Abstract: This paper proposes a robust method for face recognition under uncontrolled lighting condition. The effects of illumination variation are reduced by Gamma correction, DoG（Difference of Gauss filtering） and contrast equalization. The original image is divided into several non-overlapping cells and their corresponding histograms of Local Binary Pattern（LBP） are generated according to the characteristics of the local binary patterns in face image. PCA（Principal Component Analysis） is used to reduce the dimension of feature vector.The classification is made by SVM（Support Vector Machine）. The experimental results show that the average precision of the proposed algorithm is up to 99.68%, which indicates the performance of the proposed algorithm is better than that of the existing methods.

Key words: face recognition, Local Binary Pattern（LBP）, Principal Component Analysis（PCA）, Support Vector Machine（SVM）

摘要： 提出一种具有较强光照鲁棒性的人脸识别方法。通过Gamma校正、高斯差分（DoG）滤波和对比度均衡化算法对图像进行光照预处理，降低光照敏感度；利用局部二值模式（LBP）算子提取局部纹理特征，将图像划分为若干个不重叠的子区域，提取每个子区域LBP直方图，形成人脸图像特征，用主成分分析（PCA）进行降维处理；使用支持向量机（SVM）进行分类识别。在Yale-B数据库进行实验的结果表明，该算法的平均识别率可达99.68%，其性能优于目前该领域的典型算法。

关键词: 人脸识别, 局部二元模式, 主成分分析, 支持向量机