《计算机应用》唯一官方网站 ›› 2026, Vol. 46 ›› Issue (10): 3317-3323.DOI: 10.11772/j.issn.1001-9081.2025091139

• 多媒体计算与计算机仿真 • 上一篇    

基于多特征融合的调频连续波雷达步态识别方法

江嘉玮, 郭剑(), 袁铭蔚, 李智春, 孙力娟   

  1. 南京邮电大学 计算机学院,南京 210003
  • 收稿日期:2025-09-26 修回日期:2025-11-21 接受日期:2025-11-24 发布日期:2025-12-02 出版日期:2026-10-10
  • 通讯作者: 郭剑
  • 作者简介:江嘉玮(2001—),女,江苏南通人,硕士研究生,主要研究方向:机器学习、步态识别、行为识别
    袁铭蔚(2006—),女,江苏南通人,主要研究方向:机器学习、步态识别
    李智春(1997—),男,安徽滁州人,硕士,主要研究方向:机器学习、步态识别、无线感知
    孙力娟(1963—),女,江苏南京人,教授,博士,CCF会员,主要研究方向:物联网、无线感知、生物特征识别。
  • 基金资助:
    国家自然科学基金资助项目(62272242)

Gait recognition method based on multi-feature fusion using frequency-modulated continuous wave radar

Jiawei JIANG, Jian GUO(), Mingwei YUAN, Zhichun LI, Lijuan SUN   

  1. School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing Jiangsu 210003,China
  • Received:2025-09-26 Revised:2025-11-21 Accepted:2025-11-24 Online:2025-12-02 Published:2026-10-10
  • Contact: Jian GUO
  • About author:JIANG Jiawei, born in 2001, M. S. candidate. Her research interests include machine learning, gait recognition, behavior recognition.
    YUAN Mingwei, born in 2006. Her research interests include machine learning, gait recognition.
    LI Zhichun, born in 1997, M. S. His research interests include machine learning, gait recognition, wireless sensing.
    SUN Lijuan, born in 1963, Ph. D., professor. Her research interests include internet of things, wireless sensing, biometric recognition.
  • Supported by:
    National Natural Science Foundation of China(62272242)

摘要:

针对现有毫米波点云步态识别方法因过度依赖静态空间特征、缺乏动态运动信息建模,导致识别精度受限的问题,提出一种基于多特征融合的调频连续波(FMCW)雷达步态识别方法。首先,对点云的三维坐标进行空间对齐,以统一空间表示;在此基础上融合目标点的速度与信噪比特征,实现空间结构与动态信息的协同建模。然后,通过多尺度卷积与池化模块进一步提取跨时空步态特征。实验结果显示,所提方法在多个公开数据集与自建数据集上均取得了较高的识别精度,平均准确率为86.3%,最高达到92.2%;在不同场景下,相较于mmGaitNet和PointMLP等对比方法,它的准确率提升幅度为0.90%至32.08%。所提方法通过多维特征融合与多尺度建模有效增强步态动态特征的表达能力,从而提升毫米波雷达步态识别的精确性与鲁棒性。

关键词: 步态识别, 多特征融合, 点云, 调频连续波雷达, 深度学习

Abstract:

A gait recognition method based on multi-feature fusion using Frequency-Modulated Continuous Wave (FMCW) radar was developed to overcome the low recognition accuracy of the existing millimeter-wave point cloud gait recognition methods caused by relying on static spatial features excessively and lacking modeling of dynamic motion information. First, the three-dimensional coordinates of the point cloud were aligned spatially to obtain a unified spatial representation. On this basis, the velocity and signal-to-noise ratio features of target points were fused, and the spatial structure and dynamic information were modeled jointly. Second, cross-spatio-temporal gait features were further extracted by multi-scale convolution and pooling modules. Experimental results show that the proposed method achieves high recognition accuracy on multiple public datasets and a self-built dataset, with an average accuracy of 86.3% and a maximum accuracy of 92.2%. Across different scenarios, it has the recognition accuracy improved by 0.90% to 32.08% compared with comparative methods such as mmGaitNet and PointMLP. It can be seen that through multi-dimensional feature fusion and multi-scale modeling, the proposed method enhances the representation capability of gait dynamic features effectively, and improves the accuracy and robustness of millimeter-wave radar gait recognition.

Key words: gait recognition, multi-feature fusion, point cloud, Frequency-Modulated Continuous Wave (FMCW) radar, deep learning

中图分类号: