Image Segmentation Based on 3-D Maximum Between-Cluster Variance

Abstract

A method for image segmentation based on 3 D maximum between cluster variance (3 DMBV) is proposed.It constructs 3D observation space using not only gray distribution information of pixels,but also relevant information of neighboring pixels.Based on the competitioo,redundancy and complementation of all information,it can obtain better performance than 1 DMBV methods through efficient fusion.Theoretical analysis and experiments prove that it performs well even on ground target image with low SNR and low contrast.We also propose a recursive algorithm to implement the 3 DMBV method,which can reduce the computation time and necessary storage space.