基于注意力网络的地基SAR时序差分相位分类方法  

Time Series Differential Phase Classification of Ground-based SAR Based on Attention Network

作　　者：王彦平 崔紫维 曹琨 李洋[1] 林赟 申文杰 WANG Yanping;CUI Ziwei;CAO Kun;LI Yang;LIN Yun;SHEN Wenjie(Radar Monitoring Technology Laboratory,School of Information Science and Technology,North China University of Technology,Beijing 100144,China)

机构地区：[1]北方工业大学信息学院雷达监测技术实验室,北京100144

出　　处：《信号处理》2021年第7期1207-1216,共10页Journal of Signal Processing

基　　金：国家重点研发计划资助(2018YFC1505103);国家自然科学基金重点国际合作研究项目(61860206013);国家自然科学基金面上项目(61971456)。

摘　　要：地基合成孔径雷达(Ground Based Synthetic Aperture Radar,GBSAR)是目前露天矿山工作帮及排土场进行亚毫米级形变监测的主要技术手段之一,但监测过程中出现的多径效应造成的差分相位变化会被错误识别为形变。针对识别形变精度低的问题,本文开展了差分干涉相位时序特征表达方法的研究,并以此为基础提出了一种基于注意力网络模型的地基SAR时序差分相位分类方法,以形变变化趋势与区域范围作为依据来区分突变区域和缓变区域,通过模型预测出真实形变分布。经过实验结果证明,注意力网络模型可以较为准确的提取出形变分布,有效减少多径效应造成的误差干扰。Ground based synthetic aperture radar(GBSAR)is one of the main techniques for sub millimeter deformation monitoring of working slope and dump in open pit mine.However,the differential phase change caused by multipath effect in the monitoring process will be mistakenly identified as deformation.Aiming at the problem of low accuracy of deformation recognition,this paper studies the expression method of differential interferometric phase sequence features,and proposes a time-series differential phase classification method for ground-based SAR Based on attention network model.Based on the deformation trend and regional range,the abrupt change region and the gradual change region are distinguished,and the real deformation distribution is predicted by the model.The experimental results show that the attention network model can accurately extract the deformation distribution,and effectively reduce the error interference caused by multipath effect.

关 键 词：地基SAR 注意力机制 时序特征 多径误差 

分 类 号：P237[天文地球—摄影测量与遥感]