Thursday, February 14, 2013

Median Filter using C++ and OpenCV: Image Processing

Basic Theory

Median filter also reduces the noise in image like low pass filter, but it is better than low pass filter in the sense that it preserves the edges and other details.  The process of calculating the intensity of central pixel is same as that of low pass filtering except instead of averaging all the neighbours, we sort the window and replace the central pixel with median from the sorted window. For example, lets we have a window like this
Now we sort the given window and get the sorted array as [1 1 2 2 3 3 5 6 7]. The median of this array is 4th element i.e 3. Now we replace the central element 7 with 4. That's it.

Source Code

#include<iostream>
#include<opencv2/imgproc/imgproc.hpp>
#include<opencv2/highgui/highgui.hpp>

using namespace std;
using namespace cv;

//sort the window using insertion sort
//insertion sort is best for this sorting
void insertionSort(int window[])
{
    int temp, i , j;
    for(i = 0; i < 9; i++){
        temp = window[i];
        for(j = i-1; j >= 0 && temp < window[j]; j--){
            window[j+1] = window[j];
        }
        window[j+1] = temp;
    }
}

int main()
{
      Mat src, dst;

      // Load an image
      src = imread("book.png", CV_LOAD_IMAGE_GRAYSCALE);

      if( !src.data )
      { return -1; }

      //create a sliding window of size 9
      int window[9];

        dst = src.clone();
        for(int y = 0; y < src.rows; y++)
            for(int x = 0; x < src.cols; x++)
                dst.at<uchar>(y,x) = 0.0;

        for(int y = 1; y < src.rows - 1; y++){
            for(int x = 1; x < src.cols - 1; x++){

                // Pick up window element

                window[0] = src.at<uchar>(y - 1 ,x - 1);
                window[1] = src.at<uchar>(y, x - 1);
                window[2] = src.at<uchar>(y + 1, x - 1);
                window[3] = src.at<uchar>(y - 1, x);
                window[4] = src.at<uchar>(y, x);
                window[5] = src.at<uchar>(y + 1, x);
                window[6] = src.at<uchar>(y - 1, x + 1);
                window[7] = src.at<uchar>(y, x + 1);
                window[8] = src.at<uchar>(y + 1, x + 1);

                // sort the window to find median
                insertionSort(window);

                // assign the median to centered element of the matrix
                dst.at<uchar>(y,x) = window[4];
            }
        }

        namedWindow("final");
        imshow("final", dst);

        namedWindow("initial");
        imshow("initial", src);

      waitKey();


    return 0;
}

Output


10 comments:

  1. can u explain about this function? for(int y = 0; y < src.rows; y++)
    for(int x = 0; x < src.cols; x++)
    dst.at(y,x) = 0.0;

    for(int y = 1; y < src.rows - 1; y++){
    for(int x = 1; x < src.cols - 1; x++){

    ReplyDelete
    Replies
    1. It cleans the image, sets it to black

      Delete
  2. what the function of this code ? why use 0.0; ? at dst.at(y,x) = 0.0;

    ReplyDelete
    Replies
    1. Hello arnanda
      This is just an initialization. You can skip this step and this doesn't affect the output.

      Delete
  3. can u explain about this function? for(int y = 0; y < src.rows; y++)
    for(int x = 0; x < src.cols; x++)
    dst.at(y,x) = 0.0;

    for(int y = 1; y < src.rows - 1; y++){
    for(int x = 1; x < src.cols - 1; x++){

    ReplyDelete
    Replies
    1. The first three line is just an initialization of image pixels to 0. You can safely remove this step. The next two loops iterate over all the image pixels and apply the median filter.

      Delete
    2. we can not remove it if we use it in sobel operator. because it's use to iterates border of array

      Delete
  4. please can you tell me how i can add opencv library ?

    i didn't undestand why you've add this

    #include
    #include
    #include

    ReplyDelete
  5. my comment is not shown !!!! why

    ReplyDelete