The Research Of Gait Recognition

With the developpment of technology, modern society has more requirements on the accuracy, security and practicability of human recognition, so biometrics recognition technology has been growing rapidly. Human gait recognition is one of these techniques. Gait recognition is the process of identifying individuals by their walking manners. This thesis researchs on gait recognition based on Hidden Markov Models (HMM).In this thesis, the current situation of gait recognition is discussed, including the main issues and the prospect in this field. HMM is the important theory of this thesis and discussed in detail. Generally, gait recognition consists of three parts: preprocessing of gait sequences, feature extraction and classification. This thesis focus on the Feature extraction.In the feature extraction step, this thesis get human figure through preprocessing (including background construction, foreground detection, morphology processing) of gait sequences, and choose the distance between edge of human figure and its centroid and the longest step as gait feature; This thesis choose 5 key frames of a gait cycle, and use primary component analyse (PCA) to further reduce the observation vector dimension. A continuous HMM for every person can be constructed by training these feature sequences. Finally, in the classification step, the research calculate the output probability of the unknown gait sequences to all HMM and choose the person corresponding to maximum output probability as its classification identification.This method captures human gait structure and the transitional information, the HMM statistical property decides that it has very good robust in the gait description and recognition, decrease the affection of self-occlusion and shadow. At last, experiments show that HMM is an effective method in gait recognition field and has a broad application prospect.