Crack Detection in Single- and Multi-Light Images of Painted Surfaces using Convolutional Neural Networks

Simple item page
dc.contributor.authorDulecha, Tinsae Gebrechristosen_US
dc.contributor.authorGiachetti, Andreaen_US
dc.contributor.authorPintus, Ruggeroen_US
dc.contributor.authorCiortan, Irinaen_US
dc.contributor.authorVillanueva, Alberto Jaspeen_US
dc.contributor.authorGobbetti, Enricoen_US
dc.contributor.editorRizvic, Selma and Rodriguez Echavarria, Karinaen_US
dc.date.accessioned2019-11-06T06:02:21Z
dc.date.available2019-11-06T06:02:21Z
dc.date.issued2019
dc.description.abstractCracks represent an imminent danger for painted surfaces that needs to be alerted before degenerating into more severe aging effects, such as color loss. Automatic detection of cracks from painted surfaces' images would be therefore extremely useful for art conservators; however, classical image processing solutions are not effective to detect them, distinguish them from other lines or surface characteristics. A possible solution to improve the quality of crack detection exploits Multi-Light Image Collections (MLIC), that are often acquired in the Cultural Heritage domain thanks to the diffusion of the Reflectance Transformation Imaging (RTI) technique, allowing a low cost and rich digitization of artworks' surfaces. In this paper, we propose a pipeline for the detection of crack on egg-tempera paintings from multi-light image acquisitions and that can be used as well on single images. The method is based on single or multi-light edge detection and on a custom Convolutional Neural Network able to classify image patches around edge points as crack or non-crack, trained on RTI data. The pipeline is able to classify regions with cracks with good accuracy when applied on MLIC. Used on single images, it can give still reasonable results. The analysis of the performances for different lighting directions also reveals optimal lighting directions.en_US
dc.description.sectionheadersAnalysis and Visualisation
dc.description.seriesinformationEurographics Workshop on Graphics and Cultural Heritage
dc.identifier.doi10.2312/gch.20191347
dc.identifier.isbn978-3-03868-082-6
dc.identifier.issn2312-6124
dc.identifier.pages43-50
dc.identifier.urihttps://doi.org/10.2312/gch.20191347
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/gch20191347
dc.publisherThe Eurographics Associationen_US
dc.subjectComputing methodologies
dc.subjectSupervised learning by classification
dc.subjectCross
dc.subjectvalidation
dc.subjectApplied computing
dc.subjectFine arts
dc.titleCrack Detection in Single- and Multi-Light Images of Painted Surfaces using Convolutional Neural Networksen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
043-050.pdf
Size:
7.52 MB
Format:
Adobe Portable Document Format
Download
Collections
GCH 2019 - Eurographics Workshop on Graphics and Cultural Heritage