Cascade Boosting LBP Feature Based Classifiers for Face Recognition

摘要：

Local Binary Pattern(LBP)is a powerful means of tex-ture description that has achieved great success in faceanalysis area.In this paper,we propose a face recognitionapproach using boosted LBP-feature based classifiers.Themulti-class problem of face recognition is transformed intoa two-class one of intra-and extra-class by classifying ev-ery pair of face image as intra-class or extra-class ones.The cascade framework,is used to overcome the problemof overwhelmingly large number of samples and grosslyimbalance of the positive and negative samples.By boot-strapping negative examples,sub-training spaces (randomsubsets)are randomly generated,and then weak classifiersare learned using every sub-training space(random sub-set).The weak classifiers are combined into a strong one byimproving recognition accuracy.Experimental results onFERET database show competitive performance.

作者: Canming Ma Taizhe Tan Qunsheng Yang

作者单位: Faculty of Computer,Guangdong University of Technology,Guangzhou 510006,China

会议类型: 国际会议

会议名称: 2008 3rd International Conference on Intelligent System and Knowledge Engineering(第三届智能系统与知识工程国际会议)(ISKE 2008)

会议地点: 厦门

会议语种:英文

页码: 1100-1104

在线出版日期: 2008-11-17（万方平台首次上网日期，不代表论文的发表时间）