_DISABLE_HOUGH
- CV_IPP_CHECK()
+ if (type == CV_32FC2 && CV_IPP_CHECK_COND)
{
IppiSize srcSize = { width, height };
IppPointPolar delta = { rho, theta };
- IppPointPolar dstRoi[2] = {{(Ipp32f) -(width + height), (Ipp32f) min_theta},{(Ipp32f) (width + height), (Ipp32f) max_theta}};
+ IppPointPolar dstRoi[2] = {{(Ipp32f) min_rho, (Ipp32f) min_theta},{(Ipp32f) max_rho, (Ipp32f) max_theta}};
int bufferSize;
int nz = countNonZero(img);
int ipp_linesMax = std::min(linesMax, nz*numangle/threshold);
int linesCount = 0;
- lines.resize(ipp_linesMax);
+ std::vector<Vec2f> _lines(ipp_linesMax);
IppStatus ok = ippiHoughLineGetSize_8u_C1R(srcSize, delta, ipp_linesMax, &bufferSize);
Ipp8u* buffer = ippsMalloc_8u_L(bufferSize);
- if (ok >= 0) {ok = CV_INSTRUMENT_FUN_IPP(ippiHoughLine_Region_8u32f_C1R, image, step, srcSize, (IppPointPolar*) &lines[0], dstRoi, ipp_linesMax, &linesCount, delta, threshold, buffer);};
+ if (ok >= 0) {ok = CV_INSTRUMENT_FUN_IPP(ippiHoughLine_Region_8u32f_C1R, image, step, srcSize, (IppPointPolar*) &_lines[0], dstRoi, ipp_linesMax, &linesCount, delta, threshold, buffer);};
ippsFree(buffer);
if (ok >= 0)
{
- lines.resize(linesCount);
+ lines.create(linesCount, 1, CV_32FC2);
+ Mat(linesCount, 1, CV_32FC2, &_lines[0]).copyTo(lines);
CV_IMPL_ADD(CV_IMPL_IPP);
return;
}
- lines.clear();
setIppErrorStatus();
}
#endif
// stage 4. store the first min(total,linesMax) lines to the output buffer
linesMax = std::min(linesMax, (int)_sort_buf.size());
double scale = 1./(numrho+2);
+
+ lines.create(linesMax, 1, type);
+ Mat _lines = lines.getMat();
for( i = 0; i < linesMax; i++ )
{
LinePolar line;
int r = idx - (n+1)*(numrho+2) - 1;
line.rho = (r - (numrho - 1)*0.5f) * rho;
line.angle = static_cast<float>(min_theta) + n * theta;
- lines.push_back(Vec2f(line.rho, line.angle));
+ if (type == CV_32FC2)
+ {
+ _lines.at<Vec2f>(i) = Vec2f(line.rho, line.angle);
+ }
+ else
+ {
+ CV_DbgAssert(type == CV_32FC3);
+ _lines.at<Vec3f>(i) = Vec3f(line.rho, line.angle, (float)accum[idx]);
+ }
}
}
static void
-HoughLinesSDiv( const Mat& img,
+HoughLinesSDiv( InputArray image, OutputArray lines, int type,
float rho, float theta, int threshold,
- int srn, int stn,
- std::vector<Vec2f>& lines, int linesMax,
+ int srn, int stn, int linesMax,
double min_theta, double max_theta )
{
+ CV_CheckType(type, type == CV_32FC2 || type == CV_32FC3, "Internal error");
+
#define _POINT(row, column)\
(image_src[(row)*step+(column)])
+ Mat img = image.getMat();
int index, i;
int ri, ti, ti1, ti0;
int row, col;
if( count * 100 > rn * tn )
{
- HoughLinesStandard( img, rho, theta, threshold, lines, linesMax, min_theta, max_theta );
+ HoughLinesStandard( image, lines, type, rho, theta, threshold, linesMax, min_theta, max_theta );
return;
}
}
}
+ lines.create((int)lst.size(), 1, type);
+ Mat _lines = lines.getMat();
for( size_t idx = 0; idx < lst.size(); idx++ )
{
if( lst[idx].rho < 0 )
continue;
- lines.push_back(Vec2f(lst[idx].rho, lst[idx].theta));
+ if (type == CV_32FC2)
+ {
+ _lines.at<Vec2f>((int)idx) = Vec2f(lst[idx].rho, lst[idx].theta);
+ }
+ else
+ {
+ CV_DbgAssert(type == CV_32FC3);
+ _lines.at<Vec3f>((int)idx) = Vec3f(lst[idx].rho, lst[idx].theta, (float)lst[idx].value);
+ }
}
}
