摘要： 提出了一种新的正面人脸检测算法。该方法利用经验模式分解和匹配追踪算法来提取人脸特征，训练Bayes分类器来进行分类判决。在FERET人脸库中与特征脸(Eigenfaces)方法进行了比较，实验结果表明，该算法的计算效率和检测精度均优于特征脸方法。

关键词: 人脸检测, 经验模式分解, 匹配追踪算法, Bayes分类器

Abstract: A novel method for frontal face detection is presented. EMD (empirical mode decomposition) and matching pursuit algorithm are used for face feature extraction, and the Bayes classifier is trained for classification. The proposed method is compared with eigenfaces method on FERET face database. Experimental results demonstrate that the method has lower computational complexity and higher accuracy than Eigenfaces method.

Key words: face detection, empirical mode decomposition(EMD), matching pursuit algorithm, Bayes classifier

中图分类号: 

• TP301.6