A research team led by Prof. ZHANG Pengchang from the Xi'an Institute of Optics and Precision Mechanics (XIOPM), Chinese Academy of Sciences, has successfully engineered an automated virtual restoration system for murals based on hyperspectral imaging technology. This system was applied to reconstruct murals in Tang Dynasty tombs, achieving outstanding restoration results and providing valuable insights for cultural heritage preservation.
The study has published in Heritage Science.
Hyperspectral imaging technology, integrating feature detection and visual perception, is widely used for high-dimensional information recording and analysis of material properties of color layers on mural surfaces. Additionally, hyperspectral pseudo-color display technology enables the virtual restoration of faded colors, laying the foundation for this innovative system
In this study, researchers virtually restored murals from the Tang Dynasty tomb discovered in Baiyangzhai Village, Xi'an. They utilized superpixel division of hyperspectral images combined with the spectral binary coding algorithm to accurately identify and classify mural damaged regions. Subsequently, a standard colorimetric system was then employed to convert hyperspectral images into RGB images, effectively addressing color fading and improving multi-scale image quality by 8.42% compared to traditional pseudo-color fusion methods. Finally, a layered restoration strategy was proposed, integrating partial convolutional neural network with the Criminisi algorithm, ensuring consistency and coherence in the restoration of large-scale missing points and defaced points.
“This work offers non-contact, non-destructive testing with superior color fidelity and restoration results compared to existing methods. It provides an excellent new scientific tool for cultural heritage preservation.” said Prof. ZHANG Pengchang from XIOPM.
(Available online 22 November 2024)
Fig. Restoration effect exhibition.(Image by XIOPM)
Download：