Medical Image Segmentation Algorithm Based on Markup Improvement Combining the Watershed Algorithm and the Level Set Algorithm

Abstract

In this paper, a method is proposed to merge the level set segmentation and the watershed segmentation for edge detection in medical images. To solve the issue of over-segmentation in the watershed algorithm, this improved watershed algorithm is proposed. By identifying the foreground and background, the number of segmented zones can be adjusted. For the process of filling the valley bottom, the divided region's contour is enhanced and modified to ensure that it fits as closely as possible to the target edge. The location of the edge in the gradient map is enhanced and adjusted, while the largest connected region is applied to acquire image edge information. Experiments have shown that this technique can reduce over-segmentation in medical images.