The Study Of Face Detection Technology

Face detection find faces information in the video sequence or image and determine facial size, position, trajectory, attitude, and further human face eyes, lips and other features extraction process. Face detection come from face recognition, it is the first stage of face recognition system, is also a critical stage. In recent years, face detection and recognition technology has become the most promising means of authentication. In the e-commerce, health care, tracking and other areas of its practical application face detection have become increasingly prominent.Firstly this paper introduces the face detection technology research background and significance, and the current number of commonly used face detection algorithm. Summarize the face detection algorithm history and current status.Secondly, researched skin-based face detection method and Adaboost-based face detection technology in-depth, analysis and summarized their features, and compared their advantages and disadvantages. Proposed a Adaboost face detection algorithm that base on skin segmentation:firstly make skin segmentation by an adaptive threshold selection method in YCgCr color space. And make rectangle feature calculation, classifier training in Adaboost algorithm and so on. Make Adaboost face detection of skin segmentation candidate region, obtained face region.Thirdly, make face detection in skin detection, Adaboost detection and the algorithm of this paper, for a large number of images. And statistical analysis of the test to show the advantages of this algorithm.Finally, this study made a comprehensive summary of the work, and concluded, and made the outlook for further work.