Subversion Repositories OpenCV2-Cookbook

/*------------------------------------------------------------------------------------------*\
   This file contains material supporting chapter 8 of the cookbook:  
   Computer Vision Programming using the OpenCV Library.
   by Robert Laganiere, Packt Publishing, 2011.

   This program is free software; permission is hereby granted to use, copy, modify,
   and distribute this source code, or portions thereof, for any purpose, without fee,
   subject to the restriction that the copyright notice may not be removed
   or altered from any source or altered source distribution.
   The software is released on an as-is basis and without any warranties of any kind.
   In particular, the software is not guaranteed to be fault-tolerant or free from failure.
   The author disclaims all warranties with regard to this software, any use,
   and any consequent failure, is purely the responsibility of the user.
 
   Copyright (C) 2010-2011 Robert Laganiere, www.laganiere.name
\*------------------------------------------------------------------------------------------*/

#if !defined HARRISD
#define HARRISD

#include <vector>
#include <opencv2/core/core.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/features2d/features2d.hpp>

class HarrisDetector {

  private:

          // 32-bit float image of corner strength
          cv::Mat cornerStrength;
          // 32-bit float image of thresholded corners
          cv::Mat cornerTh;
          // image of local maxima (internal)
          cv::Mat localMax;
          // size of neighbourhood for derivatives smoothing
          int neighbourhood;
          // aperture for gradient computation
          int aperture;
          // Harris parameter
          double k;
          // maximum strength for threshold computation
          double maxStrength;
          // calculated threshold (internal)
          double threshold;
          // size of neighbourhood for non-max suppression
          int nonMaxSize;
          // kernel for non-max suppression
          cv::Mat kernel;

  public:

          HarrisDetector() : neighbourhood(3), aperture(3), k(0.1), maxStrength(0.0), threshold(0.01), nonMaxSize(3) {
         
                  setLocalMaxWindowSize(nonMaxSize);
          }

          // Create kernel used in non-maxima suppression
          void setLocalMaxWindowSize(int size) {

                  nonMaxSize= size;
                  kernel.create(nonMaxSize,nonMaxSize,CV_8U);
          }

          // Compute Harris corners
          void detect(const cv::Mat& image) {
       
                  // Harris computation
                  cv::cornerHarris(image,cornerStrength,
                             neighbourhood,// neighborhood size
                                         aperture,     // aperture size
                                         k);           // Harris parameter
       
                  // internal threshold computation
                  double minStrength; // not used
                  cv::minMaxLoc(cornerStrength,&minStrength,&maxStrength);

                  // local maxima detection
                  cv::Mat dilated;  // temporary image
                  cv::dilate(cornerStrength,dilated,cv::Mat());
                  cv::compare(cornerStrength,dilated,localMax,cv::CMP_EQ);
          }

          // Get the corner map from the computed Harris values
          cv::Mat getCornerMap(double qualityLevel) {

                  cv::Mat cornerMap;

                  // thresholding the corner strength
                  threshold= qualityLevel*maxStrength;
                  cv::threshold(cornerStrength,cornerTh,threshold,255,cv::THRESH_BINARY);

                  // convert to 8-bit image
                  cornerTh.convertTo(cornerMap,CV_8U);
       
                  // non-maxima suppression
                  cv::bitwise_and(cornerMap,localMax,cornerMap);

                  return cornerMap;
          }

          // Get the feature points vector from the computed Harris values
          void getCorners(std::vector<cv::Point> &points, double qualityLevel) {

                  // Get the corner map
                  cv::Mat cornerMap= getCornerMap(qualityLevel);
                  // Get the corners
                  getCorners(points, cornerMap);
          }

          // Get the feature points vector from the computed corner map
          void getCorners(std::vector<cv::Point> &points, const cv::Mat& cornerMap) {
                         
                  // Iterate over the pixels to obtain all feature points
                  for( int y = 0; y < cornerMap.rows; y++ ) {
   
                          const uchar* rowPtr = cornerMap.ptr<uchar>(y);
   
                          for( int x = 0; x < cornerMap.cols; x++ ) {

                                  // if it is a feature point
                                  if (rowPtr[x]) {

                                          points.push_back(cv::Point(x,y));
                                  }
                          }
                  }
          }

          // Draw circles at feature point locations on an image
          void drawOnImage(cv::Mat &image, const std::vector<cv::Point> &points, cv::Scalar color= cv::Scalar(255,255,255), int radius=3, int thickness=2) {

                  std::vector<cv::Point>::const_iterator it= points.begin();

                  // for all corners
                  while (it!=points.end()) {

                          // draw a circle at each corner location
                          cv::circle(image,*it,radius,color,thickness);
                          ++it;
                  }
          }
};

#endif