基于特征互补图像的人脸识别仿真研究

Rapid Feature Fusion of 2DCCA Based on Complementary Feature Image and Face Recognition

研究人脸识别精度问题。由于人脸图像中存在大量干扰信息的缺点,而造成了人脸识别正确率下降,为了解决上述问题,提出了一种基于特征互补图像快速特征融合算法。算法通过对人脸图像的位平面切片图像分析,采用位平面图像分解法,通过各种合成策略构造多幅样本图像。并突出高位平面图像,采用两种加权策略将每一幅人脸图像样本都生成"特征互补图像"。然后,直接用图像的二维典型相关分析(2DCCA)法对两种特征互补图像进行特征抽取。最后通过在ORL国际标准人脸库上进行的实验,结果表明,高位平面图像的典型相关鉴别特征提高了正确识别率,并且因为摒弃了原始人脸图像的大部分干扰信息所以具有更强的鲁棒性。

For there is a lot of interference information in face images,firstly proposed a new sample augment method based on Bit-cutting face images.It slices the image at eight different planes（bit-planes）.Then,it augments new training samples by combining the bit-planes images.By using two kinds of weighting strategy to highlight the high plane images,two kinds of new training samples for each one face image samples are generated called ＂two complementary feature image＂.Finally,two dimensions canonical correlation analysis（2DCCA） works on the ＂complementary feature image＂ for feature fusion.The experimental results on ORL face database verify the effectiveness of the proposed method.Because the proposed method abandoned most interfere information of the primitive face images and therefore it have a stronger robustness.