基于混合二元缺向量值有理插值的图像修复  

Image inpainting based on blending bivariate lacunary vector-valued rational interpolants

作　　者：胡大海[1] 朱晓临[1] 

机构地区：[1]合肥工业大学数学学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2012年第11期1577-1581,共5页Journal of Hefei University of Technology：Natural Science

基　　金：教育部科学技术研究重大资助项目(309017);第38批留学回国人员科研启动基金资助项目(2010JYLH0322);安徽省自然科学基金资助项目(11040606M06);国家大学生创新性实验计划资助项目(101035937)

摘　　要：文章提出了一种基于二元缺向量值有理插值的彩色图像修复方法,该方法将数字图像的每个像素点看成定义在一个平面域上以R、G、B为分量的一个向量,根据已知像素的变化趋势,相应地采用二元缺向量值Newton-Newton插值、Newton-Thiele插值、Thiele-Newton插值或Thiele-Thiele插值构造向量值插值曲面,然后对插值曲面进行采样以实现修复。实验结果表明,该算法比TV方法和CDD方法修复的效果更好,所用时间更短。A method of color image inpainting based on the bivariate lacunary vector-valued rational in- terpolants is presented. In this method, each pixel of the digital image is viewed as a vector of R, G,B values defined on a planar domain, and Newton-Newton, Newton-Thiele, Thiele-Newton, or Thiele- Thiele vector-valued interpolants are applied to constructing a vector-valued interpolation surface ac- cording to the variation trend of the known pixels, and then the damaged areas are restored based on the sampling of the interpolation surface. Experimental results show that this algorithm is more effi- cient and less time consumption than TV method and CDD method.

关 键 词：二元缺向量值有理插值 混合向量值连分式插值 彩色图像修复 

分 类 号：O173.2[理学—数学]