首页 | 本学科首页   官方微博 | 高级检索  
     检索      

一种基于小波和DCT的人脸识别算法
引用本文:杨绍华.一种基于小波和DCT的人脸识别算法[J].宁夏师范学院学报,2007,28(6):49-53.
作者姓名:杨绍华
作者单位:宁夏大学,数学计算机学院,宁夏,银川,750021
摘    要:提出了一种基于小波变换和离散余弦变换(Discrete Cosine Transform DCT)的人脸识别方法,它通过对经过小波变换的人脸图像的低频图像进行DCT变换,从而获得原始图像的频谱图像,然后利用最近邻法进行分类,在ORL人脸数据库上的实验表明这样的方法能够获得比传统主成分分析(Principal Components Analysis PCA)更好的识别性.

关 键 词:人脸识别  小波变换  离散余弦变换
文章编号:1674-1331(2007)06-0049-05
收稿时间:2007-06-25

Human Face Recognition Method Based on Wavelet Transform and DCT
YANG Shaohua.Human Face Recognition Method Based on Wavelet Transform and DCT[J].Journal of Ningxia Teachers College,2007,28(6):49-53.
Authors:YANG Shaohua
Abstract:The algorithm of face recognition based on Wavelet Transform and Discrete Cosine Trans- form was proposed in this paper.First the original image was decomposed by wavelet transform,then, the DCT method can be directly used to the low frequent sub-images obtained from the previous step.In the end,the classical K-Nearest Neighbor (KNN) method is performed on the reductions for the pattern classification.The experiment based on ORL-DATABASE shows that the above-mentioned algorithm performs faster training speed and higher accuracy of recognition than traditional PCA method.
Keywords:Face recognition  Wavelet transform  Discrete cosine transform(DCT)
本文献已被 CNKI 维普 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