Segmentation of Multi-Look Fully Polarimetric SAR Images Based on Wishart Distribution and MRF

WU Yong-hui; JI Ke-feng; LI Yu; YU Wen-xian

您当前的位置：

首页 >

文章列表页 >

Segmentation of Multi-Look Fully Polarimetric SAR Images Based on Wishart Distribution and MRF

更新时间：2025-07-16

- Segmentation of Multi-Look Fully Polarimetric SAR Images Based on Wishart Distribution and MRF
- Acta Electronica Sinica Vol. 35, Issue 12, Pages: 2302-2306(2007)
- 作者机构：
  
  国防科学技术大学电子科学与工程学院,湖南,长沙,410073
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP753
- Published：2007
- 稿件说明：
移动端阅览
WU Yong-hui, JI Ke-feng, LI Yu, et al. Segmentation of Multi-Look Fully Polarimetric SAR Images Based on Wishart Distribution and MRF[J]. Acta Electronica Sinica, 2007, 35(12): 2302-2306.
DOI：

WU Yong-hui, JI Ke-feng, LI Yu, et al. Segmentation of Multi-Look Fully Polarimetric SAR Images Based on Wishart Distribution and MRF[J]. Acta Electronica Sinica, 2007, 35(12): 2302-2306. DOI：

摘要

提出一种新的多视全极化SAR图像分割方法.将描述多视协方差矩阵的Wishart分布与马尔可夫随机场模型结合起来

利用迭代条件模型法(ICM)求取最大后验概率准则下的分割结果

其中ICM所需的初始分割图由基于Wishart分布的最大似然法获得.NASA/JPL实验室AIRSAR系统多视全极化数据的实验结果表明

与几种常用方法相比

本文方法分割精度更高

分割结果图中孤立像素少

图像连通性好.

Abstract

A novel method is proposed for segmenting multi-look fully polarimetric SAR images.The Wishart distribution of the covariance matrix is integrated with the Markov random field model

and then the iterative conditional modes (ICM) algorithm is taken to implement the maximum a posteriori estimation of pixel labels.The initial segmentation map for ICM is obtained using the Wishart-based ML.Using multi-look fully polarimetric SAR images

acquired by the NASA/JPL AIRSAR sensor

the new approach is compared with several other common ones

higher accuracy are observed

as well as less isolated pixels and better connectivity.

关键词

Keywords

references

Views

1451

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Cross-Modal SAR Target Detection via Progressive Knowledge Transfer

Enhanced PFCM Algorithm for Noisy Image Segmentation Combining Regional and Neighborhood-Level Information

SwinT-Unet: Ultrasound Image Segmentation Based on Two-Channel Self-Attention Mechanism

End-to-End Full Complex-Valued Domain SAR Target Classification Neural Network

Related Author

ZHAO Guo-wei

JIANG Jia-qing

DONG Gang-gang

WANG Xiao-peng

WANG Hai-zhou

CHEN Hao-ran

SONG Yan-tao

LU Yun-li

Related Institution

National Key Laboratory of Radar Signal Processing, Xidian University

School of Electronic and Information Engineering, Lanzhou Jiaotong University

Institute of Big Data Science & Industry, Shanxi University

School of Computer & Information Technology, Shanxi University

College of Electronic Information and Automation， Civil Aviation University of China

⁰