Pothole Detection and Reporting System Implementation Using Yolov8 and TensorFlow.js

Authors

  • Ruhil N Dept. of Computer Science and Engineering, Dr Akhilesh Das Gupta Institute of Professional Studies (Previously ADGITM), New Delhi, India
  • Sahni D Dept. of Computer Science and Engineering, Dr Akhilesh Das Gupta Institute of Professional Studies (Previously ADGITM), New Delhi, India
  • Anushaka A Dept. of Computer Science and Engineering, Dr Akhilesh Das Gupta Institute of Professional Studies (Previously ADGITM), New Delhi, India
  • Wadhwa A Dept. of Computer Science and Engineering, Dr Akhilesh Das Gupta Institute of Professional Studies (Previously ADGITM), New Delhi, India
  • Sharma A Dept. of Computer Science and Engineering, Dr Akhilesh Das Gupta Institute of Professional Studies (Previously ADGITM), New Delhi, India

DOI:

https://doi.org/10.26438/ijcse/v11i12.2631

Keywords:

Object detectio, YOLOv8, TensorFlow.js,, Road safety,, Pothole detection.

Abstract

Potholes present a substantial hazard to both road safety and the structural integrity of vehicles. This paper introduces a novel approach to pothole detection leveraging YOLOv8, an object detection algorithm, and TensorFlow.js. The proposed system aims to detect potholes accurately and swiftly by analysing live video feeds. The trained model exhibits promising performance metrics in pothole detection, with the bounding box precision at 0.822 and the mean Average Precision (mAP) value of 0.847, highlighting the model`s robustness in accurately localizing potholes. The proposed pothole detection system presents a promising solution for proactive road maintenance and safety enhancement. Its efficiency in real-time detection, combined with the adaptability of TensorFlow.js, holds the potential for widespread implementation, contributing to mitigating road hazards and infrastructure maintenance. The use of Tensorflow.js allows JavaScript developers to work with YOLOv8 reducing the dependency on Python for this purpose. The Pothole Detection and Reporting System with YOLOv8 and Tensorflow.js provides quite promising results.

References

[1] Anas Al Shaghouri, Rami Alkhatib, Samir Berjaoui, "Real-Time Pothole Detection Using Deep Learning," 2021. https://doi.org/10.48550/arXiv.2107.06356

[2] O. M. Khare, S. Gandhi, A. M. Rahalkar, S. Mane, "YOLOv8-Based Visual Detection of Road Hazards: Potholes, Sewer Covers, and Manholes," 2023.

[3] S. R. Kuthyar, R. S., V. Rasika, S. Manjesh, R. Girimaji, R. Davis Arjun, P. S., "An Intelligent Pothole Detection and Alerting System using Mobile Sensors and Deep Learning," 2021.

[4] R. Hiremath, K. Malshikare, M. Mahajan, "A Smart App for Pothole Detection Using Yolo Model," 2021.

[5] A. Kumar, S. Kumar, "Road quality management using mobile sensing," International Conference on Innovations in Information Embedded and Communication Systems, 2018.

[6] A. Mendis, G. Starzadnis, G. K. Anoris, "Real-time pothole detection using Android smartphones and accelerometers," 12th International Conference on ITS Telecommunications (ITST), pp. 668-672, 2019.

[7] B. Piao, K. Aihara, "Detecting the road surface condition by using mobile crowd sensing with a drive recorder," IEEE 20th International Conference on Intelligent Transportation Systems, 2019.

[8] Collinson Colin M. Agbesi, Ebenezer K. Gavua, Seth Okyere-Dankwa, Kwame Anim Appiah, Kofi Adu-Manu Sarpong, “Pothole Detection, Reporting and Management using Internet of Things: Prospects and Challenges”, International Journal of Emerging Science and Engineering (IJESE), 2017.

[9] Jetashri R. Gandhi, U. K. Jaliya, D. G. Thakore, “A Review Paper on Pothole Detection Methods”, International Journal of Computer Sciences and Engineering (IJCSE), 2019

[10] Young-Mok Kim, Young-Gil Kim, Seung-Yong Son, Soo-Yeon Lim, Bong-Yeol Choi and Doo-Hyun Choi, “Review of Recent Automated Pothole-Detection Methods”, Applied Sciences, 2022.

[11] Hsiu-Wen Wang, Chi-Hua Chen,Ding-Yuan Cheng, Chun-Hao Lin and Chi-Chun Lo, “A Real-Time Pothole Detection Approach for Intelligent Transportation System”, ResearchGate, 2015.

Downloads

Published

2023-12-31
CITATION
DOI: 10.26438/ijcse/v11i12.2631
Published: 2023-12-31

How to Cite

[1]
N. Ruhil, D. Sahni, A. Anushaka, A. Wadhwa, and A. Sharma, “Pothole Detection and Reporting System Implementation Using Yolov8 and TensorFlow.js”, Int. J. Comp. Sci. Eng., vol. 11, no. 12, pp. 26–31, Dec. 2023.

Issue

Vol. 11 No. 12 (2023): IJCSE December Edition

Section

Research Article

License

Creative Commons License

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

Authors contributing to this journal agree to publish their articles under the Creative Commons Attribution 4.0 International License, allowing third parties to share their work (copy, distribute, transmit) and to adapt it, under the condition that the authors are given credit and that in the event of reuse or distribution, the terms of this license are made clear.