Applying Box Filter Tutorial

Advertisements


We are going to apply Box filter that blurs an image. A Box filter could be of dimensions 3x3,5x5,9x9 e.t.c.

We are going to use OpenCV function filter2D to apply Box filter to images. It can be found under Imgproc package. Its syntax is given below:

filter2D(src, dst, ddepth , kernel, anchor, delta, BORDER_DEFAULT );

The arguments of the functions and their description is listed below:

Sr.NoArgument & Description
1src
Source image.
2dst
Destiantion image.
3ddepth:
The depth of dst. A negative value (such as -1) indicates that the depth is the same as the source.
4kernel
The kernel to be scanned through the image.
5anchor
The position of the anchor relative to its kernel. The location Point(-1, -1) indicates the center by default.
6delta
A value to be added to each pixel during the convolution. By default it is 0.
7BORDER_DEFAULT
We let this value by default.

Apart from the filter2D method, there are other methods provide by the Imgproc class. They are listed below:

Sr.NoMethod & Description
1cvtColor(Mat src, Mat dst, int code, int dstCn)
Converts an image from one colour space to another.
2dilate(Mat src, Mat dst, Mat kernel)
Dilates an image by using a specific structuring element.
3equalizeHist(Mat src, Mat dst)
Equalizes the histogram of a grayscale image.
4filter2D(Mat src, Mat dst, int ddepth, Mat kernel, Point anchor, double delta)
Convolves an image with the kernel.
5GaussianBlur(Mat src, Mat dst, Size ksize, double sigmaX)
Blurs an image using a Gaussian filter.
6integral(Mat src, Mat sum)
Calculates the integral of an image.

Example

The following example demonstrates the use of Imgproc class to apply Box filter to an image of GrayScale.

import org.opencv.core.Core;
import org.opencv.core.CvType;
import org.opencv.core.Mat;
import org.opencv.highgui.Highgui;
import org.opencv.imgproc.Imgproc;


public class convolution {
   public static void main( String[] args )
   {
   try {
      int kernelSize = 9;
      System.loadLibrary( Core.NATIVE_LIBRARY_NAME );
      Mat source = Highgui.imread("grayscale.jpg",
      Highgui.CV_LOAD_IMAGE_GRAYSCALE);
      Mat destination = new Mat(source.rows(),source.cols(),source.type());
      Mat kernel = Mat.ones(kernelSize,kernelSize, CvType.CV_32F);	      
      for(int i=0; i<kernel.rows(); i++){
         for(int j=0; j<kernel.cols(); j++){
            double[] m = kernel.get(i, j);
            for(int k =0; k<m.length; k++){
               m[k] = m[k]/(kernelSize * kernelSize);
            }
            kernel.put(i,j, m);
         }
      }	      
      Imgproc.filter2D(source, destination, -1, kernel);
      Highgui.imwrite("output.jpg", destination);
      } catch (Exception e) {
         System.out.println("Error: " + e.getMessage());
      }
   }
}

Output

Original Image

Applying Box Filter Tutorial

In this example we convolve our image with the following filter(kernel). This filter results in blurring an image as its size increases.

This original image has been convolved with the box filter of size 5 , which is given below:

Box filter of size 5.

1/251/251/251/251/25
1/251/251/251/251/25
1/251/251/251/251/25
1/251/251/251/251/25
1/251/251/251/251/25

Convolved Image(With Box Filter of size 5).

Applying Box Filter Tutorial

Convolved Image (With Box Filter of size 9).

Applying Box Filter Tutorial

Advertisements
Advertisements