The following is the explanation to the C++ code for coin detection in C++ using the tool OpenCV.
- The code will only compile in Linux environment.
- To run in windows, please use the file: ‘coin.o’ and run it in cmd. However if it does not run(problem in system architecture) then compile it in windows by making suitable and obvious changes to the code like: Use in place of .
- Compile command: g++ -w coin.cpp -o coin.exe `pkg-config –libs opencv`
- Run command: ./coin
- The image containing coin/coins has to be in the same directory as the code.
Before you run the code, please make sure that you have OpenCV installed on your // system.
Code Snippets Explained:
#include "opencv2/highgui/highgui.hpp" // highgui - an interface to video and image capturing. #include "opencv2/imgproc/imgproc.hpp" // imgproc - An image processing module that for linear and non-linear image filtering, geometrical image transformations, color space conversion and so on. #include <iostream> #include <stdio.h> // The header files for performing input and output. using namespace cv; // Namespace where all the C++ OpenCV functionality resides. using namespace std; // For input output operations. int main() { Mat image; // Mat object is a basic image container. image is an object of Mat. image=imread("coin-detection.jpg",CV_LOAD_IMAGE_GRAYSCALE); // Take any image but make sure its in the same folder. // first argument denotes the image to be loaded. // second argument specifies the image format as follows: // CV_LOAD_IMAGE_UNCHANGED (<0) loads the image as it is. // CV_LOAD_IMAGE_GRAYSCALE ( 0) loads the image in Gray scale. // CV_LOAD_IMAGE_COLOR (>0) loads the image in the BGR format. // If the second argument is not there, it is implied CV_LOAD_IMAGE_COLOR. vector coin; // A vector data type to store the details of coins. HoughCircles(image,coin,CV_HOUGH_GRADIENT,2,20,450,60,0,0 ); // Argument 1: Input image mode // Argument 2: A vector that stores 3 values: x,y and r for each circle. // Argument 3: CV_HOUGH_GRADIENT: Detection method. // Argument 4: The inverse ratio of resolution. // Argument 5: Minimum distance between centers. // Argument 6: Upper threshold for Canny edge detector. // Argument 7: Threshold for center detection. // Argument 8: Minimum radius to be detected. Put zero as default // Argument 9: Maximum radius to be detected. Put zero as default int l=coin.size(); // Get the number of coins. cout<<"\n The number of coins is: "<<l<<"\n\n"; // To draw the detected circles. for( size_t i = 0; i < coin.size(); i++ ) { Point center(cvRound(coin[i][0]),cvRound(coin[i][1])); // Detect center // cvRound: Rounds floating point number to nearest integer. int radius=cvRound(coin[i][2]); // To get the radius from the second argument of vector coin. circle(image,center,3,Scalar(0,255,0),-1,8,0); // circle center // To get the circle outline. circle(image,center,radius,Scalar(0,0,255),3,8,0); // circle outline cout<< " Center location for circle "<<i+1<<" : "<<center<<"\n Diameter : "<<2*radius<<"\n"; } cout<<"\n"; namedWindow("Coin Counter",CV_WINDOW_AUTOSIZE); // Create a window called //"A_good_name". // first argument: name of the window. // second argument: flag- types: // WINDOW_NORMAL : The user can resize the window. // WINDOW_AUTOSIZE : The window size is automatically adjusted to fit the // displayed image() ), and you cannot change the window size manually. // WINDOW_OPENGL : The window will be created with OpenGL support. imshow("Coin Counter",image); // first argument: name of the window // second argument: image to be shown(Mat object) waitKey(0); // Wait for infinite time for a key press. Return 0; // Return from main function. } End of explanation.
About the Author:
Aditya Prakash is an undergraduate student at Indian Institute of Information Technology, Vadodara. He primarily codes in C++. The motto for him is: So far so good. He plays cricket, watches superhero movies, football and is a big fan of answering questions.
If you also wish to showcase your blog here, please see GBlog for guest blog writing on GeeksforGeeks.