Human gait identity recognition system based on gait pal and pal entropy (GPPE) and distances features fusion

Gait recognition is one of the most promising modality for human identification in motion and without subject cooperation. This modality of recognition has the objective to identify humans by the manner of walking on foot, even if gait sequences are captured at a distance with low-quality image. A new identity recognition method using gait will be proposed on this paper. The Gait Pal and Pal Entropy (GPPE) image was generated and merged with four proposed distances. The fusion of features based images Gait Pal and Pal Entropy Image (GPPE) and features based distances represent the gait signature in this work. The feed-forward neural network was used for features classification. CASIA-B dataset is used in order to perform the proposed method. A better performance of this method compared to the related work methods has been shown after experimental result in terms of rate accuracy of identification and classification.