基于改进的局部二值模式和SVM的人脸识别  

Face recognition based on improved local binary mode and SVM

作　　者：庙传杰 史东承[1] MIAO Chuanjie;SHI Dongcheng(School of Computer Science & Engineering, Changchun University of Technology, Changchun 130012, China)

机构地区：[1]长春工业大学计算机科学与工程学院,吉林长春130012

出　　处：《长春工业大学学报》2020年第3期257-262,共6页Journal of Changchun University of Technology

基　　金：吉林省科技攻关计划重点项目(20150204020SF)。

摘　　要：先提取人脸图像标准的LBP特征,再分块提取且确定出相应的加权MB-LBP特征。通过支持向量机分别处理这两种特征,确定出相应的投票结果矩阵。基于一定的权重比对提取的两种特征加权融合,识别结果即融合矩阵最大值。选择ORL和AR人脸库检测了该方法的性能。The standard Local Binary Pattern(LBP)feature of face image is extracted,and then the weighted MB-LBP feature is extracted by face division.Support vector machine is used to process the two features for determining the vote matrix.The two features are fused with different weight and the recognition result is the fused matrix max value.ORL and AR face data library is applied to test the method.

关 键 词：LBP算子 人脸识别 特征提取 加权融合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]