Evaluation of deep learning models for classification of asphalt pavement distresses

Article


Apeagyei, A., Ademolake, T. E. and Adom-Asamoah, M. 2023. Evaluation of deep learning models for classification of asphalt pavement distresses. International Journal of Pavement Engineering. 24 (Art. 2180641). https://doi.org/10.1080/10298436.2023.2180641
AuthorsApeagyei, A., Ademolake, T. E. and Adom-Asamoah, M.
Abstract

Transfer learning (TL) offers a convenient methodology for exploiting the capability of deep convolutional neural networks (DCNNs) for many image classification tasks including the classification of pavement distresses. Seven state-of-the-art DCNNs were retrained to classify asphalt pavement distresses grouped into eight classes using TL techniques. The aim was to evaluate the predictive performances of the selected DCNNs in order to provide some guidelines on selection of DCNNs for pavement application. The results show some existing DCNN’s are better than others for developing pavement distress classification models using the specific TL approach adopted in the study. The predictive ability of each model varied depending on distress class as some models with very low overall accuracy showed excellent results for individual distress class(s). Based on a combination of various performance metrics including F1-score, area under ROC curve, optimal operating threshold, training time, and model size, the best performing network had a relative score that was found to be significantly higher than the next two top-performing models. The best-performing networks were characterised by lower proportions of false negative values, low ambiguity scores, and well-defined t-SNE clusters that showed clear separation between the eight different pavement distress classes considered.

KeywordsAsphalt; asphalt pavements; pavement distresses; pavement distresses classification; F1-score; transfer learning; rutting; fatigue cracking; transverse cracking; longitudinal cracking
JournalInternational Journal of Pavement Engineering
Journal citation24 (Art. 2180641)
ISSN1477-268X
Year2023
PublisherTaylor & Francis
Publisher's version
License
File Access Level
Anyone
Digital Object Identifier (DOI)https://doi.org/10.1080/10298436.2023.2180641
Publication dates
Online24 Feb 2023
Publication process dates
Accepted10 Feb 2023
Deposited02 Mar 2023
Copyright holder© 2023 The Author(s)
Permalink -

https://repository.uel.ac.uk/item/8vqqv

Download files


Publisher's version
  • 89
    total views
  • 105
    total downloads
  • 7
    views this month
  • 10
    downloads this month

