Mayur Popade, Anita Thengade
Abstract: Biometrics is a dynamic industry resulting in sustained evolution of numerous new technologies. The biometric system is needed in many key areas such as banking, airport, criminal cases, security purpose etc. Gait is an important biometric technology to recognize a human by the manner, they walk. Biometric identification based on body movements as gait recognition has gained new motivation over the past few years due to the launch of low cost depth cameras. Here, we propose a model to identify the human without coordination using gait images. In this paper we propose the biometric system to distinguish human from others through gait images based on principal component analysis algorithm and K nearest neighbor classifier. Principal component analysis is used to extract gait features. K nearest neighbor classifier is used to compute human id. Experimental analyses on CASIA A dataset show a significant performance gain in terms of accuracy. We create our own dataset for experiment using camera that provide a good view of entire human body. Furthermore, proposed approach has been experimented on our own dataset.
Keywords: Gait recognition, PCA, KNN, CASIA dataset