一种基于人脸分割的PCA和SVM人脸识别方法

PCA and SVM face recognition method based on human face segmentation

基于PCA和SVM提出了一种新的人脸分割法,将双眼、额头、鼻子、嘴等可以明显表征人脸的六类特征提取出来,舍弃双颊以及耳朵等特征量少的部位。融合上述人脸的特征识别结果,运用支持向量机的方法进行分类识别,实验结果表明,文章所提出的PCA与SVM融合的新的人脸分割方法能有效地对人脸进行分类,极大地提升识别率。

In this paper,a new face segmentation method is presented based on PCA and SVM. Eyes,forehead,nose,mouth and so on which can characterize face feature is extracted,cheeks and ears such as fewer characteristics parts are abandoned. Using the facial feature recognition results and the method of support vector machine（ SVM） to identify human face. The experimental results show that the proposed method can effectively classify human faces,greatly improve recognition rate.