Abstract
Every country has a significant problem with road connectivity in terms of development. The road damage assessment along particular thoroughfares has revealed that governmental efforts to preserve road quality whether through construction or maintenance are substantial. Potholes are typically the first thing noticed while assessing roadway quality in developing nations such as India. Intelligent Transportation Systems (ITS) have made significant strides in automation and computer vision compared to other methods such as radar, sensor bases, manual methods, etc. Recent research has shown that intelligent transport systems perform exceptionally well, especially in pothole detection and assessment. Recent advances in artificial intelligence, particularly machine learning and deep learning, have advanced robotics and automation. Modern technology has produced better results in production and cost-efficiency than traditional methods. The paper emphasizes the need for better road maintenance to decrease accidents caused by potholes on Indian roadways. Intelligent Transportation Systems (ITS) face road abnormalities beyond road damages that pose safety issues. Detecting and managing surface fissures, potholes, road signs, landslides, and animal crossings are the concerns. Deep learning and artificial intelligence can improve Intelligent Transportation Systems by providing a holistic approach to road concerns. Technology should make transportation networks safer and more efficient. The research aims to examine the effectiveness of three deep-learning object identification frameworks (YOLOv5, YOLOv6, and YOLOv7) in detecting potholes. The results demonstrate that deep learning methods are highly effective for identifying road damage within the Intelligent Transport System, especially potholes. The dataset included in this study consists of photographs depicting potholes observed on diverse categories of roadways, namely municipal, state, and national highways. The empirical findings indicate that YOLOv7 has superior efficiency as a pothole detector, with a precision rate of 93%.
Similar content being viewed by others
Data Availability
All the data and materials are available to the authors.
References
NHAI National High way Authority of India (NHAI). https://nhai.gov.in
The Economic Times. Supreme Court takes note of 3,597 deaths due to pothole-related accidents in 2017 - The Economic Times. Econ Times; 2018.
The Indian Express. Union transport minister nitin gadkari inaugurates CRRI’ s pothole repair machine. The Indian Express; 2022.
Patel S. 2022. Pothole image data-set, Kaggle. https://www.kaggle.com/datasets/sachinpatel21/pothole-image-dataset. Accessed 20 Aug 2022.
Colagrande S, Ranalli D, Tallini M. GPR research on damaged road pavements built in cut and fill sections. Transp Res Procedia. 2020;45:30–7. https://doi.org/10.1016/j.trpro.2020.02.059.
Ravi R, Bullock D, Habib A. Highway and airport runway pavement inspection using mobile lidar. Int Arch Photogramm Remote Sens Spat Inf Sci - ISPRS Arch. 2020;43:349–54. https://doi.org/10.5194/isprs-archives-XLIII-B1-2020-349-2020.
Koch C, Brilakis I. Pothole detection in asphalt pavement images. Adv Eng Inf. 2011;25:507–15. https://doi.org/10.1016/j.aei.2011.01.002.
Ozoglu F, Gökgöz T. Detection of road potholes by applying convolutional neural network method based on road vibration data. Sensors (Basel). 2023. https://doi.org/10.3390/s23229023.
Dhall R, Solanki VK. An IoT based predictive connected car maintenance approach. Int J Interact Multimed Artif Intell. 2017;4:16. https://doi.org/10.9781/ijimai.2017.433.
You C, Wen C, Luo H, Wang CLJ. Rapid traffic damage inspection in natural scenes using mobile laser scanning data. IEEE.; 2017. p. 62271–6274.
Ikidid A, El FA, Sadgal M. Multi-agent and fuzzy inference-based framework for traffic light optimization. Int J Interact Multimed Artif Intell. 2023;8:88–97. https://doi.org/10.9781/ijimai.2021.12.002.
Nienaber S, Booysen MJ. Detecting potholes using image processing techniques and real-world footage. Singapore: Springer Nature; 2021. p. 893–902.
Omar M, Kumar P. Image features selection based on computer vision techniques to detect potholes for intelligent transport system. Int Conf Inf Sci Commun Technol Appl Trends Oppor ICISCT. 2021. https://doi.org/10.1109/ICISCT52966.2021.9670066.
Bibi R, Saeed Y, Zeb A, et al. Edge ai-based automated detection and classification of road anomalies in vanet using deep learning. Comput Intell Neurosci. 2021. https://doi.org/10.1155/2021/6262194.
Guo L, Ge PS, Zhang MH, et al. Pedestrian detection for intelligent transportation systems combining AdaBoost algorithm and support vector machine. Expert Syst Appl. 2012;39:4274–86. https://doi.org/10.1016/j.eswa.2011.09.106.
