关键词: 人脸识别, 小波变换, 仿生模式识别

Abstract: Aiming at robustness of face recognition under the condition of illumination perturbations and facial variety, a face recognition method based on dyadic wavelet transform and biomimetic pattern recognition is proposed in this paper. The spline dyadic wavelet is applied to face images. The detail subbands are merged by concatenation. The facial images are treated and reduced dimension respectively by FFT（Fast Fourier Transform） and PCA（Principal Component Analysis）. The feature vectors are learned and recognized by biomimetic pattern recognition. The experimental results prove that the correct recognition rate of the method in this paper is higher than traditional recognition method.

Key words: face recognition, wavelet transform, biomimetic pattern recognition