Long distance person identification using height measurement and face recognition

This paper presents a long distance person identification system based on height measurement and face recognition using web-camera without zoom-in/zoom-out. In this system, face detection determines if a target person entering the monitoring area or not. We use calibrated camera technique to compensate distortion images. Human height is then measured by the ratio with a reference point. The measured height is utilized to make a screen process so that only a subset of the people database is chosen as candidates for next-stage face recognition. In our experimental results, the proposed architecture has a great improvement in recognition rate compared with the conventional face recognition results in the same environment of long distance.