#endif /* HAVE_OPENCL */
-void HoughLines( InputArray _image, OutputArray _lines,
+void HoughLines( InputArray _image, OutputArray lines,
double rho, double theta, int threshold,
double srn, double stn, double min_theta, double max_theta )
{
CV_INSTRUMENT_REGION()
- CV_OCL_RUN(srn == 0 && stn == 0 && _image.isUMat() && _lines.isUMat(),
- ocl_HoughLines(_image, _lines, rho, theta, threshold, min_theta, max_theta));
+ int type = CV_32FC2;
+ if (lines.fixedType())
+ {
+ type = lines.type();
+ CV_CheckType(type, type == CV_32FC2 || type == CV_32FC3, "Wrong type of output lines");
+ }
- Mat image = _image.getMat();
- std::vector<Vec2f> lines;
+ CV_OCL_RUN(srn == 0 && stn == 0 && _image.isUMat() && lines.isUMat() && type == CV_32FC2,
+ ocl_HoughLines(_image, lines, rho, theta, threshold, min_theta, max_theta));
if( srn == 0 && stn == 0 )
- HoughLinesStandard(image, (float)rho, (float)theta, threshold, lines, INT_MAX, min_theta, max_theta );
+ HoughLinesStandard(_image, lines, type, (float)rho, (float)theta, threshold, INT_MAX, min_theta, max_theta );
else
- HoughLinesSDiv(image, (float)rho, (float)theta, threshold, cvRound(srn), cvRound(stn), lines, INT_MAX, min_theta, max_theta);
-
- Mat(lines).copyTo(_lines);
+ HoughLinesSDiv(_image, lines, type, (float)rho, (float)theta, threshold, cvRound(srn), cvRound(stn), INT_MAX, min_theta, max_theta);
}
return false;
}
-inline Vec3f GetCircle(const EstimatedCircle& est)
+static inline Vec3f GetCircle(const EstimatedCircle& est)
{
return est.c;
}
+static inline Vec4f GetCircle4f(const EstimatedCircle& est)
+{
+ return Vec4f(est.c[0], est.c[1], est.c[2], (float)est.accum);
+}
+
class NZPointList : public std::vector<Point>
{
private:
Mutex& _lock;
};
-static bool CheckDistance(const std::vector<Vec3f> &circles, size_t endIdx, const Vec3f& circle, float minDist2)
+template<typename T>
+static bool CheckDistance(const std::vector<T> &circles, size_t endIdx, const T& circle, float minDist2)
{
bool goodPoint = true;
for (uint j = 0; j < endIdx; ++j)
{
- Vec3f pt = circles[j];
+ T pt = circles[j];
float distX = circle[0] - pt[0], distY = circle[1] - pt[1];
if (distX * distX + distY * distY < minDist2)
{
}
}
-static void RemoveOverlaps(std::vector<Vec3f>& circles, float minDist)
+static void GetCircleCenters(const std::vector<int> ¢ers, std::vector<Vec4f> &circles, int acols, float minDist, float dr)
+{
+ size_t centerCnt = centers.size();
+ float minDist2 = minDist * minDist;
+ for (size_t i = 0; i < centerCnt; ++i)
+ {
+ int center = centers[i];
+ int y = center / acols;
+ int x = center - y * acols;
+ Vec4f circle = Vec4f((x + 0.5f) * dr, (y + 0.5f) * dr, 0, (float)center);
+
+ bool goodPoint = CheckDistance(circles, circles.size(), circle, minDist2);
+ if (goodPoint)
+ circles.push_back(circle);
+ }
+}
+
+template<typename T>
+static void RemoveOverlaps(std::vector<T>& circles, float minDist)
{
float minDist2 = minDist * minDist;
size_t endIdx = 1;
for (size_t i = 1; i < circles.size(); ++i)
{
- Vec3f circle = circles[i];
+ T circle = circles[i];
if (CheckDistance(circles, endIdx, circle, minDist2))
{
circles[endIdx] = circle;
circles.resize(endIdx);
}
+static void CreateCircles(const std::vector<EstimatedCircle>& circlesEst, std::vector<Vec3f>& circles)
