WANG Yafu, LIANG Buge, YANG Degui, et al. Quantity identifying technology for static target of ultra-wideband through-wall radar based on CNN[J]. 2021, (4): 19-22.
WANG Yafu, LIANG Buge, YANG Degui, et al. Quantity identifying technology for static target of ultra-wideband through-wall radar based on CNN[J]. 2021, (4): 19-22. DOI: 10.13873/J.1000-9787(2021)04-0019-04.
基于CNN的超宽带穿墙雷达静目标数量识别技术
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。
Abstract
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.