新的声纳图像自动分割方法

对水下声纳图像进行目标分割是非常复杂的,它不仅取决于被分割的不同目标,还与海底混响噪声、背景区域有着紧密的联系。通过分析声纳图像的特点,提出了一种新的声纳图像自动分割方法,即利用一种快速的模糊C均值聚类方法来完成初始分割,然后利用初始分割结果对马尔可夫模型的初始参数进行估计,最后,根据马尔可夫理论进行迭代条件估计,得到精确的图像分割结果。最后利用实测数据,验证了此种算法的可行性和有效性。

Novel automatic segmentation algorithm for sonar imagery

Segmentation of underwater objects using sonar imagery is complicated by the variability of objects,noises,and back-ground.Through analyzing the features of sonar imagery,a kind of new automatic segmentation algorithm for sonar imagery is pro-posed.A fast fuzzy C-mean clustering algorithm is adopted to complete initial segmentation.Then according to the initial segmenta-tion results,the initial parameters of the Markov random field models are estimated.Lastly,iterative conditional estimation based on the Ma...