Abstract: A new method for segmentation of synthetic aperture radar（SAR） images is presented to consider spatial distributed characters between pixels of SAR images as well as the local means and variances statistics of gray level，a Gaussian autoregressive（GAR） model under a multiresolution pairwise Markov framework.Based on texture feature images from gray level co-occurrence probability statistics，we examine the texture segmentation of SAR image suing the multi-resolution maximization of the posterior marginal（MPM） estimate with corresponding unsupervised segmentation algorithm on those texture feature images.For some SAR images，compared with multiresolution pariwise Markov-GAR model texture segmentation based on gray level images，the results of experimentation show that the proposed method in this paper has a better performance on segmentation precision.

Key words: Gray Level Co-occurrence Matrices, SAR image, pairwise Markov-GAR model, multiresolution MPM, texture segmentation

摘要： 同时考虑SAR图像局部灰度均值和方差及像素空间分布特征等统计量，在以灰度共生矩阵产生的纹理统计量为特征所生成的图像上，建立多分辨双Markov-GAR模型，采用多分辨MPM的参数估计方法及相应的无监督分割算法，对SAR图像进行纹理分割。该方法用于一些高分辨SAR图像，其分割精度及分割边缘的平滑度均优于基于灰度图像上的多分辨双Markov-GAR模型纹理分割。

关键词: 灰度共生矩阵, SAR图像, 双Markov-GAR模型, 多分辨MPM, 纹理分割