基于Gabor小波特征的多姿态人脸图像识别

多姿态人脸识别在很多领域具有重要的应用价值。基于多姿态人脸图像及Gabor小波特点选取离散化参数,对人脸图像进行Gabor小波变换;然后采用两步降维法对变换系数进行降维,基于降维后的Gabor特征表示实现人脸识别。实验将互不相交的两个样本集依次作为训练集和测试集,验证了该方法在人脸识别中对于不同姿态和表情的有效性及鲁棒性。

Pose-Varied Face Recognition Based on Gabor Wavelet Features

Pose-varied face recognition is important in many research areas.Based on the analysis of Gabor wavelet and pose -varied face images,discrete Gabor wavelet is applied to face images with parameters deliberately selected.A method named two step dimension reduction is used to reduce the dimension of Gabor feature vector.Based on the reduced Gabor wavelet representation of face images and a certain distance measure,two groups of images,which share no intersection,are used as train group and test group in turn to test the robustness of Gabor wavelet representation to varied poses and deferent facial expressions.The result shows that high recognition rate for pose-varied face recognition can be obtained based on this Gabor wavelet features.