电子测量与仪器学报

邵虹,王昳昀.基于集成Gabor特征的步态识别方法[J].电子测量与仪器学报,2017,31(4):573-579

基于集成Gabor特征的步态识别方法

Gait recognition method based on integrated gabor feature

DOI：10.13382/j.jemi.2017.04.012

英文关键词:gait energy image integrated Gabor feature mean fusion differential binary encoding gait recognition

基金项目:辽宁省高等学校优秀人才支持计划(LJQ2013013)资助项目

作者	单位
邵虹	沈阳工业大学信息科学与工程学院沈阳110870
王昳昀	沈阳工业大学信息科学与工程学院沈阳110870

Author	Institution
Shao Hong	School of Information Science and Engineering, Shenyang University of Technology, Shenyang 110870, China
Wang Yiyun	School of Information Science and Engineering, Shenyang University of Technology, Shenyang 110870, China

摘要点击次数: 2570

全文下载次数: 17727

中文摘要:

步态对个人身份进行识别受到越来越多生物识别技术研究者的重视。步态能量图是一种有效的步态表征方法.通过提取步态能量图中的动态区域并利用Gabor小波变换对其特征提取,但经过 Gabor 变换后特征维数较高,必须经过有效的特征融合和选择。由此针对传统的 Gabor 特征提取后存在特征维数较高的缺点,提出了一种基于集成Gabor特征的步态识别方法。首先,采用均值融合和差分二值编码这两种集成方法,对动态区域Gabor特征图进行多尺度和多角度的集成,获得26张集成Gabor特征图;然后从26张集成Gabor特征图中选出4张作为最终的特征向量;最后将特征向量输入KNN分类器进行步态识别。实验结果表明,基于集成Gabor特征的步态识别方法,能够对步态特征进行有效分离和表达,同时降低维数并紧凑表征数据,对步态信息进行正确归类。

英文摘要:

Gait recognition for individual identification has received more and more attention from biometrics researchers. Gait Energy Image is an efficient represent method. Gabor wavelet was used to get magnitude feature of active region of gait energy image, but the images after the Gabor transform generate high dimension feature, which must be processed through effective feature fusion and selection. In order to overcome the shortcoming of high dimension feature of the traditional Gabor feature, a gait recognition method based on integrated Gabor feature is proposed in this paper. Firstly, by means of two integration methods are mean fusion and differential binary encoding, the active region Gabor feature images are integrated in a multi scale and multi angle way and 26 integrated Gabor feature images are obtained, and the 4 images from 26 integrated Gabor feature images are selected as the final feature vector. Finally, the feature vector is input KNN classifier to identify. Experimental results indicate that a gait recognition method based on integrated Gabor feature can separate and express gait features effectively, and reduce dimension and present expression data compactly, meanwhile the expressions are classified correctly.

查看全文查看/发表评论下载PDF阅读器