基于CDD模型的西藏壁画数字图像修复技术研究

Research of Tibet mural digital images inpainting using CDD model

本文介绍了数字图像修复技术中的一个主要算法-CDD算法,具体分析了数字图像的破损类型以及对应修复方法,在破损区域较大的情况下,本文将CDD模型与另一个经典算法TV模型的修复性能做了对比,试验结果表明,前者无论在连接整体性还是边缘连贯性都优于后者。在保证修复质量的前提下,该模型可以大幅度提升修复效率。

The paper introduce the repair technology of one of the main algorithm-CDD algorithm, analysis of the breakage of the digital image types and corresponding repair methods, under the condition of the damaged area is larger, in this paper we extend the classical CDD model with another TV model contrasted the performance of repair, The results show that the former both on the edge of the connection integrity and consistency are better than that of the latter, under the premise of guarantee the quality of repair, the model can greatly improve repair efficiency.