Real-world evaluation of automated defect detection in masonry bridges using 360° imagery with machine learning

Sen, A ORCID: https://orcid.org/0000-0001-8967-9475, Sun, Q, Wu, S ORCID: https://orcid.org/0000-0002-5633-2229 and Talebi, S, 2026. Real-world evaluation of automated defect detection in masonry bridges using 360° imagery with machine learning. Construction Innovation: Information, Process, Management. ISSN 1471-4175

Preview

Text 2579941_Sen.pdf - Post-print Download (774kB) | Preview

Abstract

Purpose: The purpose of this study is to evaluate different deep learning approaches, Convolutional Neural Networks (CNN), transformer, hybrid and commercial models, for automated defect detection in UK masonry railway bridges, in both laboratory and real-world settings, using high-resolution 360° imagery.

Design/methodology/approach: Expert-annotated imagery was categorised into six defect types, with SMOTE oversampling applied to mitigate class imbalance. Four widely used architectures, EfficientNet, Swin Transformer, ConvNeXt and Azure CustomVision, were benchmarked using compact variants in a two-stage design: laboratory data and real-world evaluation, to assess feasibility and generalisability.

Findings: All models achieved high performance on laboratory data (0.83–0.91 accuracy), demonstrating feasibility in controlled environments. However, when applied to real-world evaluation, accuracies declined to 0.76–0.86, with the Swin Transformer showing the greatest robustness (2% drop). This decline was largely attributable to extreme class imbalance (non-defect to defect ratio around 220:1), which caused models to favour the non-defect class. While Vegetation and Loss of Section showed moderate recall, crack detection was less reliable, likely affected by limited samples and textural similarity to other classes. Consequently, overall accuracy masked substantial class-level disparities, and ensemble modelling delivered only marginal improvements under these conditions.

Practical implications: Automated detection can streamline inspections and enhance consistency, as compact models show feasibility. However, reliable deployment requires addressing imbalance, as some defect classes (e.g. cracks) remain unreliable.

Originality/value: To the best of the authors’ knowledge, this study is the first comprehensive evaluation on masonry railway bridges with 360° imagery, which advances beyond prior laboratory environment by systematically testing generalisability in real-world sceneries, generating new insights into imbalance-driven errors and class-specific detection limits.

Item Type:	Journal article
Publication Title:	Construction Innovation: Information, Process, Management
Creators:	Sen, A., Sun, Q., Wu, S. and Talebi, S.
Publisher:	Emerald
Date:	23 February 2026
ISSN:	1471-4175
Identifiers:	Number Type 10.1108/ci-08-2025-0377 DOI 2579941 Other
Rights:	© 2026 Emerald Publishing Limited. This AAM is provided for your own personal use only. It may not be used for resale, reprinting, systematic distribution, emailing, or for any other commercial purpose without the permission of the publisher
Divisions:	Schools > School of Architecture, Design and the Built Environment
Record created by:	Laura Borcherds
Date Added:	26 Feb 2026 08:34
Last Modified:	26 Feb 2026 08:34
URI:	https://irep.ntu.ac.uk/id/eprint/55334

Actions (login required)

Edit View

Statistics

Views

Downloads