Researches On Gait Recognition Algorithms And System

Gait Recognition has an important value for theoretical study and practical application as a new biometric technology. As an ascendant field of study, there are many study points and various kinds of innovations and improvements to be worked. The two major tasks of this paper are: first, improving the gait recognition algorithms; second, building a gait recognition system.On aspect of algorithms research, the work of this paper includes: First, to solve the conflict of the current gait recognition algorithm either too complicated or the recognition rate is not high, a new rapid gait recognition method based on the silhouette is proposed. Second, the sub-space feature extraction method is applied in gait recognition, and the research focus on the algorithm of Locality Preserving Projections (LPP), the discriminant informations are introduced to the two-dimensional LPP for improvement. Compared to algorithms of OLPP, DLPP, ODLPP, our mothod avoids the loss of local information, as well as the problem of singular value. Third, the research on view-insensitive gait recognition is done here. Rules of affections on recognition performance from various views are studied and summarized through two-dimensional geometric shapes analysis. The decision-fusion method is proposed to solve the gait recognition in view invariants. All of the algorithms obtain the satisfactory results and encouraging recognition performance with the experimental verification and comparative analysis.On aspect of system construction, we also construct a gait recognition system based on CASIA gait database which has been already built by Institute of Automation Chinese Academy of Sciences. System is designed to be clear, simple, fast and effective. The three algorithms in this paper are programmed in this system and compared with a number of other algorithms which are used commonly. This system will provide a platform for further research and provide some help on experimental stage.