+{
+ std::transform(circlesEst.begin(), circlesEst.end(), std::back_inserter(circles), GetCircle);
+}
+
+static void CreateCircles(const std::vector<EstimatedCircle>& circlesEst, std::vector<Vec4f>& circles)
+{
+ std::transform(circlesEst.begin(), circlesEst.end(), std::back_inserter(circles), GetCircle4f);
+}
+
template<class NZPoints>
class HoughCircleEstimateRadiusInvoker : public ParallelLoopBody
{
return nzCount;
}
-static void HoughCirclesGradient(InputArray _image, OutputArray _circles, float dp, float minDist,
+template <typename CircleType>
+static void HoughCirclesGradient(InputArray _image, OutputArray _circles,
+ float dp, float minDist,
int minRadius, int maxRadius, int cannyThreshold,
int accThreshold, int maxCircles, int kernelSize, bool centersOnly)
{
CV_Assert(kernelSize == -1 || kernelSize == 3 || kernelSize == 5 || kernelSize == 7);
+
dp = max(dp, 1.f);
float idp = 1.f/dp;
std::sort(centers.begin(), centers.end(), hough_cmp_gt(accum.ptr<int>()));
- std::vector<Vec3f> circles;
+ std::vector<CircleType> circles;
circles.reserve(256);
if (centersOnly)
{
// Sort by accumulator value
std::sort(circlesEst.begin(), circlesEst.end(), cmpAccum);
- std::transform(circlesEst.begin(), circlesEst.end(), std::back_inserter(circles), GetCircle);
+
+ // Create Circles
+ CreateCircles(circlesEst, circles);
RemoveOverlaps(circles, minDist);
}
- if(circles.size() > 0)
+ if (circles.size() > 0)
{
int numCircles = std::min(maxCircles, int(circles.size()));
- _circles.create(1, numCircles, CV_32FC3);
- Mat(1, numCircles, CV_32FC3, &circles[0]).copyTo(_circles.getMat());
+ Mat(1, numCircles, cv::traits::Type<CircleType>::value, &circles[0]).copyTo(_circles);
return;
}
}
{
CV_INSTRUMENT_REGION()
+ int type = CV_32FC3;
+ if( _circles.fixedType() )
+ {
+ type = _circles.type();
+ CV_CheckType(type, type == CV_32FC3 || type == CV_32FC4, "Wrong type of output circles");
+ }
+
CV_Assert(!_image.empty() && _image.type() == CV_8UC1 && (_image.isMat() || _image.isUMat()));
CV_Assert(_circles.isMat() || _circles.isVector());
switch( method )
{
case CV_HOUGH_GRADIENT:
- HoughCirclesGradient(_image, _circles, (float)dp, (float)minDist,
- minRadius, maxRadius, cannyThresh,
- accThresh, maxCircles, kernelSize, centersOnly);
+ if (type == CV_32FC3)
+ HoughCirclesGradient<Vec3f>(_image, _circles, (float)dp, (float)minDist,
+ minRadius, maxRadius, cannyThresh,
+ accThresh, maxCircles, kernelSize, centersOnly);
+ else if (type == CV_32FC4)
+ HoughCirclesGradient<Vec4f>(_image, _circles, (float)dp, (float)minDist,
+ minRadius, maxRadius, cannyThresh,
+ accThresh, maxCircles, kernelSize, centersOnly);
+ else
+ CV_Error(Error::StsError, "Internal error");
break;
default:
CV_Error( Error::StsBadArg, "Unrecognized method id. Actually only CV_HOUGH_GRADIENT is supported." );
switch( method )
{
case CV_HOUGH_STANDARD:
- HoughLinesStandard( image, (float)rho,
- (float)theta, threshold, l2, linesMax, min_theta, max_theta );
+ HoughLinesStandard( image, l2, CV_32FC2, (float)rho,
+ (float)theta, threshold, linesMax, min_theta, max_theta );
break;
case CV_HOUGH_MULTI_SCALE:
- HoughLinesSDiv( image, (float)rho, (float)theta,
- threshold, iparam1, iparam2, l2, linesMax, min_theta, max_theta );
