ARTDET: Machine learning software for automated detection of art deterioration in easel paintings
Francisco M. Garcia-Moreno,
Jesús Cortés Alcaraz,
José Manuel del Castillo de la Fuente,
Luis Rodrigo Rodríguez-Simón,
María Visitación Hurtado-Torres
Affiliations
Francisco M. Garcia-Moreno
Department of Software Engineering, Computer Science School, University of Granada, C/Periodista Daniel Saucedo Aranda, 18014, Spain; Research Centre for Information and Communication Technologies (CITIC-UGR), University of Granada, 18014, Spain; Corresponding author.
Jesús Cortés Alcaraz
Department of Software Engineering, Computer Science School, University of Granada, C/Periodista Daniel Saucedo Aranda, 18014, Spain
José Manuel del Castillo de la Fuente
Department of Paint and Restoration, Faculty of Fine Arts, University of Granada, Av. Andalucía n° 38, 18071, Spain
Luis Rodrigo Rodríguez-Simón
Department of Paint and Restoration, Faculty of Fine Arts, University of Granada, Av. Andalucía n° 38, 18071, Spain
María Visitación Hurtado-Torres
Department of Software Engineering, Computer Science School, University of Granada, C/Periodista Daniel Saucedo Aranda, 18014, Spain; Research Centre for Information and Communication Technologies (CITIC-UGR), University of Granada, 18014, Spain
The increasing interest in digital preservation of cultural heritage has led to ARTDET, a machine learning software for automated detection of deterioration in easel paintings. This web application uses a pre-trained Mask R-CNN model to detect Lacune (areas of missing paint, resulting in visible support panel) from the loss of the Painting Layer (LPL) and stucco repairs. ARTDET leverages high-resolution images annotated by expert restorers. The software achieved 80.4 % recall for LPL and stucco, with a 99 % confidence score in detected damages. Available as open access resource, ARTDET aids conservators and researchers in preserving invaluable artworks.
