OpenCV image edge extraction (1) - Robert operator principle and pure algorithm implementation (C#)

Algorithm principle

Robert operator, also known as Roberts edge detection operator, is an operator that uses local difference operators to find edges. Robert operator is used to solve the gradient in the diagonal direction of the image to find edges, as shown below:
Insert image description here

Gradient calculation:
Insert image description here

Insert image description here

Calculation illustration:
Insert image description here
The first convolution calculation result of the pixel at (0, 0) is: |120×1+125×(-1)|=5

Algorithm implementation

Code implementation (pure algorithm):

            if (fileDialog.ShowDialog() == DialogResult.OK)
            {
    
    
                
                picFile = fileDialog.FileName;
                inputMat = Cv2.ImRead(picFile, ImreadModes.Grayscale);
                outMat = new Mat(new Size(inputMat.Cols, inputMat.Rows), inputMat.Type());

                for (int i = 0; i < inputMat.Rows - 1; i++)
                {
    
    

                    for (int j = 0; j < inputMat.Cols - 1; j++)
                    {
    
    
                        var left_up = inputMat.Get<byte>(i, j);
                        var right_down= inputMat.Get<byte>(i+1, j+1);
                        var result = Math.Abs(left_up - right_down);
                        outMat.Set<byte>(i, j, OpenCvSharp.Internal.Util.SaturateCast.ToByte(result));
                    }
                }
                picBox_Display.Image = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(inputMat);
                picBox_After.Image = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(outMat);
            }

Insert image description here

In the same way, define the following operator to retain the edge from the upper right to the lower left direction.
Insert image description here

OpenCV API implementation

          if (fileDialog.ShowDialog() == DialogResult.OK)
            {
    
    
                

                picFile = fileDialog.FileName;
                inputMat = Cv2.ImRead(picFile, ImreadModes.Grayscale);
                outMat = new Mat(new Size(inputMat.Cols, inputMat.Rows), inputMat.Type());

                //左上到右下方向滤波器
                InputArray kernel = InputArray.Create<int>(new int[2, 2] {
    
     {
    
     1, 0}, {
    
     0, -1 } });
                Cv2.Filter2D(inputMat, outMat, MatType.CV_16SC1, kernel);
                
                //矩阵取绝对值
                Cv2.ConvertScaleAbs(outMat, outMat);
                picBox_Display.Image = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(inputMat);
                picBox_After.Image = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(outMat);
            }

Insert image description here

Additional instructions:

The OpenCV library used in this case in .NET is OpenCvSharp4
The edge extraction of the entire image can be obtained by adding the pictures from the upper left to the lower right and the upper right to the lower left directions;

OpenCv library for .NET environment