Export as

Related outputs

Road Deterioration detection A Machine Learning-Based System for Automated Pavement Crack Identification and Analysis
Ganeshan, D., Sharif, S., Apeagyei, A. and Elmedany, W. 2023. Road Deterioration detection A Machine Learning-Based System for Automated Pavement Crack Identification and Analysis. 3ICT 2023: International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies. University of Bahrain, Bahrain 20 - 21 Nov 2023 IEEE. https://doi.org/10.1109/3ICT60104.2023.10391802
Artificial Intelligence Applications in Road Traffic Forecasting: A Review of Current Research
Khairi, S., Sharif, S., Apeagyei, A. and Abbas, A. 2023. Artificial Intelligence Applications in Road Traffic Forecasting: A Review of Current Research. 3ICT 2023: International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies. University of Bahrain, Bahrain 20 - 21 Nov 2023 IEEE. https://doi.org/10.1109/3ICT60104.2023.10391677
Utilising Convolutional Neural Networks for Pavement Distress Classification and Detection
Sharif, S., Emiola, D. I., Zoto, A., Apeagyei, A. and Elmedany, W. 2023. Utilising Convolutional Neural Networks for Pavement Distress Classification and Detection. 3ICT 2023: International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies. University of Bahrain, Bahrain 20 - 21 Nov 2023 IEEE. https://doi.org/10.1109/3ICT60104.2023.10391401
Cold Recycling of Reclaimed Asphalt Pavements
Tebaldi, G., Dave, E., Hugener, M., Falchetto, A. C., Perraton, D., Grilli, A., Lo Presti, D., Pasetto, M., Loizos, A., Jenkins, K., Apeagyei, A., Grenfell, J. and Bocci, M. 2018. Cold Recycling of Reclaimed Asphalt Pavements. in: Partl, M. N., Porot, L., Di Benedetto, H., Canestrari, F., Marsac, P. and Tebaldi, G. (ed.) Testing and Characterization of Sustainable Innovative Bituminous Materials and Systems: State-of-the-Art Report of the RILEM Technical Committee 237-SIB Springer. pp. 239-296
Recommendation of RILEM TC237-SIB on cohesion test of recycled asphalt
Tebaldi, G., Dave, E., Cannone Falchetto, A., Hugener, M., Perraton, D., Grilli, A., Lo Presti, D., Pasetto, M., Loizos, A., Jenkins, K., Apeagyei, A., Grenfell, J. and Bocci, M. 2018. Recommendation of RILEM TC237-SIB on cohesion test of recycled asphalt. Materials and Structures. 51 (Art. 117). https://doi.org/10.1617/s11527-018-1238-4
Physical and rheological characterization of carbonated bitumen for paving applications
Apeagyei, A. and Airey, Gordon D. 2018. Physical and rheological characterization of carbonated bitumen for paving applications. Materials & Design. 140, pp. 345-356. https://doi.org/10.1016/j.matdes.2017.11.069
Moisture damage evaluation of aggregate–bitumen bonds with the respect of moisture absorption, tensile strength and failure surface
Zhang, J., Airey, G. D., Grenfell, J. and Apeagyei, A. 2017. Moisture damage evaluation of aggregate–bitumen bonds with the respect of moisture absorption, tensile strength and failure surface. Road Materials and Pavement Design. 18 (4), pp. 833-848. https://doi.org/10.1080/14680629.2017.1286441
Development of a composite substrate peel test to assess moisture sensitivity of aggregate–bitumen bonds
Zhang, J., Airey, G. D., Grenfell, J., Apeagyei, A. and Barrett, M. 2016. Development of a composite substrate peel test to assess moisture sensitivity of aggregate–bitumen bonds. International Journal of Adhesion and Adhesives. 68, pp. 133-141. https://doi.org/10.1016/j.ijadhadh.2016.02.013
Time dependent viscoelastic rheological response of pure, modified and synthetic bituminous binders
Airey, G. D., Grenfell, J. R. A., Apeagyei, A., Subhy, A. and Lo Presti, D. 2016. Time dependent viscoelastic rheological response of pure, modified and synthetic bituminous binders. Mechanics of Time-Dependent Materials. 20 (3), pp. 455-480. https://doi.org/10.1007/s11043-016-9295-y
Moisture sensitivity examination of asphalt mixtures using thermodynamic, direct adhesion peel and compacted mixture mechanical tests
Zhang, J., Airey, G. D., Grenfell, J. and Apeagyei, A. 2016. Moisture sensitivity examination of asphalt mixtures using thermodynamic, direct adhesion peel and compacted mixture mechanical tests. Road Materials and Pavement Design. 19 (1), pp. 120-138. https://doi.org/10.1080/14680629.2016.1249510
Application of Fickian and non-Fickian diffusion models to study moisture diffusion in asphalt mastics
Apeagyei, A., Grenfell, J. R. A. and Airey, G. D. 2015. Application of Fickian and non-Fickian diffusion models to study moisture diffusion in asphalt mastics. Materials and Structures. 48 (5), pp. 1461-1474. https://doi.org/10.1617/s11527-014-0246-2
Moisture damage assessment using surface energy, bitumen stripping and the SATS moisture conditioning procedure
Grenfell, J., Apeagyei, A. and Airey, G. 2015. Moisture damage assessment using surface energy, bitumen stripping and the SATS moisture conditioning procedure. International Journal of Pavement Engineering. 16 (5), pp. 411-431. https://doi.org/10.1080/10298436.2015.1007235
Influence of aggregate absorption and diffusion properties on moisture damage in asphalt mixtures
Apeagyei, A., Grenfell, J. R. A. and Airey, G. D. 2015. Influence of aggregate absorption and diffusion properties on moisture damage in asphalt mixtures. Road Materials and Pavement Design. 16 (Sup 1), pp. 404-422. https://doi.org/10.1080/14680629.2015.1030827
Influence of aggregate mineralogical composition on water resistance of aggregate–bitumen adhesion
Zhang, J., Apeagyei, A., Airey, G. D. and Grenfell, J. R. A. 2015. Influence of aggregate mineralogical composition on water resistance of aggregate–bitumen adhesion. International Journal of Adhesion and Adhesives. 62, pp. 45-54. https://doi.org/10.1016/j.ijadhadh.2015.06.012
Moisture-induced strength degradation of aggregate–asphalt mastic bonds
Apeagyei, A., Grenfell, J. R. A. and Airey, G. D. 2014. Moisture-induced strength degradation of aggregate–asphalt mastic bonds. Road Materials and Pavement Design. 15 (Sup 1), pp. 239-262. https://doi.org/10.1080/14680629.2014.927951
Observation of reversible moisture damage in asphalt mixtures
Apeagyei, A., Grenfell, J. R. A. and Airey, G. D. 2014. Observation of reversible moisture damage in asphalt mixtures. Construction and Building Materials. 60, pp. 73-80. https://doi.org/10.1016/j.conbuildmat.2014.02.033
Examination of moisture sensitivity of aggregate–bitumen bonding strength using loose asphalt mixture and physico-chemical surface energy property tests
Liu, Y., Apeagyei, A., Ahmad, N., Grenfell, J. and Airey, G. 2013. Examination of moisture sensitivity of aggregate–bitumen bonding strength using loose asphalt mixture and physico-chemical surface energy property tests. International Journal of Pavement Engineering. 15 (7), pp. 657-670. https://doi.org/10.1080/10298436.2013.855312
Assessing asphalt mixture moisture susceptibility through intrinsic adhesion, bitumen stripping and mechanical damage
Grenfell, J., Ahmad, N., Liu, Y., Apeagyei, A., Large, D. and Airey, G. 2013. Assessing asphalt mixture moisture susceptibility through intrinsic adhesion, bitumen stripping and mechanical damage. Road Materials and Pavement Design. 15 (1), pp. 131-152. https://doi.org/10.1080/14680629.2013.863162