关键词: 马尔可夫随机场, 声纳图像, 分割

Abstract: Segmentation of underwater objects using sonar imagery is complicated by the variability of objects，noises，and background.Through analyzing the features of sonar imagery，a kind of new automatic segmentation algorithm for sonar imagery is proposed.A fast fuzzy C-mean clustering algorithm is adopted to complete initial segmentation.Then according to the initial segmentation results，the initial parameters of the Markov random field models are estimated.Lastly，iterative conditional estimation based on the Markov random field theory is used to obtain the final precise segmentation results.In the last part，experiments are conducted to demonstrate the feasibility and effectiveness by the data detected practically.

Key words: Markov random field, sonar imagery, segmentation