基于CFAR和CNN的SAR图像目标识别技术  

SAR Image Target Recognition Technology Based on CFAR and CNN

作　　者：张官荣[1] 赵玉 陈相 李波[1] 王建军 刘丹 ZHANG Guanrong;ZHAO Yu;CHEN Xiang;LI Bo;WANG Jianjun;LIU Dan(Aeronautics Engineering College,Air Force Engineering University,Xi'an 710000,China;School of Electronics and Information,Northwestern Polytechnical University,Xi'an 710000,China;No.93046 Unit of PLA,Shenyang 110000,China)

机构地区：[1]空军工程大学航空工程学院,西安710000 [2]西北工业大学电子信息学院,西安710000 [3]中国人民解放军93046部队,沈阳110000

出　　处：《电光与控制》2022年第7期119-125,共7页Electronics Optics & Control

摘　　要：合成孔径雷达(SAR)图像自动目标识别(ATR)技术是人工图像解译的关键技术之一,其旨在屏蔽固有噪声影响,获取感兴趣区域内表征目标的潜在特征信息,为目标识别提供有力的数据支撑。为了提升高分辨SAR图像目标识别精度,围绕算法设计中的相干斑抑制和特征提取问题,结合传统恒虚警率(CFAR)检测算法和深度卷积神经网络(DCNN)的最新研究,设计了SAR图像自动目标识别框架。实验基于MSTAR标准数据集,目标识别结果表明所构建模型的有效性。Synthetic Aperture Radar(SAR)image Automatic Target Recognition(ATR)technology is one of the key technologies of artificial image interpretationwhich aims to isolate the influence of inherent noiseobtain the potential characteristic information of the target in the region of interestand provide strong data support for target recognition.In order to improve the accuracy of target recognition in high-resolution SAR imagesfocusing on the problems of speckle suppression and feature extraction in algorithm designan automatic target recognition framework for SAR images is designed by combining the traditional Constant False Alarm Rate(CFAR)detection algorithm and the latest research of the Deep Convolutional Neural Network(DCNN).The experiment is based on MSTAR standard data setand the results of target recognition show the effectiveness of the model.

关 键 词：SAR图像目标识别 相干斑抑制 特征学习 卷积自编码网络 卷积神经网络 

分 类 号：TN959.17[电子电信—信号与信息处理]