摘要
提出基于二维Gabor小波特征(Gaborface)矩阵的人脸表征方法,二维Gaborface矩阵有别于常规的一维采样特征矢量方法。对该方法的2种运用方式:整体Gaborface表征(EGFR)和多通道Gaborface表征(MGFR)进行了研究,在ORL数据库中采用二维PCA人脸识别算法进行了实验。对比实验结果表明了该方法的有效性和可行性,特别是基于MGFR的2DPCA方法实现了100%的识别率。
This paper introduces a novel face representation method based on two-dimensional Gabor wavelet feature (Gaborface) matrices instead of the conventional transformed one dimensional feature vectors. Both the ensemble Gaborface representation and the multi-channel Gaborface representation are studied and experimented on the ORL face database using two-dimensional principal component analysis face recognition approach (2DPCA). Experimental results show the efficiency and feasibility of the method, particularly, the 100% recognition rate is achieved by the MGFR-based 2DPCA.
出处
《计算机工程》
CAS
CSCD
北大核心
2008年第2期8-10,共3页
Computer Engineering
基金
中国科学院"百人计划"择优支持基金资助项目"生物特征识别技术研究与应用"
上海浦江人才计划基金资助项目"生物特征身份证照技术研究与应用"(05PJ14111)
关键词
人脸识别
GABOR小波特征
2DPCA算法
特征提取
多通道
决策级融合
face recognition
Gaborface
Two-dimension Principal Component Analysis(2DPCA)
feature extraction
multi-channel
decisionlevel fusion