+ HoughLinesSDiv( image, l2, CV_32FC2, (float)rho, (float)theta,
+ threshold, iparam1, iparam2, linesMax, min_theta, max_theta );
break;
case CV_HOUGH_PROBABILISTIC:
HoughLinesProbabilistic( image, (float)rho, (float)theta,
#define DEBUG_IMAGES 0
#endif
+//#define GENERATE_DATA // generate data in debug mode via CPU code path (without IPP / OpenCL and other accelerators)
+
using namespace cv;
using namespace std;
{
}
- void run_test()
+ template <typename CircleType>
+ void run_test(const char* xml_name)
{
string test_case_name = getTestCaseName(picture_name, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
string filename = cvtest::TS::ptr()->get_data_path() + picture_name;
GaussianBlur(src, src, Size(9, 9), 2, 2);
- vector<Vec3f> circles;
+ vector<CircleType> circles;
const double dp = 1.0;
HoughCircles(src, circles, CV_HOUGH_GRADIENT, dp, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
highlightCircles(filename, circles, imgProc + test_case_name + ".png");
#endif
- string xml = imgProc + "HoughCircles.xml";
- FileStorage fs(xml, FileStorage::READ);
- FileNode node = fs[test_case_name];
- if (node.empty())
+ string xml = imgProc + xml_name;
+#ifdef GENERATE_DATA
{
- fs.release();
- fs.open(xml, FileStorage::APPEND);
+ FileStorage fs(xml, FileStorage::READ);
+ ASSERT_TRUE(!fs.isOpened() || fs[test_case_name].empty());
+ }
+ {
+ FileStorage fs(xml, FileStorage::APPEND);
EXPECT_TRUE(fs.isOpened()) << "Cannot open sanity data file: " << xml;
fs << test_case_name << circles;
- fs.release();
- fs.open(xml, FileStorage::READ);
- EXPECT_TRUE(fs.isOpened()) << "Cannot open sanity data file: " << xml;
}
-
- vector<Vec3f> exp_circles;
- read(fs[test_case_name], exp_circles, vector<Vec3f>());
+#else
+ FileStorage fs(xml, FileStorage::READ);
+ FileNode node = fs[test_case_name];
+ ASSERT_FALSE(node.empty()) << "Missing test data: " << test_case_name << std::endl << "XML: " << xml;
+ vector<CircleType> exp_circles;
+ read(fs[test_case_name], exp_circles, vector<CircleType>());
fs.release();
-
EXPECT_EQ(exp_circles.size(), circles.size());
+#endif
}
};
TEST_P(HoughCirclesTestFixture, regression)
{
- run_test();
+ run_test<Vec3f>("HoughCircles.xml");
+}
+
+TEST_P(HoughCirclesTestFixture, regression4f)
+{
+ run_test<Vec4f>("HoughCircles4f.xml");
}
INSTANTIATE_TEST_CASE_P(ImgProc, HoughCirclesTestFixture, testing::Combine(
GaussianBlur(src, src, Size(9, 9), 2, 2);
vector<Vec3f> circles;
+ vector<Vec4f> circles4f;
HoughCircles(src, circles, CV_HOUGH_GRADIENT, dp, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
+ HoughCircles(src, circles4f, CV_HOUGH_GRADIENT, dp, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
#if DEBUG_IMAGES
string imgProc = string(cvtest::TS::ptr()->get_data_path()) + "imgproc/";
GaussianBlur(src, src, Size(9, 9), 2, 2);
vector<Vec3f> circles;
+ vector<Vec4f> circles4f;
HoughCircles(src, circles, CV_HOUGH_GRADIENT, dp, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
+ HoughCircles(src, circles4f, CV_HOUGH_GRADIENT, dp, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
#if DEBUG_IMAGES
string imgProc = string(cvtest::TS::ptr()->get_data_path()) + "imgproc/";
for (size_t i = 0; i < circles.size(); ++i)
