摘要
光照变化条件下的人脸图像识别一直以来都是图像处理中的热点和难点问题,为了提高人脸图像的识别率,提出了一种用于非均匀光照条件下人脸识别的算法。利用对数及二维小波变换的多尺度特性提取出人脸的光照不变量,然后运用PCA+LDA方法进行人脸特征提取,并采用基于欧氏距离的最近邻分类器进行识别。通过Matlab编程实验,在YaleB人脸库中达到了较高的识别率。
Illumination variation is the most significant factor affecting the performance of face recognition, and has received much attention in recent years. A novel method to extract illumination invariant features was proposed for face recognition under varying lighting conditions, which combined wavelet transform and logarithm operation. The experiment with Matlab programs was designed on Yale B face database by PCA + LDA recognition. Minimum distance classifier was applied for its simplicity; the Euclidean metric L2 was used as distance measure. The experimental results show that the performance of the proposed method is better than other methods.
出处
《计算机应用》
CSCD
北大核心
2009年第9期2395-2397,共3页
journal of Computer Applications
基金
国家自然科学基金资助项目(60873092)
教育部新世纪优秀人才基金资助项目(NCET-06-0762)
教育部高等学校博士学科点专项科研基金资助项目(20060611009)
重庆市自然科学基金重点资助项目(CSTC2007BA2003)
关键词
人脸识别
光照不变量
小波变换
多尺度分析
欧氏距离
face recognition
illumination invariance
wavelet transform
muhiresulotion analysis
Euclidean metric