We propose a semi-supervised crack detection method that can be used for high-dimensional and multimodal acquisitions of paintings. Our dataset consists of a recent collection of images of the Ghent Altarpiece (1432), one of Northern Europe's most important art masterpieces. We build a classifier that is able to discern crack pixels from the background consisting of non-crack pixels, making optimal use of the information that is provided by each modality. To accomplish this we employ a recently developed non-parametric Bayesian classifier, that uses tensor factorizations to characterize any conditional probability. A prior is placed on the parameters of the factorization such that every possible interaction between predictors is allowed while still identifying a sparse subset among these predictors.
Cornelis, B, Dooms, A, Daubechies, I & Dunson, D 2014, Bayesian crack detection in high resolution data. in iTWIST 14 international Traveling Workshop on Interactions between Sparse models and Technology, Namur, Belgium. iTWIST'14 international Traveling Workshop on Interactions between Sparse models and Technology, Namur, Belgium, 27/08/14.
Cornelis, B., Dooms, A., Daubechies, I., & Dunson, D. (2014). Bayesian crack detection in high resolution data. In iTWIST 14 international Traveling Workshop on Interactions between Sparse models and Technology, Namur, Belgium
@inbook{2c42717156bb42c3a3ba5ee108832df2,
title = "Bayesian crack detection in high resolution data",
abstract = "We propose a semi-supervised crack detection method that can be used for high-dimensional and multimodal acquisitions of paintings. Our dataset consists of a recent collection of images of the Ghent Altarpiece (1432), one of Northern Europe's most important art masterpieces. We build a classifier that is able to discern crack pixels from the background consisting of non-crack pixels, making optimal use of the information that is provided by each modality. To accomplish this we employ a recently developed non-parametric Bayesian classifier, that uses tensor factorizations to characterize any conditional probability. A prior is placed on the parameters of the factorization such that every possible interaction between predictors is allowed while still identifying a sparse subset among these predictors.",
keywords = "Crack detection",
author = "Bruno Cornelis and Ann Dooms and Ingrid Daubechies and David Dunson",
year = "2014",
month = aug,
day = "7",
language = "English",
booktitle = "iTWIST 14 international Traveling Workshop on Interactions between Sparse models and Technology, Namur, Belgium",
note = "iTWIST'14 international Traveling Workshop on Interactions between Sparse models and Technology ; Conference date: 27-08-2014 Through 29-08-2014",
}