Mural image shedding diseases inpainting algorithm based on structure priority

A novel mural image inpainting algorithm based on structure priority has been developed to effectively repair shedding diseases that pass through contour lines in ancient murals, such as those found in the Longmen Grottoes. This approach converts the structure repair problem into an optimization task and leverages the global optimization capability of genetic algorithms to connect structural information in damaged areas. Texture is then filled using subarea optimization, resulting in a more accurate and ideal repair effect compared to traditional methods. The algorithm demonstrates superior performance in both subjective and objective evaluations, offering a promising solution for the preservation of cultural heritage without causing secondary damage.

Pen, H., Wang, S., & Zhang, Z. (2023). Mural image shedding diseases inpainting algorithm based on structure priority. Third International Conference on Artificial Intelligence and Computer Engineering (ICAICE 2022), 128. https://doi.org/10.1117/12.2671230