Potholes Detection from Dashcam using Deep Learning Approach

Authors

  • Noryanti Muhammad Centre for Mathematical Sciences, Universiti Malaysia Pahang Al-Sultan Abdullah, Lebuh Persiaran Tun Khalil Yaakob, 26300 Kuantan, Pahang, Malaysia https://orcid.org/0000-0002-6112-3720 (unauthenticated)
  • Wan Maisarah Wan Alias Centre for Mathematical Sciences, Universiti Malaysia Pahang Al-Sultan Abdullah, Lebuh Persiaran Tun Khalil Yaakob, 26300 Kuantan, Pahang, Malaysia
  • Mohd Radhie Mohd Salleh INet Spatial Sdn Bhd, 41A & 41B, Jalan Flora 1/9 Johor Bahru, 81300 Skudai, Johor, Malaysia
  • Yuki Hayashida Graduate School of Engineering, Division of Information Engineering, Mie University, Japan.

DOI:

https://doi.org/10.15282/daam.v6i2.12679

Keywords:

Convolutional Neural Network (CNN), Confusion matrix, Object detection, Potholes

Abstract

Deep learning has attracted considerable interest in the previous ten years and has established itself as a leading technology in the artificial intelligence sector. In object detection based on image processing, characteristics are extracted from images, and after that, data including category, position, and motion are collected and analyzed. Object detection same goes with potholes detection. Due to the potential for crashes and injuries, potholes can be dangerous for cyclists, pedestrians, and moving automobiles. The discovery and repair of potholes may fundamentally change with the development of automated pothole detecting systems. Road repair is necessary to prevent such traffic accidents in the future. Therefore, in order to perform road maintenance, it is necessary to identify difficulties with things like potholes. The aims of this research is to identify number of potholes on asphalt or concrete roads from the images obtained for both primary and secondary data, to investigate the accuracy of the testing by using the deep learning approach which is Convolutional Neural Network (CNN) algorithm and to differentiate between normal road and potholes by evaluating them using confusion matrix which are “actual potholes”, “predicted potholes”, “actual normal” and “predicted normal”. Data collecting has been done around Universiti Teknologi Malaysia's road. Deep learning has been applied in this study is CNN (Convolutional Neural Networks), which was used as the model of this project. Results obtained are in confusion matrix which it concludes the actual number of potholes there. The actual potholes obtained were less than expected so this means more data needed if this project extended in the future. To improve and to achieve better performance of this project, data collecting can be done by adding other places so there will be more data and resolution of the images can be sharper so the results may be more accurate. To make our roadways reliable, efficient, and long-lasting, research on pothole detection is essential.

Author Biography

  • Noryanti Muhammad, Centre for Mathematical Sciences, Universiti Malaysia Pahang Al-Sultan Abdullah, Lebuh Persiaran Tun Khalil Yaakob, 26300 Kuantan, Pahang, Malaysia

    Associate Professor Ts. Dr. Noryanti Muhammad is a researcher and academic at Universiti Malaysia Pahang Al-Sultan Abdullah (UMPSA), serving as an Associate Professor at the Centre for Mathematical Sciences. She previously held leadership roles as Director of the Data Management & Analytics Centre (DMAC) and the CoE for Artificial Intelligence & Data Science. She earned her Ph.D. in Mathematical Sciences from Durham University, United Kingdom, specializing in predictive inference using copula models for bivariate data. Her research covers a wide range of areas including nonparametric predictive inference (NPI), machine learning, statistical modelling, in various applications. With over 60 publications, Dr. Noryanti has been recognised with prestigious awards such as the PERSAMA Innovation Award and medals from CITReX and ITEX exhibitions. She actively serves as a reviewer for high impact journals, contributes to editorial and scientific committees, and maintains memberships in professional organizations such as the Institute Mathematical Sciences (IMS), Institute Statistical Malaysia (ISM), Persatuan Sains dan Matematik (PERSAMA), RSS, and IEEE. Dr. Noryanti has supervised numerous Ph.D. and Master’s students and contributed to consultancy projects for industries like KANEKA, Emery Oleochemicals, and KBC, focusing on analytics, risk management, and AI solutions. She collaborates widely across faculties, industries, and institutions both nationally and internationally, including organizations such as IMR, LPPKN, SSM, CSM, Credence, Abyres, HSI, HTAA, Thaksin University, and Newcastle Australia Institute of Higher Education. She is also the author of several academic books and modules and is a frequent invited speaker at conferences and workshops.

References

[1] Holmes BD. Road traffic crashes in Federal Capital Territory, Nigeria: explanatory factors in statistical, geospatial, and qualitative analyses. The Medical College of Wisconsin. 2019.

