基于光照归一化分块完备LBP特征的人脸识别  

Face recognition based on illumination normalization and block-based completed local binary pattern

作　　者：周巍[1] 程勇[2] 曹雪虹[1,2] 

机构地区：[1]南京邮电大学通信与信息工程学院,南京210003 [2]南京工程学院通信工程学院,南京211167

出　　处：《计算机工程与应用》2015年第11期145-149,共5页Computer Engineering and Applications

基　　金：江苏省自然科学基金(No.BK20131342)

摘　　要：针对复杂光照条件下的人脸识别,提出了一种基于光照归一化分块完备局部二值模式(B-CLBP)特征的人脸识别算法。该方法对人脸图像进行光照归一化预处理,对处理后的人脸图像进行B-CLBP特征提取,融合成B-CLBP直方图,根据最近邻准则进行分类识别。在Extended Yale B人脸库上的实验结果表明,所提算法可以有效提高复杂光照条件下的人脸识别率。Aiming at the problem of face recognition under complex illumination, an effective face recognition method based on illumination normalization and block-based completed local binary pattern is proposed. The method performs illumination normalization on face images, then extracts the B-CLBP features from the processed face images in order to form the B-CLBP histogram, applies the nearest neighbor principle for face recognition. The experimental results on Extended Yale B face databases demonstrate that the proposed method can achieve significant recognition rate under complex illumination conditions.

关 键 词：人脸识别 光照归一化 分块完备局部二值模式 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]