Visual Servoing Platform  version 3.6.1 under development (2024-10-18)
tutorial-canny.cpp
/*
* ViSP, open source Visual Servoing Platform software.
* Copyright (C) 2005 - 2023 by Inria. All rights reserved.
*
* This software is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
* See the file LICENSE.txt at the root directory of this source
* distribution for additional information about the GNU GPL.
*
* For using ViSP with software that can not be combined with the GNU
* GPL, please contact Inria about acquiring a ViSP Professional
* Edition License.
*
* See https://visp.inria.fr for more information.
*
* This software was developed at:
* Inria Rennes - Bretagne Atlantique
* Campus Universitaire de Beaulieu
* 35042 Rennes Cedex
* France
*
* If you have questions regarding the use of this file, please contact
* Inria at visp@inria.fr
*
* This file is provided AS IS with NO WARRANTY OF ANY KIND, INCLUDING THE
* WARRANTY OF DESIGN, MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
*/
#include <visp3/core/vpConfig.h>
#include <visp3/core/vpCannyEdgeDetection.h>
#include <visp3/core/vpImageFilter.h>
#include <visp3/io/vpImageIo.h>
#ifdef HAVE_OPENCV_IMGPROC
#include <opencv2/imgproc/imgproc.hpp>
#endif
#include "drawingHelpers.h"
#ifdef ENABLE_VISP_NAMESPACE
using namespace VISP_NAMESPACE_NAME;
#endif
template <class T>
void computeMeanMaxStdev(const vpImage<T> &I, float &mean, float &max, float &stdev)
{
max = std::numeric_limits<float>::epsilon();
mean = 0.;
stdev = 0.;
unsigned int nbRows = I.getRows();
unsigned int nbCols = I.getCols();
float scale = 1.f / (static_cast<float>(nbRows) * static_cast<float>(nbCols));
for (unsigned int r = 0; r < nbRows; r++) {
for (unsigned int c = 0; c < nbCols; c++) {
mean += I[r][c];
max = std::max<float>(max, static_cast<float>(I[r][c]));
}
}
mean *= scale;
for (unsigned int r = 0; r < nbRows; r++) {
for (unsigned int c = 0; c < nbCols; c++) {
stdev += (I[r][c] - mean) * (I[r][c] - mean);
}
}
stdev *= scale;
stdev = std::sqrt(stdev);
}
void setGradientOutsideClass(const vpImage<unsigned char> &I, const int &gaussianKernelSize, const float &gaussianStdev,
vpCannyEdgeDetection &cannyDetector, const unsigned int apertureSize,
)
{
// Computing the gradients
vpImage<float> dIx, dIy;
vpImageFilter::computePartialDerivatives(I, dIx, dIy, true, true, true, gaussianKernelSize, gaussianStdev,
apertureSize, filteringType);
// Set the gradients of the vpCannyEdgeDetection
cannyDetector.setGradients(dIx, dIy);
// Display the gradients
float mean, max, stdev;
computeMeanMaxStdev(dIx, mean, max, stdev);
#if (VISP_CXX_STANDARD > VISP_CXX_STANDARD_98)
std::string title = "Gradient along the horizontal axis. Mean = " + std::to_string(mean)
+ "+/-" + std::to_string(stdev) + " Max = " + std::to_string(max);
#else
std::string title;
{
std::stringstream ss;
ss << "Gradient along the horizontal axis. Mean = " << mean<< "+/-" << stdev<< " Max = " << max;
title = ss.str();
}
#endif
vpImageConvert::convert(dIx, dIx_uchar);
drawingHelpers::display(dIx_uchar, title);
computeMeanMaxStdev(dIy, mean, max, stdev);
#if (VISP_CXX_STANDARD > VISP_CXX_STANDARD_98)
title = "Gradient along the horizontal axis. Mean = " + std::to_string(mean)
+ "+/-" + std::to_string(stdev) + " Max = " + std::to_string(max);
#else
{
std::stringstream ss;
ss << "Gradient along the horizontal axis. Mean = " << mean<< "+/-" << stdev<< " Max = " << max;
title = ss.str();
}
#endif
vpImageConvert::convert(dIy, dIy_uchar);
drawingHelpers::display(dIy_uchar, title);
}
void usage(const std::string &softName, int gaussianKernelSize, float gaussianStdev, float lowerThresh, float upperThresh,
int apertureSize, vpImageFilter::vpCannyFilteringAndGradientType filteringType,
float lowerThreshRatio, float upperThreshRatio, vpImageFilter::vpCannyBackendType backend)
{
std::cout << "NAME" << std::endl;
std::cout << softName << ": software to test the vpCannyEdgeComputation class and vpImageFilter::canny method" << std::endl;
std::cout << "SYNOPSIS" << std::endl;
