摘要
提出一种利用小波分析提取人脸特征的方法。对人脸图像做小波分解,用网格划分其子图像,在各子块上提取统计特征,用其训练多分类支持向量机模型,最后用训练好的支持向量机进行人脸识别。选择ORL人脸库对该算法进行实验,与PCA算法的比较结果证明了该算法在识别性能方面的优越性。
A method that extracts facial feature based on wavelet analysis is proposed. Face images is decomposed by wavelet, using mask sizes to separate some wavelet sub-bands into small blocks, and then calculate the statistical characterization, which is then used to train the multi-class SVM model. Finally the trained SVM model is used to classify the human faces. The ORL database of faces is selected to test and evaluate the proposed algorithm. The results of the test is shown to perform very well in recognizing capability compared with PCA arithmetic for face recognition.
出处
《长春工业大学学报》
CAS
2006年第1期4-7,共4页
Journal of Changchun University of Technology
关键词
人脸识别
小波分解
特征提取
支持向量机
face recognition
wavelet decomposition
feature extraction
SVM.