{
EXPECT_EQ(circles[i][2], 0.0f) << "Did not ask for radius";
+ EXPECT_EQ(circles[i][0], circles4f[i][0]);
+ EXPECT_EQ(circles[i][1], circles4f[i][1]);
+ EXPECT_EQ(circles[i][2], circles4f[i][2]);
}
}
EXPECT_FALSE(src.empty()) << "Invalid test image: " << filename;
vector<Vec3f> circles;
+ vector<Vec4f> circles4f;
HoughCircles(src, circles, CV_HOUGH_GRADIENT, dp, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
+ HoughCircles(src, circles4f, CV_HOUGH_GRADIENT, dp, minDist, edgeThreshold, accumThreshold, minRadius, maxRadius);
#if DEBUG_IMAGES
string imgProc = string(cvtest::TS::ptr()->get_data_path()) + "imgproc/";
#endif
EXPECT_GT(circles.size(), size_t(3000)) << "Should find a lot of circles";
+ EXPECT_EQ(circles.size(), circles4f.size());
}
}} // namespace
#include "test_precomp.hpp"
+//#define GENERATE_DATA // generate data in debug mode via CPU code path (without IPP / OpenCL and other accelerators)
+
namespace opencv_test { namespace {
template<typename T>
float theta_eps;
float rho_eps;
SimilarWith<T>(T val, float e, float r_e): value(val), theta_eps(e), rho_eps(r_e) { };
- bool operator()(T other);
+ bool operator()(const T& other);
};
template<>
-bool SimilarWith<Vec2f>::operator()(Vec2f other)
+bool SimilarWith<Vec2f>::operator()(const Vec2f& other)
+{
+ return std::abs(other[0] - value[0]) < rho_eps && std::abs(other[1] - value[1]) < theta_eps;
+}
+
+template<>
+bool SimilarWith<Vec3f>::operator()(const Vec3f& other)
{
return std::abs(other[0] - value[0]) < rho_eps && std::abs(other[1] - value[1]) < theta_eps;
}
template<>
-bool SimilarWith<Vec4i>::operator()(Vec4i other)
+bool SimilarWith<Vec4i>::operator()(const Vec4i& other)
{
return cv::norm(value, other) < theta_eps;
}
template <typename T>
-int countMatIntersection(Mat expect, Mat actual, float eps, float rho_eps)
+int countMatIntersection(const Mat& expect, const Mat& actual, float eps, float rho_eps)
{
int count = 0;
if (!expect.empty() && !actual.empty())
{
- for (MatIterator_<T> it=expect.begin<T>(); it!=expect.end<T>(); it++)
+ for (MatConstIterator_<T> it=expect.begin<T>(); it!=expect.end<T>(); it++)
{
- MatIterator_<T> f = std::find_if(actual.begin<T>(), actual.end<T>(), SimilarWith<T>(*it, eps, rho_eps));
+ MatConstIterator_<T> f = std::find_if(actual.begin<T>(), actual.end<T>(), SimilarWith<T>(*it, eps, rho_eps));
if (f != actual.end<T>())
count++;
}
public:
enum {STANDART = 0, PROBABILISTIC};
protected:
- void run_test(int type);
+ template<typename LinesType, typename LineType>
+ void run_test(int type, const char* xml_name);
string picture_name;
double rhoStep;
}
};
-void BaseHoughLineTest::run_test(int type)
+template<typename LinesType, typename LineType>
+void BaseHoughLineTest::run_test(int type, const char* xml_name)
{
string filename = cvtest::TS::ptr()->get_data_path() + picture_name;
Mat src = imread(filename, IMREAD_GRAYSCALE);
- EXPECT_FALSE(src.empty()) << "Invalid test image: " << filename;
+ ASSERT_FALSE(src.empty()) << "Invalid test image: " << filename;
- string xml;
- if (type == STANDART)
- xml = string(cvtest::TS::ptr()->get_data_path()) + "imgproc/HoughLines.xml";
- else if (type == PROBABILISTIC)
- xml = string(cvtest::TS::ptr()->get_data_path()) + "imgproc/HoughLinesP.xml";
