All Issue

2024 Vol.14, Issue 2 Preview Page

Scientific Paper

Asphalt pavement pothole detection using a 3 channel based bayesian-convolutional neural network

3채널 베이스 Bayesian-Convolutional Neural Network를 활용한 아스팔트 포장 포트홀 추적 연구

Byeong-Hun Woo1
,
Dong-Ho Yoo2
,
Jae-Soon Choi3*
우 병훈1
,
유 동호2
,
최 재순3*
1Researcher Professor, Department of Civil and Environmental Engineering, Hanyang University
2Ph.D. Course, Department of Civil, Construction and Environmental Engineering, North Carolina State University
3Professor, Department of Civil and Architectural Engineering, Seokyeong University
1한양대학교 건설환경공학과 연구교수
2노스캐롤라이나주립대학교 건설환경공학과 박사과정
3서경대학교 토목건축공학과 교수
*Corresponding Author
31 December 2024. pp. 192-201
PDF XML Citation
Abstract

This study presents an approach for detecting potholes in asphalt pavements using a 3-channel based Bayesian-Convolutional Neural Network (BCNN). The BCNN leverages Bayesian inference to model the uncertainty in the weights, helping prevent overfitting and ensuring robust performance, even with limited data. BCNN was trained using RGB color-scale images and achieved remarkable accuracy in identifying potholes amidst complex environments containing shadows, vehicles, and other noise. The evaluation metrics for the BCNN showed a precision of 99.36%, a recall of 98.68%, and an F1 score of 99.02%. These results were validated against both test and validation datasets with varying levels of complexity. The findings demonstrate that BCNN provides a reliable method for detecting pavement defects, outperforming traditional CNN approaches, particularly in noisy environments.

Keywords
Bayesian-convolutional neural network
Pothole
Machine learning
Asphalt
Pavement
References
1

Ali, F., Khan, Z.H., Khattak, K.S. and Gulliver, T.A. (2023). "Evaluating the Effect of Road Surface Potholes Using a Microscopic Traffic Model", Applied Sciences, 13(15), p. 8677. https://doi.org/10.3390/app13158677

10.3390/app13158677
2

Atikur Rahman Chitholian (2020). Pothole image data set, Roboflow Universe, https://public.roboflow.com/object-detection/pothole

3

Chen, G.-H., Ni, J., Chen, Z., Huang, H., Sun, Y.-L., Ip, W.H. and Yung, K.L. (2022). "Detection of Highway Pavement Damage Based on a CNN Using Grayscale and HOG Features", Sensors, 22(7), p. 2455. https://doi.org/10.3390/s22072455

10.3390/s2207245535408070PMC9002920
4

Egaji, O.A., Evans, G., Griffiths, M.G. and Islas, G. (2021). "Real-time machine learning-based approach for pothole detection", Expert Systems with Applications, 184, p. 115562. https://doi.org/10.1016/j.eswa.2021.115562

10.1016/j.eswa.2021.115562
5

Hoang, N.-D. and Tran, V.-D. (2022). "Computer vision based asphalt pavement segregation detection using image texture analysis integrated with extreme gradient boosting machine and deep convolutional neural networks", Measurement, 196, p. 111207. https://doi.org/10.1016/j.measurement.2022.111207

10.1016/j.measurement.2022.111207
6

Huang, H., Lin, L., Tong, R., Hu, H., Zhang, Q., Iwamoto, Y., Han, X., Chen, Y.-W. and Wu, J. (2020). "UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation", ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1055-1059. https://doi.org/10.1109/ICASSP40776.2020.9053405

10.1109/ICASSP40776.2020.9053405
7

Jang, C., Kim, H.G. and Woo, B.H. (2024). "Machine Learning-Based Simulation of the Air Conditioner Operating Time in Concrete Structures with Bayesian Thresholding", Materials, 17(9). https://doi.org/10.3390/ma17092108

10.3390/ma1709210838730917PMC11084204
8

Klco, P., Koniar, D., Hargas, L. and Paskala, M. (2023). "Automated Detection Of Potholes Using YOLOv5 Neural Network", Transportation Research Procedia, 74, pp. 1150-1155. https://doi.org/10.1016/j.trpro.2023.11.255

10.1016/j.trpro.2023.11.255
9

Lee, S.-Y., Le, T.H.M. and Kim, Y.-M. (2023). "Prediction and detection of potholes in urban roads: Machine learning and deep learning based image segmentation approaches", Developments in the Built Environment, 13, p. 100109. https://doi.org/10.1016/j.dibe.2022.100109