std::cout << "\t" << softName
<< " [-i, --image <pathToImg>]"
<< " [-g, --gradient <kernelSize stdev>]"
<< " [-t, --thresh <lowerThresh upperThresh>]"
<< " [-a, --aperture <apertureSize>]"
<< " [-f, --filter <filterName>]"
<< " [-r, --ratio <lowerThreshRatio upperThreshRatio>]"
<< " [-b, --backend <backendName>]"
<< " [-h, --help]" << std::endl
<< std::endl;
std::cout << "DESCRIPTION" << std::endl;
std::cout << "\t-i, --image <pathToImg>" << std::endl
<< "\t\tPermits to load an image on which will be tested the vpCanny class." << std::endl
<< "\t\tWhen empty uses a simulated image." << std::endl
<< std::endl;
std::cout << "\t-g, --gradient <kernelSize stdev>" << std::endl
<< "\t\tPermits to compute the gradients of the image outside the vpCanny class." << std::endl
<< "\t\tFirst parameter is the size of the Gaussian kernel used to compute the gradients." << std::endl
<< "\t\tSecond parameter is the standard deviation of the Gaussian kernel used to compute the gradients." << std::endl
<< "\t\tDefault: " << gaussianKernelSize << " " << gaussianStdev << std::endl
<< std::endl;
std::cout << "\t-t, --thresh <lowerThresh upperThresh>" << std::endl
<< "\t\tPermits to set the lower and upper thresholds of the vpCanny class." << std::endl
<< "\t\tFirst parameter is the lower threshold." << std::endl
<< "\t\tSecond parameter is the upper threshold." << std::endl
<< "\t\tWhen set to -1 thresholds are computed automatically." << std::endl
<< "\t\tDefault: " << lowerThresh << " " << upperThresh << std::endl
<< std::endl;
std::cout << "\t-a, --aperture <apertureSize>" << std::endl
<< "\t\tPermits to set the size of the gradient filter kernel." << std::endl
<< "\t\tParameter must be odd and positive." << std::endl
<< "\t\tDefault: " << apertureSize << std::endl
<< std::endl;
std::cout << "\t-f, --filter <filterName>" << std::endl
<< "\t\tPermits to choose the type of filter to apply to compute the gradient." << std::endl
<< "\t\tAvailable values: " << vpImageFilter::vpGetCannyFiltAndGradTypes("<", " | ", ">") << std::endl
<< "\t\tDefault: " << vpImageFilter::vpCannyFiltAndGradTypeToStr(filteringType) << std::endl
<< std::endl;
std::cout << "\t-r, --ratio <lowerThreshRatio upperThreshRatio>" << std::endl
<< "\t\tPermits to set the lower and upper thresholds ratio of the vpCanny class." << std::endl
<< "\t\tFirst parameter is the lower threshold ratio." << std::endl
<< "\t\tSecond parameter is the upper threshold ratio." << std::endl
<< "\t\tDefault: " << lowerThreshRatio << " " << upperThreshRatio << std::endl
<< std::endl;
std::cout << "\t-b, --backend <backendName>" << std::endl
<< "\t\tPermits to use the vpImageFilter::canny method for comparison." << std::endl
<< "\t\tAvailable values: " << vpImageFilter::vpCannyBackendTypeList("<", " | ", ">") << std::endl
<< "\t\tDefault: " << vpImageFilter::vpCannyBackendTypeToString(backend) << std::endl
<< std::endl;
std::cout << "\t-h, --help" << std::endl
<< "\t\tPermits to display the different arguments this software handles." << std::endl
<< std::endl;
}
int main(int argc, const char *argv[])
{
std::string opt_img;
bool opt_gradientOutsideClass = false;
bool opt_useVpImageFilterCanny = false;
int opt_gaussianKernelSize = 3;
int opt_apertureSize = 3;
float opt_gaussianStdev = 1.;
float opt_lowerThresh = -1.;
float opt_upperThresh = -1.;
float opt_lowerThreshRatio = 0.6f;
float opt_upperThreshRatio = 0.8f;
for (int i = 1; i < argc; i++) {
std::string argv_str = std::string(argv[i]);
if ((argv_str == "-i" || argv_str == "--image") && i + 1 < argc) {
opt_img = std::string(argv[i + 1]);
i++;
}
else if ((argv_str == "-g" || argv_str == "--gradient") && i + 2 < argc) {
opt_gradientOutsideClass = true;
opt_gaussianKernelSize = atoi(argv[i + 1]);
opt_gaussianStdev = static_cast<float>(atof(argv[i + 2]));
i += 2;
}
else if ((argv_str == "-t" || argv_str == "--thresh") && i + 2 < argc) {
opt_lowerThresh = static_cast<float>(atof(argv[i + 1]));
opt_upperThresh = static_cast<float>(atof(argv[i + 2]));
