Interpretation and understanding of asphalt crack detection deep learning models using integrated gradient (I.G.) maps

Asphalt cracking poses a significant deterioration challenge in road networks. Historically, government agencies employed visual inspection methods to identify asphalt cracking. However, this approach is labour-intensive and time-consuming, prompting most authorities to transition to automated crack...

Full description

Saved in:
Bibliographic Details
Main Authors: Gihan P. Ruwanpathirana, Sadeep Thilakarathna, Shanaka Kristombu Baduge
Format: Article
Language:English
Published: Elsevier 2025-07-01
Series:Case Studies in Construction Materials
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2214509525004115
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849702198091448320
author Gihan P. Ruwanpathirana
Sadeep Thilakarathna
Shanaka Kristombu Baduge
author_facet Gihan P. Ruwanpathirana
Sadeep Thilakarathna
Shanaka Kristombu Baduge
author_sort Gihan P. Ruwanpathirana
collection DOAJ
description Asphalt cracking poses a significant deterioration challenge in road networks. Historically, government agencies employed visual inspection methods to identify asphalt cracking. However, this approach is labour-intensive and time-consuming, prompting most authorities to transition to automated crack identification methods utilising Artificial Intelligence (A.I.). Convolutional neural networks (CNNs) have gained popularity due to their ease of implementation and enhanced accuracy. Nevertheless, CNN models were perceived as opaque entities. In this study, we employed Integrated Gradient (I.G.) maps to elucidate the workings of these models and interpret CNN-based crack image voxels that contributed to the positive (cracked) output of CNN. Our findings revealed that CNN utilised voxels within the cracks to classify the cracked image segments and highlighted the image features that CNN employed to distinguish cracks from their surroundings. These interpretation techniques provide insights into the behaviour of the CNN model, enabling the refinement of Deep Learning (DL) models.
format Article
id doaj-art-3580ea8391434aee86e1d7dedb14ba65
institution DOAJ
issn 2214-5095
language English
publishDate 2025-07-01
publisher Elsevier
record_format Article
series Case Studies in Construction Materials
spelling doaj-art-3580ea8391434aee86e1d7dedb14ba652025-08-20T03:17:44ZengElsevierCase Studies in Construction Materials2214-50952025-07-0122e0461310.1016/j.cscm.2025.e04613Interpretation and understanding of asphalt crack detection deep learning models using integrated gradient (I.G.) mapsGihan P. Ruwanpathirana0Sadeep Thilakarathna1Shanaka Kristombu Baduge2Department of Infrastructure Engineering, The University of Melbourne, Melbourne, VIC, AustraliaDepartment of Infrastructure Engineering, The University of Melbourne, Melbourne, VIC, AustraliaCorresponding author.; Department of Infrastructure Engineering, The University of Melbourne, Melbourne, VIC, AustraliaAsphalt cracking poses a significant deterioration challenge in road networks. Historically, government agencies employed visual inspection methods to identify asphalt cracking. However, this approach is labour-intensive and time-consuming, prompting most authorities to transition to automated crack identification methods utilising Artificial Intelligence (A.I.). Convolutional neural networks (CNNs) have gained popularity due to their ease of implementation and enhanced accuracy. Nevertheless, CNN models were perceived as opaque entities. In this study, we employed Integrated Gradient (I.G.) maps to elucidate the workings of these models and interpret CNN-based crack image voxels that contributed to the positive (cracked) output of CNN. Our findings revealed that CNN utilised voxels within the cracks to classify the cracked image segments and highlighted the image features that CNN employed to distinguish cracks from their surroundings. These interpretation techniques provide insights into the behaviour of the CNN model, enabling the refinement of Deep Learning (DL) models.http://www.sciencedirect.com/science/article/pii/S2214509525004115Integrated gradient mapsConvolutional neural networksArtificial IntelligenceDeep learningMachine learningAsphalt pavements
spellingShingle Gihan P. Ruwanpathirana
Sadeep Thilakarathna
Shanaka Kristombu Baduge
Interpretation and understanding of asphalt crack detection deep learning models using integrated gradient (I.G.) maps
Case Studies in Construction Materials
Integrated gradient maps
Convolutional neural networks
Artificial Intelligence
Deep learning
Machine learning
Asphalt pavements
title Interpretation and understanding of asphalt crack detection deep learning models using integrated gradient (I.G.) maps
title_full Interpretation and understanding of asphalt crack detection deep learning models using integrated gradient (I.G.) maps
title_fullStr Interpretation and understanding of asphalt crack detection deep learning models using integrated gradient (I.G.) maps
title_full_unstemmed Interpretation and understanding of asphalt crack detection deep learning models using integrated gradient (I.G.) maps
title_short Interpretation and understanding of asphalt crack detection deep learning models using integrated gradient (I.G.) maps
title_sort interpretation and understanding of asphalt crack detection deep learning models using integrated gradient i g maps
topic Integrated gradient maps
Convolutional neural networks
Artificial Intelligence
Deep learning
Machine learning
Asphalt pavements
url http://www.sciencedirect.com/science/article/pii/S2214509525004115
work_keys_str_mv AT gihanpruwanpathirana interpretationandunderstandingofasphaltcrackdetectiondeeplearningmodelsusingintegratedgradientigmaps
AT sadeepthilakarathna interpretationandunderstandingofasphaltcrackdetectiondeeplearningmodelsusingintegratedgradientigmaps
AT shanakakristombubaduge interpretationandunderstandingofasphaltcrackdetectiondeeplearningmodelsusingintegratedgradientigmaps