[2] Varona J, Munoz-Bulnes J, Marcano A. Road surface type classification for smart vehicle navigation. Transportation Research Procedia. 2019;40:142–9.

[3] BH K, Kumari M, Kumari P. A review on pothole detection methods. International Journal of Research in Engineering, Science and Management. 2018;1(7):102–6.

[4] Senbagavalli M, Mahapatra S, Ruchita L. Revolutionizing road safety: an innovative approach to pothole detection and prevention using XAI and deep learning. In: 2025 4th International Conference on Sentiment Analysis and Deep Learning (ICSADL). IEEE; 2025. p. 878–84.

[5] Reddy ESTK, V R. Pothole detection using CNN and YOLO v7 algorithm. In: 2022 6th International Conference on Electronics, Communication and Aerospace Technology. IEEE; 2022.

[6] Reddy N, Varma V. Smart road maintenance: challenges and review of pothole detection methods. Journal of Civil Infrastructure. 2022;10(1):34–41.

[7] Ping P, Yang X, Gao Z. A deep learning approach for street pothole detection. In: International Conference on Big Data. IEEE; 2020.

[8] Lin TY, Dollár P, Girshick R, He K, Hariharan B, Belongie S. Feature pyramid networks for object detection. arXiv Preprint. 2016;arXiv:1612.03144.

[9] Maeda H, Sekimoto Y, Seto T, Kashiyama T, Omata H. Road damage detection and classification using deep neural networks with smartphone images. Computer-Aided Civil and Infrastructure Engineering. 2018;33(12):1127–41.

[10] Sawant S, Deshpande R, Karne K, Mehta K, Kasetwar A, Kasture A, Kasurde S. Pothole detection and road damage analysis using YOLOv9. In: International Joint Conference on Advances in Computational Intelligence. Singapore: Springer; 2025. p. 45–58.

[11] Yik W, Liew SY, Chong YW. A crowdsourced and automated approach for pothole monitoring in Malaysia. Malaysian Journal of Civil Engineering. 2021;33(1):59–72.

[12] Shaghouri M, Khodaii A, Yadollahi M. Pavement distress classification and its role in maintenance decision-making. Construction and Building Materials. 2021;276:122204.

[13] Ahmed KR. Smart pothole detection using deep learning based on dilated convolution. Sensors. 2021;21(24):8406.

[14] Ahmed A. Study on road pavement failures and pothole formation causes. International Journal of Pavement Research. 2021;6(2):45–52.

[15] Sharma R, Kanoungo R. Causes and prevention of potholes: a case study. International Journal of Advanced Research in Engineering and Management. 2015;1(4):21–5.

[16] Naveen N, Kumar V, Sinha R. Pothole detection using computer vision and machine learning. International Journal of Engineering and Technology. 2018;7(2.21):282–6.

[17] Mittal S, Bhatia A, Madan R. A survey on deep learning techniques for object detection. Archives of Computational Methods in Engineering. 2020;27(6):2089–115.

[18] Patawar A, Mehdi M, Kore B, Saval P. A pothole can be seen with two eyes: an ensemble approach to pothole detection. Machine Vision and Applications. 2025;36(3):1–20.

[19] Pawar S, Mahajan S, Pawar P, Khaire P, Dedgaonkar S, Shelke P. Road pothole detection and reporting system. In: 2025 International Conference on Emerging Smart Computing and Informatics (ESCI). IEEE; 2025. p. 1–6.

[20] Kandacharam S, Rajathilagam B, Vasudevan SK. Fusion framework for accident anticipation and incident detection in dashcam videos. SN Computer Science. 2025;6(5):1–17.

[21] Mishra M. Convolutional neural networks, explained. Medium. 2020 Sep 2.

[22] Ahmed MIB, Saraireh L, Rahman A, Al-Qarawi S, Mhran A, Al-Jalaoud J, Al-Mudaifer D, Al-Haidar F, AlKhulaifi D, Youldash M, et al. Personal protective equipment detection: A deep-learning-based sustainable approach. Sustainability. 2023;15(18):13990.

[23] Chollet F, Keras Team. Keras. GitHub. Retrieved from https://github.com/keras-team/keras, 2015.

[24] LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521(7553):436–44.

Downloads

Published

2025-09-30

Issue

Vol. 6 No. 2: September 2025

Section

Research Articles

License

Copyright (c) 2025 The Author(s)

Creative Commons License

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

How to Cite

[1]
N. Muhammad, W. M. . Wan Alias, M. R. Mohd Salleh, and Y. . Hayashida, “Potholes Detection from Dashcam using Deep Learning Approach”, Data Anal. Appl. Math., vol. 6, no. 2, pp. 1–9, Sep. 2025, doi: 10.15282/daam.v6i2.12679.

Similar Articles

You may also start an advanced similarity search for this article.