Our results illustrated that feature extracted from the averaged silhouettes which in them, the lower part of the body is eliminated are more suitable rather than those extracted from the complete averaged silhouettes. | Journal of Automation and Control Engineering, Vol. 1, No. 2, June 2013 Human Gait Recognition: A Silhouette Based Approach Negin K. Hosseini Faculty of Information Science & Technology, Universiti Kebangsaan Malaysia, Selangor, Malaysia Email: negin62_k@ Md Jan Nordin Center for Artificial Intelligence Technology, Faculty of Information Science & Technology, Universiti Kebangsaan Malaysia, Selangor, Malaysia Email: jan@ Abstract—Human gait has become an important biometric in recent years. A silhouette based method is suggested in this paper, to recognize human in video by their gait. We used averaged silhouette to represent the gait cycle. Principal Component Analysis has been used to reduce the dimensionality of the features. We applied Euclidean distance to measure the similarity of the averaged silhouettes. We implemented the algorithm on the TUMIITKGP Gait Database which has been introduced recently. Although this method is sensitive to the appearance of the subject, it has low computational cost and it is simple. We implemented two experiments on the achieved averaged silhouettes. Our results illustrated that feature extracted from the averaged silhouettes which in them, the lower part of the body is eliminated are more suitable rather than those extracted from the complete averaged silhouettes. Adelson in 1994, their methodology was extracting spatiotemporal features from the subject’s gait for recognition [4]. Afterwards, several studies implemented different gait recognition algorithms [5]-[7]. Gait recognition methods are generally divided into two different categories: model-based and appearance based. Model free gait recognition methods or appearance based methods work directly on the gait sequences. They don’t consider a model for the human body to rebuild human walking steps. They have the advantage of low computational cost in compare with model-based approaches and they also have the disadvantage of sensitivity to cloth and