i += 2;
}
else if ((argv_str == "-a" || argv_str == "--aperture") && i + 1 < argc) {
opt_apertureSize = std::atoi(argv[i + 1]);
i++;
}
else if ((argv_str == "-f" || argv_str == "--filter") && i + 1 < argc) {
opt_filteringType = vpImageFilter::vpCannyFiltAndGradTypeFromStr(std::string(argv[i + 1]));
i++;
}
else if ((argv_str == "-r" || argv_str == "--ratio") && i + 2 < argc) {
opt_lowerThreshRatio = static_cast<float>(std::atof(argv[i + 1]));
opt_upperThreshRatio = static_cast<float>(std::atof(argv[i + 2]));
i += 2;
}
else if ((argv_str == "-b" || argv_str == "--backend") && i + 1 < argc) {
opt_useVpImageFilterCanny = true;
opt_backend = vpImageFilter::vpCannyBackendTypeFromString(std::string(argv[i+1]));
i++;
}
else if (argv_str == "-h" || argv_str == "--help") {
usage(std::string(argv[0]), opt_gaussianKernelSize, opt_gaussianStdev, opt_lowerThresh, opt_upperThresh,
opt_apertureSize, opt_filteringType, opt_lowerThreshRatio, opt_upperThreshRatio, opt_backend);
return EXIT_SUCCESS;
}
else {
std::cerr << "Argument \"" << argv_str << "\" is unknown." << std::endl;
return EXIT_FAILURE;
}
}
std::string configAsTxt("Canny Configuration:\n");
configAsTxt += "\tFiltering + gradient operators = " + vpImageFilter::vpCannyFiltAndGradTypeToStr(opt_filteringType) + "\n";
#if (VISP_CXX_STANDARD > VISP_CXX_STANDARD_98)
configAsTxt += "\tGaussian filter kernel size = " + std::to_string(opt_gaussianKernelSize) + "\n";
configAsTxt += "\tGaussian filter standard deviation = " + std::to_string(opt_gaussianStdev) + "\n";
configAsTxt += "\tGradient filter kernel size = " + std::to_string(opt_apertureSize) + "\n";
configAsTxt += "\tCanny edge filter thresholds = [" + std::to_string(opt_lowerThresh) + " ; " + std::to_string(opt_upperThresh) + "]\n";
configAsTxt += "\tCanny edge filter thresholds ratio (for auto-thresholding) = [" + std::to_string(opt_lowerThreshRatio) + " ; " + std::to_string(opt_upperThreshRatio) + "]\n";
#else
{
std::stringstream ss;
ss << "\tGaussian filter kernel size = " << opt_gaussianKernelSize << "\n";
ss << "\tGaussian filter standard deviation = " << opt_gaussianStdev << "\n";
ss << "\tGradient filter kernel size = " << opt_apertureSize << "\n";
ss << "\tCanny edge filter thresholds = [" << opt_lowerThresh << " ; " << opt_upperThresh << "]\n";
ss << "\tCanny edge filter thresholds ratio (for auto-thresholding) = [" << opt_lowerThreshRatio << " ; " << opt_upperThreshRatio << "]\n";
configAsTxt += ss.str();
}
#endif
std::cout << configAsTxt << std::endl;
vpCannyEdgeDetection cannyDetector(opt_gaussianKernelSize, opt_gaussianStdev, opt_apertureSize,
opt_lowerThresh, opt_upperThresh, opt_lowerThreshRatio, opt_upperThreshRatio,
opt_filteringType);
vpImage<unsigned char> I_canny_input, I_canny_visp, dIx_uchar, dIy_uchar, I_canny_imgFilter;
if (!opt_img.empty()) {
// Detection on the user image
vpImageIo::read(I_canny_input, opt_img);
}
else {
// Detection on a fake image of a square
I_canny_input.resize(500, 500, 0);
for (unsigned int r = 150; r < 350; r++) {
for (unsigned int c = 150; c < 350; c++) {
I_canny_input[r][c] = 125;
}
}
}
// Initialization of the displays
I_canny_visp = I_canny_imgFilter = dIx_uchar = dIy_uchar = I_canny_input;
vpImage<unsigned char> *p_dIx = nullptr, *p_dIy = nullptr, *p_IcannyImgFilter = nullptr;
if (opt_gradientOutsideClass) {
p_dIx = &dIx_uchar;
p_dIy = &dIy_uchar;
}
if (opt_useVpImageFilterCanny) {
p_IcannyImgFilter = &I_canny_imgFilter;
}
drawingHelpers::init(I_canny_input, I_canny_visp, p_dIx, p_dIy, p_IcannyImgFilter);
// Computing the gradient outside the vpCannyEdgeDetection class if asked
if (opt_gradientOutsideClass) {
setGradientOutsideClass(I_canny_input, opt_gaussianKernelSize, opt_gaussianStdev, cannyDetector, opt_apertureSize,
opt_filteringType, dIx_uchar, dIy_uchar);
}
I_canny_visp = cannyDetector.detect(I_canny_input);
float mean, max, stdev;
computeMeanMaxStdev(I_canny_input, mean, max, stdev);
#if (VISP_CXX_STANDARD > VISP_CXX_STANDARD_98)
