基于上下文和隐类属的小波域马尔可夫随机场SAR图像分割

被引:9
作者
张强
吴艳
机构
[1] 西安电子科技大学电子工程学院
关键词
SAR图像分割; 多尺度分割; 小波域混合长拖尾模型; 隐类属马尔可夫随机场; 上下文模型;
D O I
暂无
中图分类号
TN958 [雷达:按体制分]; TP391.41 [];
学科分类号
080203 ;
摘要
该文针对合成孔径雷达(Synthetic Aperture Radar,SAR)图像含有大量的乘性斑点噪声的特点,提出了一种小波域隐类属的马尔可夫随机场(Markov Random Field,MRF)图像分割算法来抑制噪声的影响。考虑到小波的聚集性和持续性,该算法重新构造了待分图像小波域模型——以类属为隐状态的混合长拖尾模型,将隐类属的马尔可夫随机场推广到小波域上,并用改进的上下文模型估计尺度间转移概率,最后推导出了新的最大后验(Maximum A Posteriori,MAP)分割公式。仿真结果证明,该算法具有鲁棒性能够有效地抑制噪声对图像的影响,得到准确的分割结果。
引用
收藏
页码:211 / 215
页数:5
相关论文
共 9 条
[1]  
Segmentation of SAR images based on Markov random field model. Lankoande O,Hayat M M,and Balu Santhanam. 2005IEEE International Conference on System,Man and Cybernetics,United States . 2005
[2]  
SAR speckle reduction using wavelet denoising and Markov random field modeling. Xie Hua,Pierce L E,Ulaby F T. IEEE Transactions on Geoscience and Remote Sensing . 2002
[3]  
Wavelet-based statistical signal processing using hidden markov models. Crouse M S,Nowak R D,Baraniuk R G. IEEE Transactions on Signal Processing . 1998
[4]  
Multiscale image segmentation using wavelet-domain hidden markov models. Choi H,Baraniuk R. IEEE Transactions on Image Processing . 2001
[5]  
A theory formultiresolution signal decomposition. Mallat SG. IEEE Transactions on Pattern Analysis and Machine Intelligence . 1989
[6]  
Multiscale SAR image segmentation using wavelet domain hidden Markov tree models. Venkatachalam V,Choi H,and Baraniuk R G. Proceedings of SPIE-The InternationalSociety for Optical Engineering . 2000
[7]  
Unsupervised segmentation of synthetic aperture radar sea ice imagery using a novel Markov random field model. Deng Huawu,and Clausi D A. IEEE Transactions on Geoscience and Remote Sensing . 2005
[8]  
Segmentation of textured images using Gibbs random fields. Derin H,and Cole W. CVGIP . 1986
[9]  
A multiscale random field model for Bayesian image segmentation. Bouman C A,and Shapiro M. IEEE Trans.on Image Processing . 1994