基于CNN的超宽带穿墙雷达静目标数量识别技术  

Quantity identifying technology for static target of ultra-wideband through-wall radar based on CNN

作　　者：王亚夫 梁步阁[1] 杨德贵[1] 朱政亮 WANG Yafu;LIANG Buge;YANG Degui;ZHU Zhengliang(School of Aeronautics and Astronautics,Central South University,Changsha 410083,China)

机构地区：[1]中南大学航空航天学院,湖南长沙410083

出　　处：《传感器与微系统》2021年第4期19-22,共4页Transducer and Microsystem Technologies

基　　金：国家重点研发计划资助项目(2018YFC0810202)。

摘　　要：本文基于静止目标的超宽带(UWB)雷达模型的构建,提出利用卷积神经网络(CNN)进行人体数量的识别方法,针对实测数据对该方法进行实验分析,并与传统方法进行比较,结果表明:将超宽带穿墙雷达与CNN结合对人体数量识别准确率较高,并且去除数据预处理环节对识别效果没有影响。Based on the construction of the ultra-wideband(UWB)radar model of stationary targets,a method for identifying the number of human bodies using convolutional neural network(CNN)is proposed.The method is analyzed based on the actual measurement data and compared with the traditional methods.The results show that the ultra-wideband through-wall radar and the method of CNN combined has a high accuracy in identifying the number of human bodies,and the removal of data preprocessing links has no effect on the recognition effect.

关 键 词：超宽带(UWB)穿墙雷达 卷积神经网络 人体数量识别 静目标 

分 类 号：TN95[电子电信—信号与信息处理]