摘要
提出了一种基于Log-Gabor滤波和局部二值模式(local binary patterns,LBP)算子的光照不变人脸识别方法.该方法首先对人脸图像进行对数变换预处理,有效改善剧烈光照变化对人脸图像的不利影响.然后采用Log-Gabor滤波器与图像进行卷积,得到不同尺度和不同方向下的人脸Log-Gabor特征图像.在此基础上,再使用LBP算子对Log-Gabor图像进行描述,最后将所有的Log-Gabor图像的LBP特征进行简单连接,作为人脸的特征向量.将所提出的方法在YaleB数据库上进行实验,实验结果表明该方法能够有效提高复杂光照条件下的人脸识别率.
An illumination invariant face recognition method based on Log Gabor filters and local binary patterns（LBP） descriptor is proposed.First,in order to reduce influences caused by intense illumination changes,we have used the logarithmic transformation to preprocess face images.Then we use Log Gabor filters to convolve with each face image to abtain local features under different scales and orientations.finally,by treating Log Gabor face images of different scales and orientations as input,we apply Local Binary Patterns operators to each of them to acquire histogram features,and concatenate all features to form feature vectors for the following classification.Experimental results on Yale B and Extended Yale B prove that the proposed method can effectively improve the face recognition accuracy under complex illumination conditions.
出处
《厦门大学学报（自然科学版）》
CAS
CSCD
北大核心
2014年第3期359-363,共5页
Journal of Xiamen University：Natural Science
基金
国家重点实验室开放基金(BUAA-VR-14KF-01)
教育部留学回国人员科研启动基金
