Image Edge Detection Algorithm of Machined Parts Based on Mathematical Morphology

At present, the edge detection method filters image edge points according to extreme image points, resulting in an unclear detected edge image and a low average energy value. An edge detection algorithm for machining part images based on mathematical morphology is proposed. The binary morphology of mathematical morphology is used to remove image noise. Using a two-dimensional wavelet transform, the edge information matrix in the horizontal and vertical directions of the image is calculated to determine the edge scale of the image. The mask operator in two directions of the image is calculated as a multiscale and multidirectional detection operator. According to the multiscale and multidirectional detection operator, the target edge detection process is designed to smooth the denoised machined part image, screen the image target edge points according to the maximum value points of the image edge gradient, segment the image target and background, and realize target edge detection. The images of machined parts collected in the normal machining room are divided into two categories according to the noise gradient, and the two groups of methods are compared. Experimental results: Compared with the method selected in this experiment, the average energy value of the detected target edge image is higher.