Gait Recognition Algorithm Based On The Fusion Using Body Dynamic And Static Feature

With the development of society, the security problem becomes more and more important. The gait recognition, a new biometric technology, aims to identify individuals by extracting the change characteristics of individuals from their different walk ways. Comparing to other biological characteristics, the gait recognition is non-contact, easily acquired at a distance, difficult to concealment and can be carried out to detect and identify by the low-resolution image sequences. To operate successfully, the established biometric such as face, fingerprint usually require proximal sensing or physical contact. However, the gait the only one biometric that can be easily perceived at a distance. So, from the view of video surveillance, the gait recognition is the biometric with the most potential in the case of long-distance.The gait recognition includes moving target detection, feature extraction and classification and recognition. Based on the study of the various of the gait recognition algorithms, this thesis proposes a gait recognition algorithm based on the fusion of static feature and dynamic feature and makes a detail research on the three parts.Because of video image sequences with plain background, the thesis analysises the moving target detection methods, extracting the image based on the improved background subtraction algorithm, removing noise based on morphology and extracting the moving silhouettes of walking figure by the Fourier descriptor as the static feature. The thesis calculates the stride frequency and the stride as the dynamic features in a gait cycle and fuses the human silhouettes feature, the stride and stride frequency. The thesis analysises addition fusion, minimum value fusion, maximum value fusion, the Choquet fuzzy integral fusion and proposes an addition fusion rules based on the weight. Finally, the decision-making uses the nearest neighbor fuzzy classifier. The thesis designs the simulation experiment on the CASIA gait database. The experimental results demonstrate that the recognition has been good improvement after fusion and is much better than those based on single feature.