Java DIP - Sobel Operator

Sobel operator is very similar to Prewitt operator. It is also a derivative mask and is used for edge detection. Sobel operator is used to detect two kinds of edges in an image: Vertical direction edges and Horizontal direction edges.

We are going to use OpenCV function filter2D to apply Sobel operator to images. It can be found under Imgproc package. Its syntax is given below −

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

The function arguments are described below −

Sr.No.	Argument
1	src It is source image.
2	dst It is destination image.
3	depth It is the depth of dst. A negative value (such as -1) indicates that the depth is the same as the source.
4	kernel It is the kernel to be scanned through the image.
5	anchor It is the position of the anchor relative to its kernel. The location Point (-1, -1) indicates the center by default.
6	delta It is a value to be added to each pixel during the convolution. By default it is 0.
7	BORDER_DEFAULT We let this value by default.

Apart from the filter2D method, there are other methods provide by the Imgproc class. They are described briefly −

Sr.No.	Method & Description
1	cvtColor(Mat src, Mat dst, int code, int dstCn) It converts an image from one color space to another.
2	dilate(Mat src, Mat dst, Mat kernel) It dilates an image by using a specific structuring element.
3	equalizeHist(Mat src, Mat dst) It equalizes the histogram of a grayscale image.
4	filter2D(Mat src, Mat dst, int depth, Mat kernel, Point anchor, double delta) It convolves an image with the kernel.
5	GaussianBlur(Mat src, Mat dst, Size ksize, double sigmaX) It blurs an image using a Gaussian filter.
6	integral(Mat src, Mat sum) It calculates the integral of an image.

Example

The following example demonstrates the use of Imgproc class to apply Sobel operator 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 = new Mat(kernelSize,kernelSize, CvType.CV_32F) {
            {
               put(0,0,-1);
               put(0,1,0);
               put(0,2,1);

               put(1,0-2);
               put(1,1,0);
               put(1,2,2);

               put(2,0,-1);
               put(2,1,0);
               put(2,2,1);
            }
         };	      
         
         Imgproc.filter2D(source, destination, -1, kernel);
         Highgui.imwrite("output.jpg", destination);
         
      } catch (Exception e) {
         System.out.println("Error: " + e.getMessage());
      }
   }
}

Output

When you execute the given code, the following output is seen −

Original Image

This original image is convolved with the Sobel operator of vertical edges, which is given below −

Vertical Direction

-1	0	1
-2	0	2
-1	0	1

Convolved Image(Vertical Direction)

This original is convolved with the Sobel operator of horizontal edges, which is given below −

Horizontal Direction

-1	-2	-1
0	0	0
1	2	1