多阶段堆叠融合下的敦煌壁画修复  

Inpainting of Dunhuang Murals Under Multi-Stage Stacking Fusion

作　　者：邬开俊[1] 单宏全 田彬 俞天秀[2] 周颜林 WU Kaijun;SHAN Hongquan;TIAN Bin;YU Tianxiu;ZHOU Yanlin(School of Electronics and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China;Institute of Cultural Digitization,Dunhuang Research Institute,Dunhuang 730030,China)

机构地区：[1]兰州交通大学电子与信息工程学院,兰州730070 [2]敦煌研究院文物数字化研究所,敦煌730030

出　　处：《电子科技大学学报》2024年第4期551-559,共9页Journal of University of Electronic Science and Technology of China

基　　金：甘肃省自然科学基金(21JR7RA300)。

摘　　要：为解决敦煌破损壁画修复难的问题,提出一种多阶段堆叠融合下的壁画修复模型,在第一阶段进行粗修复以及最大化地进行所需特征信息的提取时采用Unet结构,为增强该阶段的修复效果,又加入对抗网络(LsGan);在第二阶段进行精细化修复,主要修复纹理以及细化遮挡区域,引入多头连接和多尺度分支堆叠模块并对其进行改进,对壁画进行多阶段信息提取并对其进行融合;最后为了修复全局细节,采用大感受野的编解码器结构并且引入轻量级通用可视化(Mobile ViT)模块,同时分离了壁画图像的通道并进行分析,引入对通道不敏感的极化注意力。实验结果表明,该模型很好地解决了壁画纹理以及细节修复的难题,相较于所选修复算法最优数据,在掩码面积为5%~20%时,峰值信噪比提高了3.312,在掩码面积为5%~80%时,平均峰值信噪比提高了1.02。A fresco restoration model under multi-stage stacking fusion is proposed to solve the problem of difficult restoration of broken frescoes at Dunhuang.The Unet structure is used for coarse repair in the first stage and for maximizing the extraction of the required feature information,and the adversarial network(LsGan)is added to enhance the repair effect in this stage.In the second stage,the restoration is refined,mainly by restoring textures and refining occluded areas,introducing and improving the multi-head connection and multi-scale branching stacking modules,extracting and fusing multi-stage information from the murals.Finally,in order to consider the global details of the fix,the codec structure of the big feeling field is used and the mobile ViT module is introduced.The channels of the mural images are also separated and analyzed,introducing polarization attention that is insensitive to the channels.The experimental results show that the model can solve the difficult problem of fresco texture and detail restoration well.Compared with the optimal data of the selected restoration algorithm,the peak signal-to-noise ratio improved by 3.312 at a mask area of 5%to 20%,and the average peak signal-to-noise ratio is improved by 1.02 at a mask area ranging from 5%to 80%.

关 键 词：敦煌壁画修复 多阶段堆叠 大感受野 多头嵌套 多分支融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]