中文摘要：

提出一种基于小波低频子带的统一特征空间奇异值分解的人脸识别方法.首先利用小波变换提取人脸图像的低频子带,然后作傅立叶变换,获得低频子带的频谱;将低频子带的频谱投影到标准脸奇异值分解的统一特征空间,投影系数作为人脸识别特征;采用L_1距离进行最小距离分类,实验结果表明,该方法比统一特征空间奇异值分解方法可获得更高的识别率.

英文摘要：

This paper proposes a face recognition method using uniform eigen-space singular value decomposi- tion(SVD)based on Wavelet low-frequency band.Wavelet Transform is applied to obtain low-frequency band of facial image.By using Fourier transform,the spectrum representation of low-frequency band is obtained. Then the spectrum is projected into a uniform eigen-spaoe that is obtained from SVD of standard face image and the resultant coefficients are used as the feature of the facial image for recognition.The L_1...