Design And Implementation Of Intelligent Monitoring System Based On Face Detection And Tracking

Face detection and face tracking are widely applied in the field of video surveillance, human computer interaction, feature recognition, content based image retrieval, image coding, etc. This paper studied on Face detection, face tracking, and their application on the video surveillance system. The typical face detection and face tracking algorithms were discussed, and and an smart video surveillance system was developed using the improved face detection and face tracking techniques.First, an improved adaptive ellipse skin color model was developed to enhance the skin color detection ability under abnormal lighting condition.Second, an adaptive background updating strategy was proposed to improve the effect of background substration.Thid, motion and color information was cooperated into the AdaBoost face detection algorithm to implement an multi-feature fusion based real-time face detector. Experimental results had shown the that the proposed algorithm can improve face detection ability and processing speed.Fourth, target eliminating and multi-acessory information were introduced into the Mean Shift face tracking algorithm to solve occlusion and the individual corresponding problems in multi-face tracking.Based on aforementioned techniques, an smart surveillance system for banking business was realized in this paper. Experimental results had shown the good performance of the proposed system.