摘要: 提出一种利用脚摆动特征进行步态识别的方法。对步态序列图像进行背景提取、图像差分、阈值分割、形态学后处理后,提取行走时的脚摆角作为特征参数,再分别采用BP神经网络、最近邻分类器和K近邻分类器法对这些特征数据进行识别分类与比较分析。实验结果表明,与同类方法相比,该方法可以更快速地进行步态识别,且识别性能较好。
关键词:
步态识别,
脚摆角,
BP神经网络,
最近邻分类器,
K近邻分类器
Abstract: This paper proposes a gait recognition method based on foot swing characteristics. It conducts background subtraction, image difference, threshold segmentation, morphological post-processing, and then extracts the angle which is defined by the toe swing around heel as the characteristic parameter. These parameters are used for identifying by using Back Propagation(BP) neural network, Nearest Neighbor(NN) classifier and K-nearest neighbor classifier respectively. Experimental results show that considering foot swing angle as gait characteristic provides a quick and simple solution of gait recognition.
Key words:
gait recognition,
deflection angle of feet,
Back Propagation(BP) neural network,
Nearest Neighbor(NN) classifier,
K-Nearest Neighbor(KNN) classifier
中图分类号:
李一波, 卑珊珊, 刘婉竹, 刘金英. 基于行走时脚摆角的步态识别方法[J]. 计算机工程, 2012, 38(14): 132-134.
LI Yi-Bei, BEI Shan-Shan, LIU Wan-Zhu, LIU Jin-Yang. Gait Identification Method Based on Deflection Angle of Feet When Walking[J]. Computer Engineering, 2012, 38(14): 132-134.