Gait Recognition Based On Body Silhouette

As enhance of the security requirement in the modern society, Biometrics Recognition, as an approach for society security, has been greatly researched and developed.Early psychological studies into gait suggested that gait was a unique personal characteristic. People can identify persons by observing their walking. Gait Recognition research involves in following areas:gait detection, gait representation, gait recognition and gait database. In this paper, the fore three areas are mainly researched.On the gait extraction, a median filter model is used to subtract the background of the scene; then use the background subtraction method to obtain gait body area, combined with automatic histogram threshold segmentation and morphological algorithm to eliminate noise interference. Finally, neighborhood boundaries tracking algorithm is used to compute the coordinates of the body contours.On the gait description, for each image sequence, cyclic width of gait analysis is performed to extract key frames, then extract and compute the gait contour to make the eigenvector.A wavelet boundary descriptor based on boundary-centroid distance is proposed to improve the efficiency of exciting methods. This descriptor is invariant to shape translation scaling and rotation. Furthermore, it can represent shape very intuitively, similar to people's feeling. In addition, it can be used to extract feature under multi-resolution. Therefore, the feature can describe gait very well for the realization of gait recognition.On the gait recognition, a new recognition method based on wavelet boundary descriptor and support vector machine (SVM) is presented. The feature used here to represent shape is the wavelet boundary descriptor mentioned above. With this descriptor, a new recognition framework based on SVM is constructed to realize real-time object-efficient gait recognition.Finally, all kinds of the approaches are realized by programming. We present a Gait Recognition System developed by ourselves. It is a platform for new algorithm research and test.