关键词: 人脸识别, 特征提取, 双树复小波变换, 独立成分分析, 相关系数

Abstract: A novel face recognition method is proposed by adopting the dual-tree complex wavelet transform (DT-CWT) and independent component analysis (ICA). The DT-CWT is applied to face images to extract the feature vectors. The dimension of the salient feature vectors is reduced by principal component analysis (PCA). ICA further reduces the feature redundancies and derives independent feature vectors for the correlation-based classifier. DT-CWT has the capability of selectivity on scale and orientation, and preserves more information in frequency domain. Features extracted by DT-CWT and ICA can obtain the excellent performance on classification. Extensive experimental results demonstrated the validity of the proposed method using the ORL and AR database.

Key words: face recognition, feature extraction, dual-tree complex wavelet transform (DT-CWT), independent component analysis (ICA), correlation coefficient