基于小波变换和LBP算子的人脸识别研究

Research on Face Recognition Based on Wavelet Transform and LBP Operator

为了提高人脸识别效果,针对当前人脸识别方法存在的局限性,如识别准确率低,耗时长等,提出了基于小波变换与LBP算子的人脸识别方法。首先采集待识别的人脸图像,并采用小波变换对人脸图像作多尺分解,通过对高频子图像进行处理,消除光照变化对人脸识别的干扰,然后采用LBP算子提取人脸图像纹理特征,将特征连接组合在一起产生特征向量,最后采用k近邻算法根据特征向量建立人脸识别的分类器。采用标准人脸识别数据集Yale-B与AR作为测试对象,测试结果表明,小波变换与LBP算子能够克服当前人脸识别方法的弊端,提高人脸识别的准确率,并且人脸识别效率得到了明显改善,整体人脸识别效果要优于当前其它方法,为后续人脸处理打下了良好的基础。

In order to improve the effect of face recognition,aiming at the limitations of current face recognition methods,such as low recognition accuracy and long time consuming,a face recognition method based on wavelet transform and LBP operator is proposed.Firstly,the face image to be recognized is collected,and the wavelet transform is used to decompose the face image into multi scale.The high-frequency sub images are processed to eliminate the interference of illumination changes on face recognition.Then,the LBP operator is used to extract the texture features of the face image,and the features are connected and combined to generate the feature vector.Finally,the K-nearest neighbor algorithm is used to establish the classification model of face recognition according to the feature vector.Yale-B are AR are used as testing databases.The test results show that the wavelet transform and LBP operator overcome the shortcomings of the current face recognition methods,improve the accuracy of face recognition,and the efficiency of face recognition has been significantly improved.The whole face recognition effect is better than other current methods,which lays a good foundation for the subsequent face processing foundation.