TensorFlow-Native Implementation for Crack Detection in Concrete Structures

This paper presents a TensorFlow-native implementation for automated crack detection in concrete structures, addressing the critical need for efficient and objective infrastructure monitoring. Leveraging a Convolutional Neural Network architecture with 24.8 million parameters, the model was trained...

Full description

Saved in:
Bibliographic Details
Main Authors: Ayebare, Memory, Chavula, Petros, Mugisha, Simon, Byamukama, Willbroad
Format: Article
Language:English
Published: Mesopotamian Journal of Civil Engineering 2026
Subjects:
Online Access:http://hdl.handle.net/20.500.12493/3053
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1856816785624399872
author Ayebare, Memory
Chavula, Petros
Mugisha, Simon
Byamukama, Willbroad
author_facet Ayebare, Memory
Chavula, Petros
Mugisha, Simon
Byamukama, Willbroad
author_sort Ayebare, Memory
collection KAB-DR
description This paper presents a TensorFlow-native implementation for automated crack detection in concrete structures, addressing the critical need for efficient and objective infrastructure monitoring. Leveraging a Convolutional Neural Network architecture with 24.8 million parameters, the model was trained on a large-scale dataset of 40,000 images, each with a 227x227 RGB resolution. The methodology, incorporating specific framework optimizations and a rigorous training configuration, achieved a remarkable overall classification accuracy of 99.375% on the validation dataset. The model demonstrated balanced performance with precision values of 0.993 and 0.994, recall values of 0.994 and 0.993, and F1-scores of 0.994 and 0.994 for both "No Crack" and "Crack" classes. This high accuracy, coupled with balanced metrics, underscores the model's effectiveness and reliability for practical applications. The proposed solution significantly enhances real-time structural health monitoring systems, mitigating the limitations of traditional manual inspections and facilitating proactive maintenance strategies for concrete infrastructure
format Article
id oai:idr.kab.ac.ug:20.500.12493-3053
institution KAB-DR
language English
publishDate 2026
publisher Mesopotamian Journal of Civil Engineering
record_format dspace
spelling oai:idr.kab.ac.ug:20.500.12493-30532026-01-04T00:00:34Z TensorFlow-Native Implementation for Crack Detection in Concrete Structures Ayebare, Memory Chavula, Petros Mugisha, Simon Byamukama, Willbroad Concrete crack detection Convolutional Neural Network TensorFlow Structural health monitoring Deep learning. This paper presents a TensorFlow-native implementation for automated crack detection in concrete structures, addressing the critical need for efficient and objective infrastructure monitoring. Leveraging a Convolutional Neural Network architecture with 24.8 million parameters, the model was trained on a large-scale dataset of 40,000 images, each with a 227x227 RGB resolution. The methodology, incorporating specific framework optimizations and a rigorous training configuration, achieved a remarkable overall classification accuracy of 99.375% on the validation dataset. The model demonstrated balanced performance with precision values of 0.993 and 0.994, recall values of 0.994 and 0.993, and F1-scores of 0.994 and 0.994 for both "No Crack" and "Crack" classes. This high accuracy, coupled with balanced metrics, underscores the model's effectiveness and reliability for practical applications. The proposed solution significantly enhances real-time structural health monitoring systems, mitigating the limitations of traditional manual inspections and facilitating proactive maintenance strategies for concrete infrastructure 2026-01-03T14:58:24Z 2026-01-03T14:58:24Z 2025 Article Ayebare, M., Chavula, P., Mugisha, S., & Willbroad, B. (2025). TensorFlow-Native Implementation for Crack Detection in Concrete Structures. Mesopotamian Journal of Civil Engineering, 2025, 97-108. 3006-1148 http://hdl.handle.net/20.500.12493/3053 en Attribution-NonCommercial-NoDerivs 3.0 United States http://creativecommons.org/licenses/by-nc-nd/3.0/us/ application/pdf Mesopotamian Journal of Civil Engineering
spellingShingle Concrete crack detection
Convolutional Neural Network
TensorFlow
Structural health monitoring
Deep learning.
Ayebare, Memory
Chavula, Petros
Mugisha, Simon
Byamukama, Willbroad
TensorFlow-Native Implementation for Crack Detection in Concrete Structures
title TensorFlow-Native Implementation for Crack Detection in Concrete Structures
title_full TensorFlow-Native Implementation for Crack Detection in Concrete Structures
title_fullStr TensorFlow-Native Implementation for Crack Detection in Concrete Structures
title_full_unstemmed TensorFlow-Native Implementation for Crack Detection in Concrete Structures
title_short TensorFlow-Native Implementation for Crack Detection in Concrete Structures
title_sort tensorflow native implementation for crack detection in concrete structures
topic Concrete crack detection
Convolutional Neural Network
TensorFlow
Structural health monitoring
Deep learning.
url http://hdl.handle.net/20.500.12493/3053
work_keys_str_mv AT ayebarememory tensorflownativeimplementationforcrackdetectioninconcretestructures
AT chavulapetros tensorflownativeimplementationforcrackdetectioninconcretestructures
AT mugishasimon tensorflownativeimplementationforcrackdetectioninconcretestructures
AT byamukamawillbroad tensorflownativeimplementationforcrackdetectioninconcretestructures