10.1016/j.dibe.2022.100109
10

Leyva, R., Sanchez, V., Epiphaniou, G. and Maple, C. (2024). "Data-agnostic Face Image Synthesis Detection using Bayesian CNNs", Pattern Recognition Letters, 183, pp. 64-70. https://doi.org/10.1016/j.patrec.2024.04.008

10.1016/j.patrec.2024.04.008
11

Nhat-Duc, H., Nguyen, Q.-L. and Tran, V.-D. (2018). "Automatic recognition of asphalt pavement cracks using metaheuristic optimized edge detection algorithms and convolution neural network", Automation in Construction, 94, pp. 203-213. https://doi.org/10.1016/j.autcon.2018.07.008

10.1016/j.autcon.2018.07.008
12

Sapkota, R., Meng, Z., Churuvija, M., Du, X., Ma, Z. and Karkee, M. (2024). "Comprehensive Performance Evaluation of YOLO11, YOLOv10, YOLOv9 and YOLOv8 on Detecting and Counting Fruitlet in Complex Orchard Environments", ArXiv, pp. 1-26. https://doi.org/10.48550/arXiv.2407.12040

10.32388/E9Y7XI
13

Sun, Z., Caetano, E., Pereira, S. and Moutinho, C. (2023). "Employing histogram of oriented gradient to enhance concrete crack detection performance with classification algorithm and Bayesian optimization", Engineering Failure Analysis, 150, p. 107351. https://doi.org/10.1016/j.engfailanal.2023.107351

10.1016/j.engfailanal.2023.107351
14

Woo, B.-H., Lee, J.-B., Kim, J.-S. and Kim, H. G. (2023). "Corrosion state assessment of the rebar via Bayesian inference", Construction and Building Materials, 392, p. 131791. https://doi.org/10.1016/j.conbuildmat.2023.131791

10.1016/j.conbuildmat.2023.131791
15

Woo, B.-H., Yoo, D.-H., Jeon, I.-K., Kim, J.-S., Yoo, K.-S. and Kim, H.G. (2022). "Estimation of the thermal conductivity of cement composites using bayesian statistical approach", Composites Part B: Engineering, 243. https://doi.org/10.1016/j.compositesb.2022.110073

10.1016/j.compositesb.2022.110073
16

Woo, B.-H., Yoo, K.-S. and Choi, J.-S. (2024). "Improving the Accuracy of Pothole Tracking and Prediction using Spiking Neural Network", Journal of Korean Asphalt Institute, 14(1), pp. 28-36. https://doi.org/10.22702/jkai.2024.14.1.4

17

Xiang, S., Jiang, S., Liu, X., Zhang, T. and Yu, L. (2022). "Spiking VGG7: Deep Convolutional Spiking Neural Network with Direct Training for Object Recognition", Electronics (Switzerland), 11(13). https://doi.org/10.3390/electronics11132097

10.3390/electronics11132097
18

Ye, W., Jiang, W., Tong, Z., Yuan, D. and Xiao, J. (2021). "Convolutional neural network for pothole detection in asphalt pavement", Road Materials and Pavement Design, 22(1), pp. 42-58. https://doi.org/10.1080/14680629.2019.1615533

10.1080/14680629.2019.1615533
19

Zhang, C. and Diao, C. (2023) "A Phenology-guided Bayesian-CNN (PB-CNN) framework for soybean yield estimation and uncertainty analysis", ISPRS Journal of Photogrammetry and Remote Sensing, 205, pp. 50-73. https://doi.org/10.1016/j.isprsjprs.2023.09.025

10.1016/j.isprsjprs.2023.09.025
20

Zhong, X., Peng, X., Yan, S., Shen, M. and Zhai, Y. (2018). "Assessment of the feasibility of detecting concrete cracks in images acquired by unmanned aerial vehicles", Automation in Construction, 89, pp. 49-57. https://doi.org/10.1016/j.autcon.2018.01.005

10.1016/j.autcon.2018.01.005
Information
  • Publisher :KOREAN ASPHALT INSTITUTE
  • Publisher(Ko) :한국아스팔트학회
  • Journal Title :Journal of the Korean Asphalt Institute
  • Journal Title(Ko) :한국아스팔트학회지
  • Volume : 14
  • No :2
  • Pages :192-201
  • Received Date : 2024-10-15
  • Revised Date : 2024-12-15
  • Accepted Date : 2024-12-31
  • DOI :https://doi.org/10.22702/jkai.2024.14.2.18
Journal Informaiton Journal of the Korean Asphalt Institute Journal of the Korean Asphalt Institute
Online submission
  • KISTI Current Status
  • KISTI Cited-by
  • KCI 문헌 유사도 검사
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close