Evaluation of automated detection of pavement defects using YOLOv3: impact of collection techniques

Authors

  • Gabriel Tavares de Melo Freitas Instituto Federal do Ceará
  • Ernesto Ferreira Nobre Júnior Universidade Federal do Ceará
  • Aline Calheiros Espíndola Universidade Federal de Alagoas

DOI:

https://doi.org/10.58922/transportes.v32i2.2796

Keywords:

Pavement, Data collection, Deep learning

Abstract

This study involved training six neural networks with tailored configurations to automatically detect problems in pavements, utilizing the YOLOv3 framework. The acquisition of images and videos depicting pavement defects was conducted using smartphones and action cameras, leading to the organization of six distinct datasets. Every neural network was subjected to training and validation with the goal of attaining optimal accuracy in automated object detection. Implementing YOLOv3 facilitated effective defect surveys, enhancing the assessment of pavement quality, and offering valuable information for decision-making in road transport management. Upon concluding the investigation, it was determined that the framing method with the highest efficacy attained a precision rate of 98%. The results demonstrate the efficacy of YOLOv3 in accurately detecting defects, underscoring the significance of data collecting and framing methods, and adding to the current body of knowledge on automated pavement defect detection.

Downloads

Download data is not yet available.

References

Balbo, J.T. (2007) Pavimentação Asfáltica: Materiais, Projeto e Restauração. São Paulo: Oficina de Textos.

Espíndola, A.C.; G.T.M. Freitas and E.F. Nobre Jr. (2021) Pothole and patch detection on asphalt pavement using deep convolutional neural network. In Proceedings of the Joint XLII Ibero-Latin-American Congress on Computational Methods in Engineering; III Pan- American Congress on Computational Mechanics, ABMEC-IACM. Rio de Janeiro: ABMEC, p. 1-7. Available at: (accessed 03/23/2022).

Everingham, M.; L.V. Gool; C.K.I. Williams et al. (2010) The pascal visual object classes (voc) challenge. International Journal of Computer Vision, v. 88, n. 2, p. 303-338. DOI: 10.1007/s11263-009-0275-4. DOI: https://doi.org/10.1007/s11263-009-0275-4

Haykin, S. (1998) Neural Networks: A Comprehensive Foundation. Hoboken: Prentice Hall PTR.

Hoang, N.D. (2018) An artificial intelligence method for asphalt pavement pothole detection using least squares support vector machine and neural network with steerable filter-based feature extraction. Advances in Civil Engineering, v. 2018, p. 7419058. DOI: http://doi.org/10.1155/2018/7419058. DOI: https://doi.org/10.1155/2018/7419058

Khan, A.I. and S. Al-Habsi (2020) Machine learning in computer vision. Procedia Computer Science, v. 167, n. 13, p. 1444-1151. DOI: 10.1016/j.procs.2020.03.355. DOI: https://doi.org/10.1016/j.procs.2020.03.355

Maeda, H.; Y. Sekimoto; T. Seto et al. (2018) Road damage detection using deep neural networks with images captured through a smartphone. Computer-Aided Civil and Infrastructure Engineering, v. 33, p. 1127-41. DOI: 10.1111/mice.12387. DOI: https://doi.org/10.1111/mice.12387

Padilla, R.; S.L. Netto and E.A.D. Silva (2020) A survey on performance metrics for object-detection algorithms. In Proceedings of the 2020 International Conference on Systems, Signals and Image Processing (IWSSIP). USA: IEEE, p. 237-242. DOI: 10.1109/ IWSSIP48289.2020.9145130. DOI: https://doi.org/10.1109/IWSSIP48289.2020.9145130

Paterson, W.D. (1987) Road Deterioration and Maintenance Effects: Models for Planning and Management. Baltimore: The Johns Hopkins University Press.

Prince, S.J. (2012) Computer Vision: Models, Learning, and Inference. Cambridge: Cambridge University Press. DOI: 10.1017/ CBO9780511996504. DOI: https://doi.org/10.1017/CBO9780511996504

Radovic, M.; O. Adarkwa and Q. Wang (2017) Object recognition in aerial images using convolutional neural networks. Journal of Imaging, v. 3, n. 2, p. 21. DOI: 10.3390/jimaging3020021. DOI: https://doi.org/10.3390/jimaging3020021

Redmon, J.; S. Divvala; R. Girshick et al. (2016) You only look once: unified, real-time object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). USA: IEEE, pp. 779-788. Available at: <https://www.cv-foundation. org/openaccess/content_cvpr_2016/html/Redmon_You_Only_Look_CVPR_2016_paper.html> (accessed 03/19/2021). DOI: https://doi.org/10.1109/CVPR.2016.91

Sasaki, Y. (2007) The truth of the F-measure. Teach Tutor Mater, v. 1, n. 5, p. 1-5. Available at: <https://nicolasshu.com/assets/ pdf/Sasaki_2007_The%20Truth%20of%20the%20F-measure.pdf> (accessed 02/12/2021).

Sholevar, N.; A. Golroo and S.R. Esfahani (2022) Machine learning techniques for pavement condition evaluation, Automation in Construction, v. 136, p. 104190. DOI: 10.1016/j.autcon.2022.104190. DOI: https://doi.org/10.1016/j.autcon.2022.104190

Downloads

Published

2024-05-21

How to Cite

Tavares de Melo Freitas, G., Ferreira Nobre Júnior, E., & Calheiros Espíndola, A. (2024). Evaluation of automated detection of pavement defects using YOLOv3: impact of collection techniques. TRANSPORTES, 32(2). https://doi.org/10.58922/transportes.v32i2.2796

Issue

Vol. 32 No. 2 (2024)

Section

Artigos

License

Copyright (c) 2024 Gabriel Tavares de Melo Freitas, Ernesto Ferreira Nobre Júnior, Aline Calheiros Espíndola

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Authors who submit papers for publication by TRANSPORTES agree to the following terms:

  1. Authors retain copyright and grant TRANSPORTES the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

  2. Authors may enter into separate, additional contractual arrangements for the non-exclusive distribution of this journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in TRANSPORTES.

  3. Authors are allowed and encouraged to post their work online (e.g., in institutional repositories or on their website) after publication of the article. Authors are encouraged to use links to TRANSPORTES (e.g., DOIs or direct links) when posting the article online, as TRANSPORTES is freely available to all readers.

  4. Authors have secured all necessary clearances and written permissions to published the work and grant copyright under the terms of this agreement. Furthermore, the authors assume full responsibility for any copyright infringements related to the article, exonerating ANPET and TRANSPORTES of any responsibility regarding copyright infringement.

  5. Authors assume full responsibility for the contents of the article submitted for review, including all necessary clearances for divulgation of data and results, exonerating ANPET and TRANSPORTES of any responsibility regarding to this aspect.