std::string title("Input of the Canny edge detector. Mean = " + std::to_string(mean) + "+/-" + std::to_string(stdev) + " Max = " + std::to_string(max));
#else
std::string title;
{
std::stringstream ss;
ss << "Input of the Canny edge detector. Mean = " << mean << "+/-" << stdev << " Max = " << max;
title = ss.str();
}
#endif
drawingHelpers::display(I_canny_input, title);
drawingHelpers::display(I_canny_visp, "Canny results on image " + opt_img);
if (opt_useVpImageFilterCanny) {
float cannyThresh = opt_upperThresh;
float lowerThresh(opt_lowerThresh);
vpImageFilter::canny(I_canny_input, I_canny_imgFilter, opt_gaussianKernelSize, lowerThresh, cannyThresh,
opt_apertureSize, opt_gaussianStdev, opt_lowerThreshRatio, opt_upperThreshRatio, true,
opt_backend, opt_filteringType);
drawingHelpers::display(I_canny_imgFilter, "Canny results with \"" + vpImageFilter::vpCannyBackendTypeToString(opt_backend) + "\" backend");
}
drawingHelpers::waitForClick(I_canny_input, true);
return EXIT_SUCCESS;
}
Class that implements the Canny's edge detector. It is possible to use a boolean mask to ignore some ...
vpImage< unsigned char > detect(const vpImage< vpRGBa > &I_color)
Detect the edges in an image. Convert the color image into a gray-scale image.
void setGradients(const vpImage< float > &dIx, const vpImage< float > &dIy)
Set the Gradients of the image that will be processed.
static void convert(const vpImage< unsigned char > &src, vpImage< vpRGBa > &dest)
static std::string vpCannyBackendTypeToString(const vpCannyBackendType &type)
Cast a vpImageFilter::vpCannyBackendTypeToString into a string, to know its name.
static void canny(const vpImage< unsigned char > &I, vpImage< unsigned char > &Ic, const unsigned int &gaussianFilterSize, const float &thresholdCanny, const unsigned int &apertureSobel)
static std::string vpCannyBackendTypeList(const std::string &pref="<", const std::string &sep=" , ", const std::string &suf=">")
Get the list of available vpCannyBackendType.
static std::string vpCannyFiltAndGradTypeToStr(const vpCannyFilteringAndGradientType &type)
Cast a vpImageFilter::vpCannyFilteringAndGradientType into a string, to know its name.
vpCannyFilteringAndGradientType
Canny filter and gradient operators to apply on the image before the edge detection stage.
Definition: vpImageFilter.h:90
@ CANNY_GBLUR_SOBEL_FILTERING
Apply Gaussian blur + Sobel operator on the input image.
Definition: vpImageFilter.h:91
vpCannyBackendType
Canny filter backends for the edge detection operations.
Definition: vpImageFilter.h:75
@ CANNY_VISP_BACKEND
Use ViSP.
Definition: vpImageFilter.h:77
static vpCannyFilteringAndGradientType vpCannyFiltAndGradTypeFromStr(const std::string &name)
Cast a string into a vpImageFilter::vpCannyFilteringAndGradientType.
static vpCannyBackendType vpCannyBackendTypeFromString(const std::string &name)
Cast a string into a vpImageFilter::vpCannyBackendTypeToString.
static std::string vpGetCannyFiltAndGradTypes(const std::string &pref="<", const std::string &sep=" , ", const std::string &suf=">")
Get the list of available vpCannyFilteringAndGradientType.
static void computePartialDerivatives(const cv::Mat &cv_I, cv::Mat &cv_dIx, cv::Mat &cv_dIy, const bool &computeDx=true, const bool &computeDy=true, const bool &normalize=true, const unsigned int &gaussianKernelSize=5, const float &gaussianStdev=2.f, const unsigned int &apertureGradient=3, const vpCannyFilteringAndGradientType &filteringType=CANNY_GBLUR_SOBEL_FILTERING)
Compute the partial derivatives (i.e. horizontal and vertical gradients) of the input image.
static void read(vpImage< unsigned char > &I, const std::string &filename, int backend=IO_DEFAULT_BACKEND)
Definition: vpImageIo.cpp:147
Definition of the vpImage class member functions.
Definition: vpImage.h:131
void resize(unsigned int h, unsigned int w)
resize the image : Image initialization
Definition: vpImage.h:542
unsigned int getCols() const
Definition: vpImage.h:171
unsigned int getRows() const
Definition: vpImage.h:212