Abstract—In this paper considering a new human gait recognition system based on Radon transform which gives a high precision recognition rate. Innovation of this paper allocate to feature extraction and usage of them during process by combined neural networks . feature extraction is based on the Radon transform of binary silhouettes .in this paper For each gait sequence, the transformed silhouettes are used after background estimation and human detection in the scene to make each related template’s. then set of all templates is used to subspace projection by following PCA method and earning final decimated feature vector for each persons in database. consequently earned feature vector for each person's is applied to multilayer perceptron neural network and set of all neural networks feed to final neural network for final decision . Experimental results is performed over a suitable data base include 10 samples for ten person which each sample have 130 frames approximately. 97% recognition rate of the proposed system is obtained over 10 samples test patterns.

Index Terms—Human Gait recognition,Background estimation, Radon Transform,Y-T plan,X-T plan , Feature extraction,PCA, Neural network classifier.

Manuscript received June, 2010. S.M.Hosseini is with the Islamic Azad University Of Firoozkooh , electrical engineering group , iran. Phone: +989127756374: m.hosseini@Birjand.ac.ir .hasan farsi is with university of birjand,iran,hfarsi@birjand.ac.ir. Peyman nouri is with Islamic azad university of Tabriz, payman_nouri@ya and Abbas Narsabadi is with the tarbiaat moaalem University of sabzevar,iran. He is now with the Department of Mechatronics engineering, abbas.nasrabadi@gmail.com.

Cite: Seyyed Meysam Hosseini, Abbas Nasrabadi, Peyman Nouri and Hasan Farsi, "A Novel Human Gait Recognition System," International Journal of Computer and Electrical Engineering vol. 2, no. 6, pp. 1043-1047, 2010.