Abstract: A hidden Markov model (HMM) human recognition algorithm based on gait energy image (GEI) is presented.First a preprocess technique is used to segment the moving silhouettes from the walking figure.The algorithm obtains the gait quasi-periodicity through analyzing the width information of the lower limbs′ gait contour edge, and the mean GEI is calculated from gait periodic.K-means clustering algorithm is extracted observation sequence from GEI.Then the observation sequence is modeled by using 1-D discrete HMM, and the model is used for human recognition.The proposed algorithm is evaluated on USF Gait Database.Experimental results show that the proposed approach is valid and has encouraging recognition performance.