Artificial Intelligence-Based color Reconstruction of Mogao Grottoes Murals Using Computer Vision Techniques
Year: 2025
Type: article
Abstract: The Mogao Grottoes murals have deteriorated over centuries due to environmental exposure, pigment degradation, and natural ageing, making cultural heritage preservation difficult. AI and computer vision can identify, classify, and reconstruct faded pigments, revolutionizing color restoration. This reconstructs faded mural sections using deep learning, image processing, and pigment data implemented through TensorFlow, PyTorch and OpenCV. The study uses high-resolution Digital Dunhuang database images of Mogao Grottoes murals and 50 pigments categorized by color, stability, and chemical composition. CNNs and deep learning-based color mapping algorithms detect fading and suggest color restorations of pigments. AI reconstructions along with history accuracy through expert evaluations and pigment records. Artificial intelligence-driven mural conservation detects faded pigments, precisely reconstructs missing sections, and matches restored colors to historical authenticity, improving accuracy, efficiency, and scalability. Scientifically, AI-based digital heritage conservation outperforms manual restoration. AI preserves and faithfully reconstructs cultural heritage sites using historical artworks using global digital pigment database and deep learning-driven restoration models. The first reproducible and scientific model (CNN, GAN and deep learning-based color mapping algorithms) using AI-based color restoration and historical pigment analysis in Mogao Grottoes murals was created.
Authors: Yi Zhang, Thirawut Bunyasakseri
Institution: Naresuan University
Language: English
Subfield: Archeology
Field: Arts and Humanities
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4409293302