基于AdaBoost算法的快速人脸检测研究

Fast face detection based on the AdaBoost algorithm

研究了AdaBoost算法的原理和训练过程,分析了导致算法准确性下降的原因.使用高斯混合模型对人脸肤色建模,将可能存在人脸的区域从彩色图像中分离出来,再使用AdaBoost算法对该区域进行检测能够提高检测速度和准确率.实验结果验证了算法的准确性.

This paper presents the theory of the AdaBoost algorithm, and analyzes the causes for the decrease of its accuracy. By using Gaussian Mixture Model（GMM） to simulate the skin color model, separating the region of the possible human face from the whole image and scanning this region by using the AdaBoost algorithm, the experi- mental results show that this approach can improve the detection speed and accuracy of the system.