基于同步压缩短时傅里叶变换的毫米波雷达人体动作识别  

Millimeter wave radar human actionrecognition based on synchrosqueezing short⁃time Fourier transform

作　　者：陈晓楠[1] 汪恩铭 于欣瑶 李姝雅 CHEN Xiaonan;WANG Enming;YU Xinyao;LI Shuya(Information Science and Technology College,Dalian Maritime University,Dalian 116000,China)

机构地区：[1]大连海事大学信息科学技术学院,辽宁大连116000

出　　处：《现代电子技术》2023年第9期46-49,共4页Modern Electronics Technique

摘　　要：针对人体动作识别问题,提出一种基于同步压缩短时傅里叶变换的人体动作识别方法。使用毫米波雷达进行人体动作数据的采集,将采集到的数据进行同步压缩短时傅里叶变换得到其时频图;然后使用卷积神经网络对不同动作进行微多普勒特征提取并分类。在数据采集部分,使用毫米波雷达进行数据采集,有效地避免了外界因素的影响;在时频分析部分,使用窗函数优化的同步压缩短时傅里叶变换提高了时频聚集性。实验结果表明,该人体动作识别系统对不同人体动作的识别率可达到91.7%。A method of human action recognition based on synchrosqueezing short⁃time Fourier transform(SSTFT)is proposed.The millimeter wave radar is adopted to collect human motion data,and then the data is synchronously compressed and short⁃time Fourier transform is used for it to obtain its time⁃frequency map.The convolutional neural network is used to extract the micro⁃Doppler features of different actions.In the data acquisition part,the use of millimeter⁃wave radarfor data acquisition can effectively avoid the influence of external factors.In the time⁃frequency analysis part,the SSTFT optimized with the window function improves the time⁃frequency aggregation.The experimental results show that the recognition rate of the human action recognition system for different human actions can reach 91.7%.

关 键 词：人体动作识别 毫米波雷达 同步压缩 短时傅里叶变换 数据采集 特征提取 时频分析 

分 类 号：TN958.94-34[电子电信—信号与信息处理]