Human Face Location And Track In Video Image Sequences

Face detecting and tracking is the first step of the face information processing. In this paper, studies on automatic face tracking in video image sequences are carried out. A systematic summary of previous work is given first. Then this paper introduces a face detection algorithm in static image under complex backgrounds into video image sequences. First of all, a determination criterion of eye location is established by the priori knowledge of geometrical facial features. Secondly, a range of threshold values that would separate estimated from the facial image histogram. Thirdly, with the progress increase of the threshold, whether the eye blocks appear is determined by the criterion of eye location. At last, the 2-D correlation coefficient is used as a symmetry similarity measure to check the factuality of the two detected eyes. This paper also combines skin color information to improve the veracity of eye location. Experimental results demonstrate that the detection algorithm is efficient under complex background in video image sequences. Finally, this paper also presents a new method for face tracking, which is based on the motion prediction and histogram match, and based on skin color model verification. The tracking method deals with center moment matching in human eyes image, motion prediction to the position of the search region, verification of the matching result using Mahalanobis confidence measurement, and the skin color model could be refreshed along with target facial change. This face tracking system enables the tracking of face has a better tracking effect, and the track speed of the system reaches 12～19 frame per second.