Magnetic Resonance Brain Image Segmentation and Reconstruction Technique Based on Genetic Fuzzy Clustering Technique

摘要

Medical image segmentation a classic difficult problem in medical image processing. Image segmentation based on clustering technology is widely applied in medical image segmentation. Based on deep analysis on image segmentation based on fuzzy clustering technology, for the existing problems and characteristics of MR brain image segmentation, the paper proposes a magnetic resonance brain image segmentation method combining automatic threshold, genetic algorithm and fuzzy clustering algorithm. The experimental results prove that the segmentation quality and segmentation velocity is higher than that of single rapid fuzzy cluster, which solves the problem of the present fuzzy cluster that initial clustering center and initial membership matrix is difficult to be determined and algorithm iteration is easy to be local extreme.