中文摘要：

人脸识别是生物识别领域一项重要的内容,在身份验证和公共安全方面越来越被重视。人脸识别易受光照、表情、饰物等外界因素的影响,提高识别的鲁棒性和准确率成为该领域研究的热点。PCA是目前人脸识别技术中较为常用的方法,但是该方法鲁棒性差,识别率不能满足实际应用。使用Gabor小波提取特征的方法能够提高识别鲁棒性,但数据量成倍增加,不能满足识别实时性。该文研究了一种基于Gabor小波特征提取的PCA人脸识别方法。首先通过Gabor小波对人脸图像进行滤波,使用Gabor小波幅值特性与人脸图像卷积提取人脸特征,提高识别鲁棒性,然后通过主成分分析法(PCA)进行数据降维,降低计算复杂度,最后通过ORL人脸库进行仿真实验,实验结果表明该方法对于光照、姿态鲁棒性较高,识别率明显优于直接进行PCA的传统方法,平均识别率提高4.458%。

英文摘要：

Face recognition is one of the important contents in the biometric identification field,and more and more attention is paid to it by the identity verification and public security field. Face recognition is easy to be influenced by external factors such as illumination, expression and decorations, so the robustness and accuracy of the recognition is the hot study topic in this field. PCA is a common method in face recognition technology, but the method has bad robustness and can not meet the practical ap-plication. Using Gabor wavelet transform method can improve the robustness, but the amount of data increases exponentially, which can not meet the real-time performance. Face recognition based on gabor wavelet feature extraction and PCA is studied in this paper.Firstly, the Gabor wavelet is used to filter the face image. Then, PCA is used to reduce data dimension, finally, the ORL is used to simulation experiment. The experimental results show that the proposed method is better than the traditional method of PCA. The average recognition rate is increased by 4.458%.