基于Gamma和AdaBoost的人脸检测系统设计

Design of face detection system based on Gamma and AdaBoost

为了有效地实现人脸的检测效果,文章在AdaBoost算法基础上提出一个改进的人脸检测算法.为了有效地消除光照和成像对人脸的影响,该算法将Canny修剪算法和伽马矫正算法进行结合,有效地消除光照和成像设备对人脸的影响.并利用Visual C++和OpenCV等开发工具设计了一个人脸检测系统.本系统采用20×20的人脸图像和背景图像各1 000张训练了一个7层的级联分类器,每一层构成的强分类器由一组基于Haar特征的弱分类器构成.该系统通过自选137幅包含人脸和背景的图片对系统进行测试,获得94.72%的正确检测率以及26.42%的误检率.

In order to achieve the face detection effectively, this paper proposed an improved algorithm of face detection based on the AdaBoost algorithm. The algorithm effectively combined Canny pruning and Gamma cor- rection algorithm, and effectively eliminated the influence of illumination and imaging equipment of face. And it uses the development tools such as Visual C ＋ ＋ and OpenCV to design a face detection system. The system u- ses 1 000 images with face and 1 000 background images to train a seven-stage cascade classifier, the size of these images are 20 - 20. Each classifier is a strong classifier and each strong classifier is composed of a set of weak classifiers based on the Haar feature. The system is tested by 137 images of faces and background, and the result is that the system gets the correct detection rate of 94.72%, and the error detection rate of 26. 42%.