A Novel Method of Medical Image Segmentation Based on the Multi-modal Function Optimization

摘要

According to the Complex medical images are not always described using the parametric method with prior probability, leading to the difference between the actual physical model and the basic hypothesis of the model, namely the problem of "model mismatch", the method of medical image segmentation based on the multi-modal function optimization is proposed in this paper. We proposed a density model of the nonparametric orthogonal polynomials for image data, the novel Particle Swarm Optimization method is used to resolve the multi-modal function optimization problem, On the basis of the heuristic optimization search, the novel method was successful in multi-modal function optimization. The FCM clustering algorithm is used to segment image with local optimal solution as the cluster center.