A Comparative Evaluation Study of Automated Gait Recognition based on Spatiotemporal Feature and Different Neural Network Classifiers

New areas of applications such as: human-computer interaction, access control, surveillance, activity monitoring and clinical analysis depend on hepatic technology. Gait analysis has been explored thoroughly during the last decade as a behavioral biometric feature which doesn't require subject interaction. In this paper, persons can be recognized from their gait regardless of the angle of walking seen. The performance of four artificial neural networks (ANNs) based classifiers was evaluated and tested, based on spatiotemporal features. The results show that discrete wavelet transforms and support vector machine recognition technique provides a recognition rates up to 94%. Moreover, it is characterized by speed and accuracy compared with other classifiers.