Abstract: To overcome the limitations of traditional face recognition methods under variations in position, illumination and expression, a novel method of face recognition based on Nonsubsampled Shearlet Transform（NSST） and Center-Symmetric Local Directional Pattern（CSLDP） is proposed. A face image is decomposed with NSST and CSLDP operator is utilized to get CSLDP feature maps from sub-images. CSLDP feature maps are divided into several blocks and the concatenated histogram calculated over each block is used as the face feature. Nearest neighbor classifier is used to classify the faces. Experimental results on ORL, YALE CAS-PEAL-R1 face databases demonstrate that the proposed descriptor is simple and effective, and also robust to variations of posture, illumination and face expression.

Key words: face recognition, Nonsubsampled Shearlet Transform（NSST）, Center-Symmetric Local Directional Pattern（CSLDP）, nearest neighbor classifier

摘要： 针对人脸识别中姿态、光照和表情等变化造成的识别率不高的问题，提出一种非采样Shearlet变换（NSST）与中心对称局部方向模式相结合的人脸识别算法。采用NSST分解人脸图像，得到低频子带图像和高频子带图像，计算子带图像中心对称局部方向模式，分块统计直方图特征信息，将直方图串接起来作为人脸图像的特征向量，利用最近邻分类器分类识别。在ORL、YALE和CAS-PEAL-R1人脸库上进行测试，实验结果表明所提方法简单有效，且对姿态、光照和表情变化具有较好鲁棒性。

关键词: 人脸识别, 非采样Shearlet变换, 中心对称局部方向模式, 最近邻分类器