Journal of Hebei University of Water Resources and Electric Engineering

Journal of Hebei University of Water Resources and Electric Engineering ›› 2020, Vol. 30 ›› Issue (4): 38-43.DOI: 10.16046/j.cnki.issn2096-5680.2020.04.006

• Technology Theory and Application • Previous Articles     Next Articles

Face Recognition Based on K-L Transform and Singular Value Decomposition

WANG Yin-hua1，2   

  1. 1.Department of Electrical Engineering,Tongling College,244000,Tongling,Anhui,China;
    2.Engineering Technology Research Center of Optoelectronic Technology Appliance,AnHui Province,244000,Tongling,Anhui,China

  • Received:2020-07-20 Revised:2020-09-11 Online:2020-12-31 Published:2021-01-30

基于K-L变换和奇异值分解的人脸识别

王银花1，2   

  1. 1.铜陵学院电气工程学院，安徽省铜陵市翠湖四路1335号 244000;
    2.光电子应用安徽省工程技术研究中心，安徽省铜陵市翠湖四路1335号 244000

  • 作者简介:王银花(1977-)，女，汉族，安徽无为人，副教授，工学硕士｡研究方向:智能信号处理｡E-mail:wyhahu@tlu.edu.cn
  • 基金资助:
    安徽省高校自然科学研究重点项目(KJ2017A472)

PDF

Abstract: Face classification is a hot research issue in the field of face image processing.In order to get agood face recognition effect,a face recognition method based on K-L transform and singular value decomposition is proposed.Firstly,K-L transform is used to train the face features in the face database to reduce the size of training samples.Secondly,a face training database and a test database are constructed.Finally,the sample class with the smallest distance is selected as the recognition result used by Euclidean distance method.The experimental results show that the face recognition accuracy of this method is high.

Key words: face processing, K-L transform original feature, singular value decomposition

摘要： 人脸识别是当今人脸图像处理领域的研究热点。为了得到良好的人脸识别效果,提出基于K-L变换和奇异值分解的人脸识别方法。首先采用K-L变换对人脸数据库中的人脸特征进行训练,以减少训练样本的规模,其次构建人脸训练数据库和测试数据库,最后采用欧式距离法选取距离最小的样本类别作为识别结果。实验结果表明,该方法人脸识别正确率较高。

关键词:  , 人脸识别, K-L变换, 奇异值分解

CLC Number: