摘要
提出一种基于稀疏表示的方法,采用CASIA-B和CUSD步态数据库进行步态识别.首先对步态序列中心化及归一化处理,之后提取了步态的主动能量图像(AEI),AEI很好地表达了步态中的动态信息,以此作为步态的特征图像,并对特征AEI采用两种方式稀疏表示:一是采用基于重构误差的方法建立字典、更新字典及分解系数;二是采用基于区分辨别字典的方式建立字典、更新字典及分解系数.系数分解采用的是正交匹配追踪算法.实验证明提出的方法识别准确性高,识别速度快,适合实时性要求高的场合.
A method based on sparse representation for gait recognition was proposed using the CASIA B and CUSD database. The gait silhouette was centralized and normalized first, then the AEI（active energy image） was calculated based on the previous operation and used as the feature image for gait recognition. Two methods were used to establish the dictionary and calculate the decomposition coefficients for the sparse representation, one for reconstruction, and the other for discriminatiorL OMP （orthogonal matching pursuit ） algorithm was used for coefficients decomposition. The result of experiment shows that the proposed method can effectively recognize the gait, and the accuracy of the recognition is high and recognizing speed is much faster.
出处
《东北大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2012年第1期43-46,共4页
Journal of Northeastern University(Natural Science)
基金
国家自然科学基金资助项目(61174145)
关键词
稀疏表示
字典
正交匹配追踪算法
主动能量图像
系数分解
sparse representation
dictionary
OMP （orthogonal matching pursuit）
AEI （active energy image）
coefficient decomposition
