基于AdaBoost的人脸特征挑选与特征融合算法研究

针对人脸识别中的特征挑选和特征融合问题进行研究。结合已有的基于AdaBoost的人脸特征挑选方法,挑选出最具分类能力的特征,并将挑选出的多类人脸特征在特征层进行融合,得到一个统一的人脸特征用于模式分类。通过在FERET人脸库上的实验表明,其识别方法具有良好的识别效果。

A Facial Feature Selection and Feature Fusion Algorithm Based on AdaBoost

This paper focuses on the study of facial feature selection and feature fusion problem. Combining with the existing facial feature selection method based on AdaBoost, the paper can choose different features which have the best classification ability. Then these features of different kinds will be fused on the feature level to get a normalized facial feature which will be used in the pattern classification. The experiment results on FERET Face Database show that the proposed method improves the recognition p...