殷君君2,,
曾亮1,,,
杨健1,,
1.清华大学电子工程系 北京 100084
2.北京科技大学计算机与通信工程学院 北京 100083
基金项目:国家自然科学基金(61771043)，中央高校基本科研业务费专项资金(FRF-IDRY-19-008, FRF-GF-19-017B)
详细信息
作者简介:朱庆涛(1997–)，男，清华大学电子工程系在读硕士研究生，研究方向为极化SAR图像处理。E-mail: zqt19@mails.tsinghua.edu.cn
殷君君(1983–)，女，北京科技大学计算机与通信工程学院副教授，研究方向为雷达极化应用的基础理论，极化合成孔径雷达图像理解、图像分割、数据融合，海洋遥感及生态环境变化监测。E-mail: yinjj07@gmail.com
曾亮：曾 亮(1990–)，男，清华大学电子工程系在读博士研究生，主要研究方向为极化雷达信号处理、微波遥感、精确制导。E-mail: zengliang14@mails.tsinghua.edu.cn
杨健：杨 健(1965–)，男，湖北襄阳人，分别在西北工业大学和日本新潟大学获得学士、硕士和博士学位，2000年回国，现在为清华大学教授，博士生导师，研究方向为极化雷达理论及其应用。E-mail: yangjian_ee@tsinghua.edu.cn
通讯作者:曾亮 zengliang14@mails.tsinghua.edu.cn
杨健 yangjian_ee@tsinghua.edu.cn
责任主编：陈思伟 Corresponding Editor: CHEN Siwei中图分类号:TP75
计量
文章访问数:790
HTML全文浏览量:324
PDF下载量:156
被引次数:0
出版历程
收稿日期:2020-08-29
修回日期:2020-11-05
网络出版日期:2020-11-20
Polarimetric SAR Image Affine Registration Based on Neighborhood Consensus
ZHU Qingtao1,,YIN Junjun2,,
ZENG Liang1,,,
YANG Jian1,,
1. Department of Electronic Engineering, Tsinghua University, Beijing 100084, China
2. School of Computer and Communication Engineering, University of Science and Technology, Beijing 100083, China
Funds:The National Natural Science Foundation of China (61771043), The Fundamental Research Funds for the Central Universities (FRF-IDRY-19-008, FRF-GF-19-017B)
More Information
Corresponding author:ZENG Liang, zengliang14@mails.tsinghua.edu.cn;YANG Jian, yangjian_ee@tsinghua.edu.cn
摘要
摘要:极化SAR图像的配准是极化SAR图像处理的基础，需要具备较高的精度与速度。基于深度学习的极化SAR图像配准大多数是结合图像块特征的匹配与基于随机抽样一致性的参数迭代估计来实现的。目前尚未实现端到端的基于深度卷积神经网络的一步仿射配准。该文提出了一种基于弱监督学习的端到端极化SAR图像配准框架，无需图像切块处理或迭代参数估计。首先，对输入图像对进行特征提取，得到密集的特征图。在此基础上，针对每个特征点保留k对相关度最高的特征点对。之后，将该4D稀疏特征匹配图输入4D稀疏卷积网络，基于邻域一致性进行特征匹配的过滤。最后，结合输出的匹配点对置信度，利用带权最小二乘法进行仿射参数回归，实现图像对的配准。该文采用RADARSAT-2卫星获取的德国Wallerfing地区农田数据以及PAZ卫星获取的中国舟山港口地区数据作为测试图像对。通过对升降轨、不同成像模式、不同极化方式、不同分辨率的极化SAR图像对的配准测试，并与4种现有方法进行对比，验证了该方法具有较高的配准精度与较快的速度。
关键词:邻域一致性/
仿射变换/
极化SAR/
图像配准/
稀疏卷积神经网络
Abstract:As the base of Synthetic Aperture Radar (SAR) image processing, the registration of polarimetric SAR images requires high accuracy and a fast speed. Most methods used to register polarimetric SAR images based on deep learning are combined with patch matching and iterative estimation, e.g. the random sample consensus algorithm. However, end-to-end deep convolutional neural networks have not been used in the non-iterative affine registration of polarimetric SAR images. This paper proposes a framework for end-to-end polarimetric SAR image registration that is based on weakly-supervised learning and uses no image patch processing or iterative parameter estimation. First, feature extraction is performed on input image pairs to obtain dense feature maps with the most relevant k matches kept for each feature point. To filter the matched feature pairs, the 4D sparse feature matching maps are then fed into a 4D sparse convolutional network based on neighborhood consensus. Lastly, the affine parameters are solved by the weighted least square method according to the degree of confidence of the matches, which enables the affine registration of the input image pair. As test image pairs, we use farmland data from Wallerfing, Germany obtained by the RADARSAT-2 satellite and Zhoushan port data from China obtained by the PAZ satellite. Comprehensive experiments were conducted on polarimetric SAR image pairs using different orbit directions, imaging modes, polarization types and resolutions. Compared with four existing methods, the proposed method was found to have high accuracy and a fast speed.
Key words:Neighborhood consensus/
Affine transformation/
Polarimetric SAR/
Image registration/
Sparse convolutional neural network
PDF全文下载地址:
https://plugin.sowise.cn/viewpdf/198_85f37a53-da30-4ca1-9f6b-bd833a144bbd_R20120
