Dunhuang murals image restoration method based on generative adversarial network
Abstract Murals are an important part of China’s cultural heritage. After more than a 1000 years of exposure to the sun and wind, most of these ancient murals have become mottled, with damage such as cracking, mold, and even large-scale detachment. It is an urgent work to restore these damaged mural...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2024-02-01
|
Series: | Heritage Science |
Subjects: | |
Online Access: | https://doi.org/10.1186/s40494-024-01159-8 |
_version_ | 1797273621963472896 |
---|---|
author | Hui Ren Ke Sun Fanhua Zhao Xian Zhu |
author_facet | Hui Ren Ke Sun Fanhua Zhao Xian Zhu |
author_sort | Hui Ren |
collection | DOAJ |
description | Abstract Murals are an important part of China’s cultural heritage. After more than a 1000 years of exposure to the sun and wind, most of these ancient murals have become mottled, with damage such as cracking, mold, and even large-scale detachment. It is an urgent work to restore these damaged murals. The technique of digital restoration of mural images refers to the reconstruction of structures and textures to virtually fill in the damaged areas of the image. Existing digital restoration methods have the problems of incomplete restoration and distortion of local details. In this paper, we propose a generative adversarial network model combining a parallel dual convolutional feature extraction depth generator and a ternary heterogeneous joint discriminator. The generator network is designed with the mechanism of parallel extraction of image features by vanilla convolution and dilated convolution, capturing multi-scale features simultaneously, and reasonable parameter settings reduce the loss of image information. A pixel-level discriminator is proposed to identify the pixel-level defects of the captured image, and its joint global discriminator and local discriminator discriminate the generated image at different levels and granularities. In this paper, we create the Dunhuang murals dataset and validate our method on this dataset, and the experimental results show that the method of this paper has an overall improvement in the evaluation metrics of PSNR and SSIM compared with the comparative methods. The restored resultant image is more in line with the subjective vision of human beings, which achieves the effective restoration of mural images. |
first_indexed | 2024-03-07T14:47:01Z |
format | Article |
id | doaj.art-c347cbfcc6c14b8195a4dec033404af8 |
institution | Directory Open Access Journal |
issn | 2050-7445 |
language | English |
last_indexed | 2024-03-07T14:47:01Z |
publishDate | 2024-02-01 |
publisher | SpringerOpen |
record_format | Article |
series | Heritage Science |
spelling | doaj.art-c347cbfcc6c14b8195a4dec033404af82024-03-05T19:55:45ZengSpringerOpenHeritage Science2050-74452024-02-0112112010.1186/s40494-024-01159-8Dunhuang murals image restoration method based on generative adversarial networkHui Ren0Ke Sun1Fanhua Zhao2Xian Zhu3School of Information and Communication Engineering, Communication University of ChinaSchool of Information and Communication Engineering, Communication University of ChinaSchool of Information and Communication Engineering, Communication University of ChinaSchool of Information and Communication Engineering, Communication University of ChinaAbstract Murals are an important part of China’s cultural heritage. After more than a 1000 years of exposure to the sun and wind, most of these ancient murals have become mottled, with damage such as cracking, mold, and even large-scale detachment. It is an urgent work to restore these damaged murals. The technique of digital restoration of mural images refers to the reconstruction of structures and textures to virtually fill in the damaged areas of the image. Existing digital restoration methods have the problems of incomplete restoration and distortion of local details. In this paper, we propose a generative adversarial network model combining a parallel dual convolutional feature extraction depth generator and a ternary heterogeneous joint discriminator. The generator network is designed with the mechanism of parallel extraction of image features by vanilla convolution and dilated convolution, capturing multi-scale features simultaneously, and reasonable parameter settings reduce the loss of image information. A pixel-level discriminator is proposed to identify the pixel-level defects of the captured image, and its joint global discriminator and local discriminator discriminate the generated image at different levels and granularities. In this paper, we create the Dunhuang murals dataset and validate our method on this dataset, and the experimental results show that the method of this paper has an overall improvement in the evaluation metrics of PSNR and SSIM compared with the comparative methods. The restored resultant image is more in line with the subjective vision of human beings, which achieves the effective restoration of mural images.https://doi.org/10.1186/s40494-024-01159-8Murals restorationGenerative adversarial networkBinary convolution mechanismTernary joint discriminator |
spellingShingle | Hui Ren Ke Sun Fanhua Zhao Xian Zhu Dunhuang murals image restoration method based on generative adversarial network Heritage Science Murals restoration Generative adversarial network Binary convolution mechanism Ternary joint discriminator |
title | Dunhuang murals image restoration method based on generative adversarial network |
title_full | Dunhuang murals image restoration method based on generative adversarial network |
title_fullStr | Dunhuang murals image restoration method based on generative adversarial network |
title_full_unstemmed | Dunhuang murals image restoration method based on generative adversarial network |
title_short | Dunhuang murals image restoration method based on generative adversarial network |
title_sort | dunhuang murals image restoration method based on generative adversarial network |
topic | Murals restoration Generative adversarial network Binary convolution mechanism Ternary joint discriminator |
url | https://doi.org/10.1186/s40494-024-01159-8 |
work_keys_str_mv | AT huiren dunhuangmuralsimagerestorationmethodbasedongenerativeadversarialnetwork AT kesun dunhuangmuralsimagerestorationmethodbasedongenerativeadversarialnetwork AT fanhuazhao dunhuangmuralsimagerestorationmethodbasedongenerativeadversarialnetwork AT xianzhu dunhuangmuralsimagerestorationmethodbasedongenerativeadversarialnetwork |