首页 | 本学科首页   官方微博 | 高级检索  
     

基于低分辨率局部二值模式的人脸识别
引用本文:戴金波,肖霄,赵宏伟.基于低分辨率局部二值模式的人脸识别[J].吉林大学学报(工学版),2013,43(2):435-438.
作者姓名:戴金波  肖霄  赵宏伟
作者单位:1. 吉林大学计算机科学与技术学院,长春130012;长春师范学院计算机科学与技术学院,长春130032
2. 吉林大学计算机科学与技术学院,长春,130012
基金项目:国家自然科学基金项目(61101155);吉林省科技发展计划项目(20101504)
摘    要:为提高人脸识别的准确度,提出了一种基于低分辨率局部二值模式的人脸识别方法。该方法将原始人脸图像滤波下采样处理成低分辨率图像,将其划分成若干块矩形块图像,对每一块图像进行局部二值模式计算,统计出每一块LBP图谱的直方图,再连接在一起成为这幅图片的最终特征向量。经实验表明,该算法在ORL和YALE上均取得了更好的识别效果,且对光照、表情、姿势等的变化具备鲁棒性。

关 键 词:计算机应用  局部二值模式  低分辨率  特征提取  人脸识别

Human face recognition based on low resolution local binary pattern
DAI Jin-bo,XIAO Xiao,ZHAO Hong-wei.Human face recognition based on low resolution local binary pattern[J].Journal of Jilin University:Eng and Technol Ed,2013,43(2):435-438.
Authors:DAI Jin-bo  XIAO Xiao  ZHAO Hong-wei
Affiliation:1(1.College of Computer Science and Technology,Jilin University,Changchun 130012,China;2.Department of Computer Science and Technology,Changchun Normal University,Changchun 130032,China)
Abstract:In order to improve the accuracy of human face recognition,a novel Low Resolution Local Binary Pattern(LRLBP) method is proposed.First,the low-resolution image is obtained from the original image through wave filter and down-sampling.Second,the obtained image is divided into non-overlapping rectangular regions and the histogram of each region is computed independently.Finally,these histograms are concatenated as the feature vector of the human face.Experiment results on ORL and Yale face databases show that the proposed LRLBP method can get high human face recognition rate and that the method is robust to illumination,face expression and position variations.
Keywords:computer application  local binary pattern  low resolution  feature extraction  face recognition
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号