Siamese Convolutional Neural Networks to Quantify Crack Pattern Similarity in Masonry Facades. Issue 1 (2nd January 2023)

• Record Type: Journal Article
• Authors: Rozsas, Arpad; Slobbe, Arthur; Huizinga, Wyke; Kruithof, Maarten; Ajithkumar Pillai, Krishna; Kleijn, Kelvin; Giardina, Giorgia
• Is Part Of: International journal of architectural heritage. Volume 17:Issue 1(2023)
• Page Start: 147
• Page End: 169
• Publication Date: 2023-01-02
• Subjects: Crack patterns -- deep neural network -- machine learning -- masonry structure -- regression -- similarity measure; Architecture -- Conservation and restoration -- Periodicals; Historic buildings -- Periodicals; Historic preservation -- Planning -- Periodicals; 721.0288
• DOI: 10.1080/15583058.2022.2134062
• Languages: English
• ISSNs: 1558-3058

Abstract

ABSTRACT: This paper proposes an automated approach to predict crack pattern similarities that correlate well with assessment by structural engineers. We use Siamese convolutional neural networks (SCNN) that take two crack pattern images as inputs and output scalar similarity measures. We focus on 2D masonry facades with and without openings. The image pairs are generated using a statistics-based approach and labelled by 28 structural engineering experts. When the data is randomly split into fit and test data, the SCNNs can achieve good performance on the test data ( R 2 ≈ 0.9 ). When the SCNNs are tested on "unseen" archetypes, their test R 2 values are on average 1% lower than the case where all archetypes are "seen" during the training. These very good results indicate that SCNNs can generalise to unseen cases without compromising their performance. Although the analyses are restricted to the considered synthetic images, the results are promising and the approach is general.