一种改进的AdaBoost算法在人脸检测系统中的应用

针对复杂背景条件下人脸检测的检测率低、速度慢的问题,提出了一种改进的AdaBoost算法,与遗传算法相结合,产生了一种识别率高、泛化能力好的强分类器,文中称之为GA-AdaBoost算法。该算法首先训练多个支持向量机作为弱分类器,然后用AdaBoost算法将多个弱分类器组合成一个强分类器,在组合的同时采用遗传算法对各弱分类器的权值进行全局寻优。最后,通过试验与传统AdaBoost进行对比,表明了该算法具有识别率高和速度快的优越性。

Application of Improved AdaBoost Algorithm in Face Detection System

To solve the problem that the face detection is slow and its rate is low in complicated background,this paper presents an algorithm named GA-AdaBoost,using genetic algorithm to improve the performance of AdaBoost,has advantages of higher identification rate and better generalization performance.The algorithm first trains some support vector machines as weak classifiers,and then uses AdaBoost algorithm to embody the weak classifiers into a strong classifier,while using genetic algorithm tooptimize weights of weak classifiers for global optimization.Experimental result demonstrates that GA-AdaBoost achieved better generalization performance and higher identification rate than the existing AdaBoost methods.