Medical image segmentation based on the minimum spanning tree
Xiaodong Long, Ziyu Sheng, Jian Sun · Proceedings of the 2020 4th International Conference on Electronic Information Technology and Computer Engineering · 2020
Due to the complex background of medical image, the diversity of target features and the multiple effects of noise, the medical image segmentation has always been a difficult and hot topic in this field. To tackle this challenge, the paper combines the minimum spanning tree (MST) image segmentation model and proposes a new weight calculation standard. The weight calculation formula effectively solves the influence of noise in the medical image, and preserves the details in the image during the segmentation process, which is convenient for subsequent medical processing. The formula fully considers the spatial distance and vector relationship between pixels, and has good stability and application effect. We have proved through experiments that the segmentation algorithm has certain advancement among similar methods.