Zhang Z, Ai X, Chan CK, Dahnoun N. An efficient algorithm for pothole detection using stereo vision ICASSP Int Conf Acoust Speech Signal Process Proc. IEEE; 2014. p. 564–8.
Rajagopal A, Ramachandran A, Shankar K, et al. Fine-tuned residual network-based features with latent variable support vector machine-based optimal scene classification model for unmanned aerial vehicles. IEEE Access. 2020;8:118396–404. https://doi.org/10.1109/ACCESS.2020.3004233.
Maeda H, Kashiyama T, Sekimoto Y, et al. Generative adversarial network for road damage detection. Comput Civ Infrastruct Eng. 2021;36:47–60. https://doi.org/10.1111/mice.12561.
Wang YJ, Ding M, Kan S, et al. Deep proposal and detection networks for road damage detection and classification, 2018 IEEE international conference on big data (Big data). USA: IEEE; 2018. p. 5224–7.
Gajjar K, Van NT, Wilm T, Mercorelli P. Vision-based deep learning algorithm for detecting potholes. J Phys Conf Ser. 2022. https://doi.org/10.1088/1742-6596/2162/1/012019.
Zhang H, Wang Y, Liu Y. IFD : an intelligent fast detection for real-time image information in industrial IoT. Appl Sci. 2022;12(15):7847.
Asad MH, Khaliq S, Yousaf MH, et al. Pothole detection using deep learning: a real-time and ai-on-the-edge perspective. Adv Civ Eng. 2022. https://doi.org/10.1155/2022/9221211.
Rajagopal A, Joshi GP, Ramachandran A, et al. A deep learning model based on multi-objective particle swarm optimization for scene classification in unmanned aerial vehicles. IEEE Access. 2020;8:135383–93. https://doi.org/10.1109/ACCESS.2020.3011502.
Pillai MS, Chaudhary G, Khari M, Crespo RG. Real-time image enhancement for an automatic automobile accident detection through CCTV using deep learning. Soft Comput. 2021;25:11929–40. https://doi.org/10.1007/s00500-021-05576-w.
Heo DH, Choi JY, Kim SB, et al. Image-based pothole detection using multi-scale feature network and risk assessment. Electron. 2023. https://doi.org/10.3390/electronics12040826.
Redmon J, Divvala S, Girshick R, Farhadi A. You only look once Unified, real-time object detection. Proc Comput Soc Conf Comput Vis Pattern Recognit 2016-Decem. IEEE; 2016.
Redmon J, Farhadi A. Yolo V2.0. Cvpr. 2016;2017:187–213.
Redmon J, Farhadi A. YOLOv3: an incremental improvement. arXiv preprint. 2018. arXiv: 1804.02767.
Bochkovskiy A, Wang C-Y, Liao H-YM. YOLOv4: optimal speed and accuracy of object detection. Comput Vis Pattern Recognit arXiv. 2020;2004:1–17.
Omar M, Kumar P. Detection of roads potholes using YOLOv4 2020. Int Conf Inf Sci Commun Technol ICISCT. IEEE; 2020.
Li S, Li Y, Li Y, et al. YOLO-FIRI: Improved YOLOv5 for infrared image object detection. IEEE Access. 2021;9:141861–75. https://doi.org/10.1109/ACCESS.2021.3120870.
Li C et al. YOLOv6: a single-stage object detection framework for industrial applications. 2022. arXiv: 2209.02976.
Wang CY, Bochkovskiy A, Liao HYM. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. IEEE; 2022.
Ahmed KR. Smart pothole detection using deep learning based on dilated convolution. Sensors. 2021;21(24):8406. https://doi.org/10.3390/s21248406.
Funding
No funding is provided for this research.
Author information
Authors and Affiliations
Contributions
Mohd Omar: Implementation, Writing—original draft, Pradeep Kumar: Suggestion, Proofreading.
Corresponding author
Ethics declarations
Conflict of Interest
The authors declare that they do not have any competing interests.
Research Involving Human and/or Animals
No involvement of human and animals for this research.
Informed Consent
The authors give their consent to publish this original research work.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This article is part of the topical collection “Security for Communication and Computing Application” guest edited by Karan Singh, Ali Ahmadian, Ahmed Mohamed Aziz Ismail, R S Yadav, Md. Akbar Hossain, D. K. Lobiyal, Mohamed Abdel-Basset, Soheil Salahshour, Anura P. Jayasumana, Satya P. Singh, Walid Osamy, Mehdi Salimi and Norazak Senu.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Omar, M., Kumar, P. PD-ITS: Pothole Detection Using YOLO Variants for Intelligent Transport System. SN COMPUT. SCI. 5, 552 (2024). https://doi.org/10.1007/s42979-024-02887-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s42979-024-02887-1