Abstract: Motivated by the deficiency of the Dual-Tree Complex Wavelet Transform(DT-CWT) for feature extraction along horizontal and vertical directions, a novel face recognition method is proposed by combining DT-CWT and Gabor wavelet. Face feature vectors are conducted by connecting the features extracted by DT-CWT in 6 directions and Gabor wavelet in the direction of 0° and 90°. Fisherfaces method is exploited to construct the linear subspace. Euclidean distance based classifier is adopted for classification. Experimental results demonstrate the validity of the proposed method with the ORL database.

Key words: face recognition, Dual-Tree Complex Wavelet Transform(DT-CWT), Gabor wavelet, Fisherfaces method

摘要： 针对双树复小波变换(DT-CWT)不能直接提取水平和垂直2个方向特征的不足，提出一种结合DT-CWT和Gabor小波的人脸识别方法。将Gabor小波提取的0°和90°特征与DT-CWT提取的6个方向特征连接起来共同构成人脸特征向量，采用Fisherfaces方法构建特征向量的线性子空间，应用基于欧氏距离的分类器实现分类。在ORL数据库上的实验结果证明了该方法的有效性。

关键词: 人脸识别, 双树复小波变换, Gabor小波, Fisherfaces方法

CLC Number: 

• TP311