摘要： 针对传统人脸识别方法中二维Gabor滤波器及下采样方法的局限性，提出一种融合Log-Gabor的统计采样与支持向量机(SVM)的人脸识别方法。该方法使用Log-Gabor滤波器替代传统的二维Gabor滤波器提取特征，运用给出的统计下采样方法代替传统的下采样方法来初步降维，并使用主成分分析法进一步降维和应用SVM进行识别。在基于ORL与FERET人脸库的实验结果表明，该方法具有较高识别率和较强鲁棒性。

关键词: 人脸识别, Log-Gabor滤波器, 特征提取, 主成分分析, 支持向量机

Abstract: Aiming at shortcomings of 2D-Gabor filters and down-sampling of traditional methods, a new method is proposed using statistical sampling of Log-Gabor Features and Support Vector Machine(SVM) for face recognition. A bank of Log-Gabor filters is applied on images to extract features. The statistical sampling and Principal Component Analysis(PCA) are used successively in order to reduce dimensions. SVM is employed for classification. Experimental results based on the ORL and FERET face database show that the novel method achieves high efficiency and strong robustness.

Key words: face recognition, Log-Gabor filter, feature extraction, Principal Component Analysis(PCA), Support Vector Machine(SVM)

中图分类号: 

• TP18