Gait Recognition via Flexible Measurement

We present a new gait identification method based on dynamic time warping (DTW), as video surveillance system requires high accuracy and precision. It could reduce computational cost of gait recognition, significantly improve the recognition rate for gait and meet the demand of video surveillance. The characters of human appearance have been utilized to extract entire binary image of human silhouette from video sequences, herein the centroid of silhouette is obtained. As the contour is unfolded clockwise, every video sequence could be converted to a normalized 1D (one-dimension) distance. The DTW-based distance between probe sequences and standard ones has been calculated to compare with the threshold. Finally, the similarity is induced to identify human gait by DTW distance. The average accuracy improves to 92.8% in gait recognition. Therefore, it is applicable for video surveillance in different scenarios.