基于小波融合的人脸图像增强方法  

Face Image Enhancement Method Based on Wavelet Fusion

作　　者：仇国庆[1] 龙毅[1] 刘芙蓉[2] 

机构地区：[1]重庆邮电大学智能仪器仪表及工业自动化与测试技术实验室,重庆400065 [2]重庆邮电大学通信与信息工程学院,重庆400065

出　　处：《电视技术》2014年第11期187-189,共3页Video Engineering

摘　　要：针对人脸识别中图像预处理问题,提出了一种基于高频强调滤波和连续均值量化变换的图像增强算法。高频强调滤波能够突出高频分量,同时也保留了高通滤波后丢失的低频分量,而连续均值量化变换是一种新的有效的图像增强方法。然后采用小波融合两者处理后的图像。实验表明,该方法能够改善整体视觉效果,证明该方法的有效性。With the image preprocessing issue in face recognition,an enhancement method is proposed ,which is disposed by high frequency emphasize filter and successive mean quantization transform（SMQT）. The high frequency emphasize filter can emphasize on high-frequency components and keep lost low-frequency components, SMQT is a new and efficient method of image enhancement. Then synthesizes their information by using wavelet method. The experiments show that the method can improve the image whole visual and demonstrates this method effective.

关 键 词：人脸识别 高频强调滤波 连续均值量化变换 小波 

分 类 号：TN911.73[电子电信—通信与信息系统]