基于遗传算法和AdaboostSVM的人脸检测算法

Face Detection Based on Genetic Algorithms and AdaboostSVM Algorithm

将遗传算法用于以SVM为弱分类器的Adaboost人脸检测算法。先根据样本训练集中的人脸和非人脸样本训练出弱分类器SVM,然后利用Adaboost算法将多个SVM弱分类器级联组合成一个强分类器,并且在组合的过程中采用遗传算法对各个弱分类器权值进行全局寻优,最终得到检测结果。通过与传统Adaboost以及AdaboostSVM进行对比试验,表明论文方法具有更高的检测效果。

An algorithm using genetic algorithm to improve the face detection in Adaboost with SVM based weak classifiers is proposed. Firstly,the method trains the weak classifier of support vector machine （ SVM） according to human face samples and non-face samples in the training sample set, then uses Adaboost algorithm to embody the weak classifiers into a strong classifier, while using genetic algorithm to optimize weights of weak classifiers for global optimization, and f in a l detection result is given. Experimen-tal result demonstrates that GA- AdaboostSVM achieved better detection performance than the traditional Adaboost and AdaboostS- VM methods.