基于二维线性判别分析的彩色人脸识别  

Color Face Recognition Based on 2D Linear Discriminant Analysis

作　　者：白晓明[1] 王成章[2] 石勤[3] 

机构地区：[1]首都经济贸易大学信息学院,北京100070 [2]中央财经大学应用数学学院,北京100081 [3]北京工业大学计算机学院,北京100124

出　　处：《北京工业大学学报》2010年第12期1717-1721,共5页Journal of Beijing University of Technology

基　　金：国家自然科学基金资助项目(10801004);北京市属高等学校人才强教深化计划资助项目(RHR201007117);首都经济贸易大学校级重点资助项目(2010XJZ002)

摘　　要：为了提高人脸识别的准确率,提出了一种基于二维线性判别分析的彩色人脸识别算法,直接对彩色人脸图像不同通道的颜色信息进行编码,采用矩阵表示模型描述人脸对象;融合人脸的类别特征,采用二维线性判别分析的算法提取彩色人脸的分类特征;根据投影后的特征矩阵,采用最近邻分类的算法进行人脸识别.利用CVL和CMU PIE彩色人脸数据库进行实验.结果表明,本文提出的彩色人脸识别算法能有效地提高人脸识别的准确率,明显优于对比的算法.A novel recognition approach coined two-dimension linear discriminant analysis based color face recognition is proposed in this paper to improve the accuracy of face recognition.A matrix-representation model which encodes the color information directly is employed to represent the color face image.Based on the representation model,two-dimension linear discriminant analysis algorithm is exploited to extract facial features for recognition task.The nearest neighborhood classifier is adopted to identify the color face samples based on the facial feature matrix.Experimental results from CVL and CMU PIE color face databases verify the effectiveness of the proposed approach.

关 键 词：人脸识别 线性判别分析 主成分分析 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]