This paper utilizes the periodicity of swing distances to estimate gait period. It shows good adaptability to low quality silhouette images. Gait moment image (GMI) is implemented based on the estimated gait period. GMI is the gait probability image at each key moment in gait period. It reduces the noise of the silhouettes extracted from low quality videos by gait probability distribution at each key moment. Moment deviation image (MDI) is generated by using silhouette images and GMIs. As a good complement of gait energy image (GEI), MDI provides more motion features than the basic GEI. MDI is utilized together with GEI to represent a subject. The nearest neighbor classifier is adopted to recognize subjects. The proposed algorithm is evaluated on the USF gait database, and the performance is compared with the baseline algorithm and two other algorithms. Experimental results show that this algorithm achieves a higher total recognition rate than the other algorithms.