Defect Detection and Classification of Cultural Heritage Buildings Using Deep Learning

Srujan Gokak, Prem Khichade, Nikhil Heggalagi, Aditya Billowria, Shashank Hegde

2025

Abstract

This paper discusses the application of deep learning models, specifically MobileNetV3 and Grad-CAM, in the detection and classification of defects in cultural heritage buildings. A dataset of images of heritage sites was used, and the MobileNetV3-based model resulted in a classification accuracy of 91.5% on the test dataset, effectively identifying sites with structural defects that may require conservation efforts. Grad-CAM visualizations were used to produce heatmaps that highlighted critical regions influencing the model’s predictions, enhancing interpretability and trust in AI-driven assessments. The training process included data augmentation, learning rate scheduling, and model pruning, reducing the model size by 20% without affecting performance. This lightweight and efficient framework demonstrates the potential of integrating advanced deep learning models with explainable AI techniques to improve accuracy in defect localization and classification and support preservation initiatives for cultural heritage sites.

Paper Citation

in Harvard Style

Gokak S., Khichade P., Heggalagi N., Billowria A. and Hegde S. (2025). Defect Detection and Classification of Cultural Heritage Buildings Using Deep Learning. In Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT; ISBN 978-989-758-763-4, SciTePress, pages 619-626. DOI: 10.5220/0013633700004664

in Bibtex Style

@conference{incoft25,
author={Srujan Gokak and Prem Khichade and Nikhil Heggalagi and Aditya Billowria and Shashank Hegde},
title={Defect Detection and Classification of Cultural Heritage Buildings Using Deep Learning},
booktitle={Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT},
year={2025},
pages={619-626},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013633700004664},
isbn={978-989-758-763-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT
TI - Defect Detection and Classification of Cultural Heritage Buildings Using Deep Learning
SN - 978-989-758-763-4
AU - Gokak S.
AU - Khichade P.
AU - Heggalagi N.
AU - Billowria A.
AU - Hegde S.
PY - 2025
SP - 619
EP - 626
DO - 10.5220/0013633700004664
PB - SciTePress