步态能量图和KFDA的步态识别研究  

Gait recognition based on GEI and KFDA

作　　者：王竣[1] 王修晖[1] 

机构地区：[1]中国计量大学信息工程学院,浙江杭州310018

出　　处：《中国计量学院学报》2016年第2期216-222,共7页Journal of China Jiliang University

基　　金：国家自然科学基金资助项目(No.61303146)

摘　　要：为了有效地获取步态连续性的动态特征,快速准确地进行身份识别.特提出了一个基于步态能量图(Gait Energy Image,GEI)和核Fisher判别分析(Kernel-based Fisher Discrimination Analysis,KFDA)的分类识别算法.算法首先以步态能量图(GEI)按列向量作为输入,求得最优子空间W_(opt)和α_(opt).利用提取步态能量图(GEI)的步态信息向量计算在α_(opt)上的投影,并计算其投影轨迹.在分类阶段,采用最近邻分类器(Nearest neighbor classifier).最终在中科院自动化研究所CASIA B步态数据库上进行实验,对比多项式、高斯径向基核函数和其他四种算法的结果显示,本文算法取得了较高的识别率.To effectively captures the dynamic features of the gait and accelerate the authentication and identification, a novel gait recognition algorithm was presented and a gait recognition algorithm based on GEI （gait energy image） and KFDA （kernel-based fisher discrimination analysis） was introduced. Firstly, the vector of the gait energy image（GEI） was used as the input to get the optimal subspace Wopt and αopt. Then the extracted GEI vectors were used to compute the projection on αopt with its projected path calculated. The nearest neighbor classifier was used for the classification. This method was evaluated on the CASIA B gait database. The comparison of the polynomial, the Gaussian kernel function and other four algorithms shows that the proposed method can obtain stable classification and performs satisfactorv recognition results_

关 键 词：步态能量图 核FISHER判别分析 多项式核函数 高斯径向基核函数 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]