COLOR IMAGE SEGMENTATION BASED ON BAYESIAN FRAMEWORK AND LEVEL SET
Level set method is a power tool for tracking the curve evolution. From the view, we propose a novel color image segmentation method based on Bayesian and level set. Firstly, we regard the color information of each pixel as a stochastic variable that obeys certain probability distribution and deduce the segmentation model by Bayesian Maximum A Posteriori (MAP). Then we construct the corresponding energy function. Finally, we obtain the curve evolution equation based on multivariate normal distribution by variational method. We choose color remote sensing images and natural images to validate the proposed approach. The experimental results show that it is a feasible and effective approach to color image segmentation.
Bayesian Mazimum A Posteriori (MAP) Level set Color image segmentation Building detection
XI-LIWANG LIN-JUAN WANG
College of Computer Science, Shanxi Normal University, Xian 710062, China
国际会议
2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)
昆明
英文
3484-3489
2008-07-12（万方平台首次上网日期，不代表论文的发表时间）