Medical Image Segmentation Based on Modified Ant Colony Algorithm with GVF Snake Model

Abstract

In order to distinguish normal tissues and abnormal pathological changes in the clinic diagnose and pathology, it is required to segment the medical images. The snake model is an important method of getting the contour of the object in the image segmentation. However, it has many defects in some fields such as concavity processing, local optimization, convergence speed and segmentation precision. Aiming at the problem existing in the snake model about falling into its local optimization, a new method of medical image segmentation based on modified ant colony algorithm with GVF snake model is proposed. With adding crowded degree function to ant colony algorithm, the overall traversal ability is increased and the capacity of finding optimal solution is enhanced. The contrast experiments proved that the method in this paper is superior to the segmentation using snake model only in convergence speed, global search performance, and the precision of finding global optimal solution.