A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Automation in Documentation of Ageing Masonry Infrastructure Through Image-Based Techniques and Machine Learning
https://orcid.org/http://orcid.org/0000-0002-3997-2443
https://orcid.org/http://orcid.org/0000-0002-8604-8659
Visual inspection and manual documentation of masonry is a time consuming and subjective process. This paper aims to improve automation in documentation and assessment of ageing masonry infrastructure through image-based techniques and machine learning. A large dataset of annotated images has been developed to train a deep learning model. Different convolutional neural networks were investigated to identify the most suitable for the task. The results presented are combined with previous work to generate high quality geometrical and numerical models of masonry infrastructure. This implementation of deep learning, for segmentation and localisation of bricks in masonry, highlights the potential of recent technologies for the automation of structural inspection, documentation, and analysis of cultural heritage.
Automation in Documentation of Ageing Masonry Infrastructure Through Image-Based Techniques and Machine Learning
https://orcid.org/http://orcid.org/0000-0002-3997-2443
https://orcid.org/http://orcid.org/0000-0002-8604-8659
Visual inspection and manual documentation of masonry is a time consuming and subjective process. This paper aims to improve automation in documentation and assessment of ageing masonry infrastructure through image-based techniques and machine learning. A large dataset of annotated images has been developed to train a deep learning model. Different convolutional neural networks were investigated to identify the most suitable for the task. The results presented are combined with previous work to generate high quality geometrical and numerical models of masonry infrastructure. This implementation of deep learning, for segmentation and localisation of bricks in masonry, highlights the potential of recent technologies for the automation of structural inspection, documentation, and analysis of cultural heritage.
Automation in Documentation of Ageing Masonry Infrastructure Through Image-Based Techniques and Machine Learning
https://orcid.org/http://orcid.org/0000-0002-3997-2443
https://orcid.org/http://orcid.org/0000-0002-8604-8659
Visual inspection and manual documentation of masonry is a time consuming and subjective process. This paper aims to improve automation in documentation and assessment of ageing masonry infrastructure through image-based techniques and machine learning. A large dataset of annotated images has been developed to train a deep learning model. Different convolutional neural networks were investigated to identify the most suitable for the task. The results presented are combined with previous work to generate high quality geometrical and numerical models of masonry infrastructure. This implementation of deep learning, for segmentation and localisation of bricks in masonry, highlights the potential of recent technologies for the automation of structural inspection, documentation, and analysis of cultural heritage.
Automation in Documentation of Ageing Masonry Infrastructure Through Image-Based Techniques and Machine Learning
Lecture Notes in Civil Engineering
Rizzo, Piervincenzo (editor) / Milazzo, Alberto (editor) / Loverdos, Dimitrios (author) / Sarhosis, Vasilis (author)
European Workshop on Structural Health Monitoring ; 2022 ; Palermo, Italy
2022-06-22
9 pages
Article/Chapter (Book)
Electronic Resource
English
| Taylor & Francis Verlag | 2025
Documentation and evaluation of historic masonry arch bridges by means of geomatic techniques
| British Library Conference Proceedings | 2007
Developing Masonry Research Infrastructure
| British Library Conference Proceedings | 1994