摘要
我国传统LBP人脸识别技术只是针对局部信息进行识别而忽略全局信息,在特殊环境中,当人脸受到光线、背景变化等原因而造成模糊现象时,只采用局部识别会导致识别效率和准确度降低。对此,文章提出一种基于LBP特征集成学习的人脸识别算法C-MB-LBP,该算法主要是将人体的脸部图像进行分块并得出LBP特征,根据中心像素以及分块的灰度值进行计算,得到新的LBP特征,最后再利用近邻分类器对其特征进行识别,从而识别人脸。实验结果表明,文章所提算法对于人脸识别不再仅限于局部的识别,更注重全局识别,使人脸识别效率显著提高,识别时间大大降低。
China’s traditional LBP face recognition technology only recognizes local information,thesewilllower recognition efficiency and accuracyin a special environmentwhenlights and backgroundsare changed. This paper is thus motivated to propose a face recognition algorithm based on LBP feature integrated learning technology. Firstly, this algorithm is mainly to block the human face image, obtain LBP features. Then it calculates new LBP featuresaccording to the central pixel and the gray value.Finally, the algorithmrecognizes the facefeaturesin terms of the neighbor classifier.Experimental results show that the proposed algorithm is no longer limited to local recognition, but pays more attention to global recognition, which significantly improves the efficiency of face recognition and greatly reduces the recognition time.
作者
胡念
张四平
王梅
Hu Nian;Zhang Siping;Wang Mei(School of software,Hunan College of Information,Changsha 410200,China)
出处
《信息通信》
2020年第8期38-40,共3页
Information & Communications
基金
长沙市社科规划一般项目“基于‘互联网+’推进智慧医疗建设研究”,(项目编号:2020csskktzc63)
湖南省教育厅科学研究项目“基于云计算人脸表情特征提取与识别的研究”,(项目编号:XJK016CXX005)。
关键词
LBP
特征集成学习
人脸识别技术
LBP
Feature integrated learning
Face recognition technology