Optimal Design of Image Edge Detection Algorithm Based on Mathematical Morphology Analysis

As an important technology of digital image processing, ED (edge detection) has always been a research hotspot in the field of computer vision and image processing. Several commonly used ED methods mostly determine whether a pixel is located on the boundary of an object by detecting the state of each pixel and its neighborhood. Although these methods are simple and sensitive to noise, the edges will be broken. In this paper, an ED algorithm based on mathematical morphology is proposed. By studying the multi-scale ED algorithm of various structural elements and morphological operation combinations, an algorithm with good noise filtering performance and accurate image edge positioning is designed. The research results show that the optimized algorithm is obviously superior to other algorithms, and the detected edges are clear and rich. The weighted adaptive morphological ED algorithm based on multi-scale and multi-structure constructed in this paper is robust to noise and blurred images, so that these thin edges and other obvious edges can be retained after the images are blurred or disturbed by noise. This achievement has positive significance for image processing.