New approach of gait recognition for human ID

A new gait recognition approach for human ID based on Fourier descriptors is presented in this paper. Body silhouettes are extracted by background subtraction, and cyclic gait analysis is performed to extract key frames. The body contours are processed by Fourier descriptors with data dimension being reduced, and templates are created. The recognition is achieved by nearest neighbor algorithm. We evaluate the proposed gait recognition method on the SOTON gait database, and the experimental results demonstrate that our approach has a good recognition performance.