| 摘要: |
| 针对单频段雷达利用微多普勒特征识别人体动作能力有限的问题，提出了一种基于双频段调频连续波(Frequency Modulation Continuous Wave,FMCW)雷达的人体动作识别方法。首先利用K频段与C频段两部不同频段的FMCW雷达分别对人体不同动作进行探测收集到回波数据，对回波数据进行预处理分别得到距离时间、距离多普勒与微多普勒时频谱图像；然后，运用主成分分析法对图像进行特征提取得到特征向量，对提取到的特征向量进行特征级融合；最后，将融合后的特征作为支持向量机的输入从而实现人体动作识别。采用雷达实测数据的实验结果表明，基于双频段FMCW雷达联合工作的探测方法对五种人体动作的识别正确率为96.25%，优于单个频段FMCW雷达单独工作时的动作识别正确率。 |
| 关键词: 人体动作识别 双频段调频连续波雷达 特征提取 主成分分析 |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目（61671310）；航空科学基金（2019ZC054004）；辽宁省兴辽英才计划项目基金(XLYC1907134)；辽宁省百千万人才工程项目基金(2018B21) |
|
| Human motion recognition based on dual-band FMCW radar |
| QU Lele,ZHANG Dingyuan,YANG Tianhong,SUN Yanpeng |
| (College of Electronic and Information Engineering,Shenyang Aerospace University,Shenyang 110136,China) |
| Abstract: |
| To solve the problem of limited recognition ability of the single-band frequency modulation continuous wave(FMCW） radar sensor,a human motion recognition method based on dual-band FMCW radar is proposed.First,two FMCW radar sensors with K-band and C-band are used to detect different human activities and collect the echo data.The echo data is preprocessed to generate range-time image,range-Doppler image and micro-Doppler image.Then,the principal component analysis is used to extract the feature vectors from the image and the feature level fusion is performed on the extracted feature vectors.Finally,the fused features are used as input of support vector machine to realize human motion recognition.The experimental results show that the recognition accuracy of five kinds of human activities using dual-band FMCW radar sensor is 96.25%,which is higher than that of single-band FMCW radar sensor. |
| Key words: human motion recognition dual-band FMCW radar feature extraction principal component analysis |