+ string xml = string(cvtest::TS::ptr()->get_data_path()) + "imgproc/" + xml_name;
Mat dst;
Canny(src, dst, 100, 150, 3);
- EXPECT_FALSE(dst.empty()) << "Failed Canny edge detector";
+ ASSERT_FALSE(dst.empty()) << "Failed Canny edge detector";
- Mat lines;
+ LinesType lines;
if (type == STANDART)
HoughLines(dst, lines, rhoStep, thetaStep, threshold, 0, 0);
else if (type == PROBABILISTIC)
HoughLinesP(dst, lines, rhoStep, thetaStep, threshold, minLineLength, maxGap);
String test_case_name = format("lines_%s_%.0f_%.2f_%d_%d_%d", picture_name.c_str(), rhoStep, thetaStep,
- threshold, minLineLength, maxGap);
+ threshold, minLineLength, maxGap);
test_case_name = getTestCaseName(test_case_name);
- FileStorage fs(xml, FileStorage::READ);
- FileNode node = fs[test_case_name];
- if (node.empty())
+#ifdef GENERATE_DATA
{
- fs.release();
- fs.open(xml, FileStorage::APPEND);
- EXPECT_TRUE(fs.isOpened()) << "Cannot open sanity data file: " << xml;
- fs << test_case_name << lines;
- fs.release();
- fs.open(xml, FileStorage::READ);
+ FileStorage fs(xml, FileStorage::READ);
+ ASSERT_TRUE(!fs.isOpened() || fs[test_case_name].empty());
+ }
+ {
+ FileStorage fs(xml, FileStorage::APPEND);
EXPECT_TRUE(fs.isOpened()) << "Cannot open sanity data file: " << xml;
+ fs << test_case_name << Mat(lines);
}
+#else
+ FileStorage fs(xml, FileStorage::READ);
+ FileNode node = fs[test_case_name];
+ ASSERT_FALSE(node.empty()) << "Missing test data: " << test_case_name << std::endl << "XML: " << xml;
- Mat exp_lines;
- read( fs[test_case_name], exp_lines, Mat() );
+ Mat exp_lines_;
+ read(fs[test_case_name], exp_lines_, Mat());
fs.release();
+ LinesType exp_lines;
+ exp_lines_.copyTo(exp_lines);
int count = -1;
if (type == STANDART)
- count = countMatIntersection<Vec2f>(exp_lines, lines, (float) thetaStep + FLT_EPSILON, (float) rhoStep + FLT_EPSILON);
+ count = countMatIntersection<LineType>(Mat(exp_lines), Mat(lines), (float) thetaStep + FLT_EPSILON, (float) rhoStep + FLT_EPSILON);
else if (type == PROBABILISTIC)
- count = countMatIntersection<Vec4i>(exp_lines, lines, 1e-4f, 0.f);
+ count = countMatIntersection<LineType>(Mat(exp_lines), Mat(lines), 1e-4f, 0.f);
#if defined HAVE_IPP && IPP_VERSION_X100 >= 810 && !IPP_DISABLE_HOUGH
- EXPECT_GE( count, (int) (exp_lines.total() * 0.8) );
+ EXPECT_LE(std::abs((double)count - Mat(exp_lines).total()), Mat(exp_lines).total() * 0.25)
+ << "count=" << count << " expected=" << Mat(exp_lines).total();
#else
- EXPECT_EQ( count, (int) exp_lines.total());
+ EXPECT_EQ(count, (int)Mat(exp_lines).total());
#endif
+#endif // GENERATE_DATA
}
void HoughLinesPointSetTest::run_test(void)
TEST_P(StandartHoughLinesTest, regression)
{
- run_test(STANDART);
+ run_test<Mat, Vec2f>(STANDART, "HoughLines.xml");
}
TEST_P(ProbabilisticHoughLinesTest, regression)
{
- run_test(PROBABILISTIC);
+ run_test<Mat, Vec4i>(PROBABILISTIC, "HoughLinesP.xml");
+}
+
+TEST_P(StandartHoughLinesTest, regression_Vec2f)
+{
+ run_test<std::vector<Vec2f>, Vec2f>(STANDART, "HoughLines2f.xml");
+}
+
+TEST_P(StandartHoughLinesTest, regression_Vec3f)
+{
+ run_test<std::vector<Vec3f>, Vec3f>(STANDART, "HoughLines3f.xml");
}
TEST_P(HoughLinesPointSetTest, regression)