Face recognition based on dual-tree complex wavelet feature
基于二元树复小波特征表示的人脸识别方法

Approximate shift invariance,good directional selectivity,computational efficiency properties of Dual-tree Complex Wavelet Transform(DTCWT) make it a good candidate for representing the face features.This paper proposed a new dual-tree complex wavelet feature representation for face recognition.We got the directional and spatial/frequency characteristics of the patterns and classify different face patterns in terms of these characteristics.Experimental results show that the proposed DTCWT based feature extraction,combined with the Fisherface recognition method,(DCF) can extract optimally discriminate face feature and achieve high recognition rate and generalization.In addition,our experiments verify that the proposed feature extraction based on DTCWT outperforms which based on Gabor wavelet.