摘要
Image inpainting is an important part of image science,but in the past,researches were focused on gray value image inpainting.In this paper,we investigate the inpainting effects of some variational models of color image diffusion.Five variational models for color image inpainting are proposed and their Split Bregman algorithms are designed.Their regularizers are LTV(Layered Total Variation) regularizer,CTV(Color Total Variation) regularizer,MTV(Multichannel Total Variation) regularizer,PA(Polyakov Action) regularizer and RPA(Reduced Polyakov Action) regularizer respectively.In order to compare their performances,we use the same data term...Some numerical experiments show the differences of the above mentioned models for color image inpainting.
Image inpainting is an important part of image science, but in the past, researches were focused on gray value image inpainting. In this paper, we investigate the inpainting effects of some variational models of color image diffusion. Five variational models for color image inpainting are proposed and their Split Bregman algorithms are designed. Their regularizers are LTV （Layered Total Variation） regularizer, CTV（Color Total Variation） regularizer, MTV（Multichannel Total Variation） regularizer, PA （Polyakov Action） regularizer and RPA （Reduced Polyakov Action） regularizer respectively. In order to compare their performances, we use the same data term... Some numerical experiments show the differences of the above mentioned models for color image inpainting.
出处
《科技信息》
2011年第6期153-156,共4页
Science & Technology Information
