一种彩色图像复原新方法:基于滑动窗口的支持向量回归算法  

Novel color image restoration method:sliding window based support vector regression algorithm

作　　者：黎明[1] 杨杰[1] 王辉[1] 蔡念[1] 

机构地区：[1]上海交通大学图像处理与模式识别研究所,上海200240

出　　处：《红外与激光工程》2006年第z4期79-82,共4页Infrared and Laser Engineering

基　　金：上海市教委基金(05NZ20);中国博士后科学基金(2005037503)

摘　　要：提出一种彩色图像复原新方法,将彩色图像按 RGB 三个通道进行分解,针对每个通道分别采用滑动窗口操作,直至遍历整幅图像,从而获得三个训练集。然后应用支持向量机进行回归分析,建立清晰图像与模糊图像之间的对应关系,从而得出彩色图像复原网络模型。最后根据该模型对待测模糊图像进行复原校正.实验结果显示新算法能很好地对模糊彩色图像进行复原,复原效果优于维纳滤波等经典滤波算法,且计算量远远小于迭代盲反卷积方法。A novel color image restoration method is presented.Firstly,a color image is separated into RGB channels,and an operation with a sliding window is employed to each channel,thus three training sets are obtained. Then the correlation between clear image and blurry image is constructed via regression analysis of the training sets by a support vector machine,and then the model for image restoration is constructed here.Finally,the blurry images to be tested are restored by this model.The experimental results show that the proposed method is most effective in color image restoration and superior to traditional restoration methods like Wiener filtering and iterative blind deconvolution method.

关 键 词：图像复原 支持向量回归 支持向量机 RGB色彩空间 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]