Face Detection Method Research and Implementation Based on Adaboost

This paper describes and deduces the theory of Haar-like features, Integral image and AdaBoost algorithm, which were proposed by Paul Viola, and then researches its improvement. We combine Microsoft Visual C++6.0 with OpenCV Function library to develop the software, and achieve the function of real-time face detection. According to experimental results, we can conclude that the improved algorithm can detect the real-time face better, having higher success rate and lower false alarm rate. Meanwhile, with analyzing the experimental process and the results, this paper proposes using Skin-Color model to integrate the AdaBoost algorithm, and then using subsidiary discriminant function to improve the detection process.