Tutorial 6 - Assessing the pixel values of an image OpenCv C++

Today I'm going to show you how to assessing the pixel values of an image using OpenCv.

Digital color images are made out of pixels.

First we look at what is a pixel. 

Pixel

Pixel or a dot means a small physical point or an addressable element/point in an image. A pixel has its own coordinates which means that a pixel is corresponds to any one value. If we considering an 8 bit gray scale image the value of a pixel in that image must be a value between 0 and 255. This value is tallied to the intensity of the light photons striking at that point. Each pixel stores a value which is proportional to the light intensity at a given location. 

Let’s see how to get the number of pixels in an image  

Total number of pixels = number of rows x number of columns

Or the number of pixels in the image will equal to the total number of (x,y) coordinate pairs of pixels. (Address of an image can be written in x,y coordinates).

Following example will guide you to get the number of channels in an image.

Example 1

============================================================ 

#include "opencv2/core/core.hpp"

#include "opencv2/highgui/highgui.hpp"

#include "opencv2/imgproc/imgproc.hpp"

#include "iostream"

using namespace cv;

using namespace std;

int main()

{

          Mat img;

         

          img = imread("C:/Users/hashan/Desktop/progtpoint.jpg", CV_LOAD_IMAGE_GRAYSCALE);

                //img = imread("C:/Users/hashan/Desktop/progtpoint.jpg", CV_LOAD_IMAGE_COLOR);

         

          if (img.empty()) {

                   cout << "Error" << endl;

                   return -1;

          }

          namedWindow("New_image", CV_WINDOW_AUTOSIZE);

          imshow("New_image", img);

          // print number of channels in image

          cout << "image channels: " << img.channels() <<  endl;

          // check if image is a single channel

          if (img.channels() == 1) {

                   Scalar pixel = img.at<uchar>(10, 10);

                   cout << "Pixel value at 5,6 cordinate : " << pixel.val[0] << endl;

          }

    // check if image is a 3 channel

          else if (img.channels() == 3) {

         

                   Vec3b pixel = img.at<Vec3b>(10, 10);

                   int blue = pixel.val[0];

                   int green = pixel.val[1];

                   int red = pixel.val[2];

          }

          else {

                   cout << "this is not a single channel image" << endl;

          }

          waitKey(0);

          return 0;

}

 =============================================================

Explanation 

This code is carried out for a grey scale image which only have single channel. The coordinates of a pixel is 10, 10.

pixel.val[0]  contains a value from 0 to 255.

So the output is,

Let’s try the code for a color image.

You need to comment following line in the code

 img = imread("C:/Users/hashan/Desktop/progtpoint.jpg", CV_LOAD_IMAGE_GRAYSCALE);

And uncomment the following line

//img = imread("C:/Users/hashan/Desktop/progtpoint.jpg", CV_LOAD_IMAGE_COLOR);

Then run the code

I got the output as this. 

Let’s accessing all pixels of a gray color image and a RGB image

Example 2

==============================================================

#include "opencv2/core/core.hpp"

#include "opencv2/highgui/highgui.hpp"

#include "opencv2/imgproc/imgproc.hpp"

#include "iostream"

using namespace cv;

using namespace std;

int main()

{

          Mat img;

          img = imread("C:/Users/hashan/Desktop/pixel.jpg", CV_LOAD_IMAGE_GRAYSCALE);

          //img = imread("C:/Users/hashan/Desktop/pixel.jpg", CV_LOAD_IMAGE_ANYCOLOR);

         

          if (img.empty()) {

                   cout << "Error" << endl;

                   return -1;

          }

          namedWindow("New_image", CV_WINDOW_NORMAL);

          imshow("New_image", img);

          // print number of channels in image

          cout << "image channels: " << img.channels() <<  endl;

          cout << "Pixel value: " << endl;

          // check if image is a single channel

          if (img.channels() == 1) {

                   for (int y = 0; y < img.cols; y++) {

                             for (int x = 0; x < img.rows; x++) {

                                      Scalar pixel = img.at<uchar>(x, y);

                                      cout << pixel.val[0] << ",";

                             }

                             cout << endl;

                   }

          }

          else if (img.channels() == 3) {

                   for (int y = 0; y < img.cols; y++) {

                             for (int x = 0; x < img.rows; x++) {

                                      Vec3b pixel = img.at<Vec3b>(x, y);

                                      int blue = pixel.val[0];

                                      int green = pixel.val[1];

                                      int red = pixel.val[2];

                                      cout << "[" << blue << " " << green << " " << red << "],";

                             }

                             cout << endl;

                   }

          }

          else {

                   cout << "this is not a single channel image" << endl;

          }

          waitKey(0);

          return 0;

}

============================================================= 

My outputs are follows

·        When I used a gray color image

When I used a RGB image

• Thank you very much for reading this tutorial.
• If you have problems regarding this tutorial feel free to make a comment below or contact us by sending message through our facebook page or by sending email to this address progtpoint@gmail.com 
• Follow us on facebook and Twitter also subcirbe us youtube