International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 10 | Oct 2024
p-ISSN: 2395-0072
www.irjet.net
REAL TIME POTHOLE DETECTION USING AI AND ML Miss. Pallavi Chede1, Miss. Shraddha Chitale2, Miss. Arti Mansuke3, Miss. Pratiksha Ghodake4, Prof. S. P. Vidhate5 1,2,3,4Department of Computer Engineering, Shri Chhatrapati Shivaji Maharaj College of Engineering, India
, India --------------------------------------------------------------------------***----------------------------------------------------------------------pothole locations using GPS, providing municipalities and Abstract 5Professor, Dept. of Computer Engineering, Shri Chhatrapati Shivaji Maharaj college of Engineering
road maintenance crews with real-time data on road conditions.
Accidents as a result of choppy road conditions can harm drivers, passengers, and peoples.
This project leverages computer vision techniques and deep learning algorithms to train models capable of identifying potholes in various lighting and weather conditions. By automating the detection process, the system can potentially reduce human errors and the time required to inspect roads manually, contributing to safer driving environments and more efficient road maintenance
Monitoring the state of the roads is important to developing a network of safe and enjoyable mobility. Road accidents are occur due to poor road situations. Due to the growing wide variety of potholes, coincidence charges are growing 12 months after year. Because road preservation is generally completed manually, it takes a long time, includes attempt, and is liable to human mistake. Since potholes are one of the main reason of road accidents. using machine learning and computer vision techniques we can discover the pothole and give alert hence road safety is increase. A system is measuring pothole length, breath, depth and detect and classify them. To discover potholes, the system uses two algorithm YOLO (You Only Look Once) and CNN (Convolutional Neural Network ).
2. RELATED WORK [1] Amit Mishra (2023) “ Road to Repair (R2R) ”, An Afrocentric Sensor-Based Solution to Enhanced Road Maintenance. It is used sensor to detect pothole and store the data in sensors. [2] Neng cheng Chen , Xiang Zhang , and Yuhang Guan (2022) “Real-Time Road Pothole Mapping Based on Vibration Analysis in Smart City” real-time road pothole mapping system using vibration analysis and Spatio-temporal trajectory fusion.
Keywords: You Only Look Once (YOLO), Convolutional Neural Network (CNN), Machine Learning, Pothole Detection , computer vision.
[3] Zener Sukra Lie1, Winda Astuti1 and Sofyan Tan1 (2020) “Pothole detection system design with proximity sensor to provide motorcycle with warning system and increase road safety driving”.
1.INTRODUCTION In this project, we propose a Live Pothole Detection System using Artificial Intelligence (AI) and Machine Learning (ML) techniques to identify potholes in real-time
[4] S. M. R. Ghosh (2019) "Deep Learning for Computer Vision”. This review paper provides a overview of some of the most significant deep learning and in this the computer vision problems are used that is CNN.
The system uses YOLO (You Only Look Once), a realtime object detection algorithm, to identify potholes in road images, and OpenCV for depth estimation. The integration of these tools enables the system to detect potholes without the need for specialized sensors or expensive equipment.
[5] Moazzam et al. (2013) “Metrology and Visualization of Potholes using the Microsoft Kinect Sensor” proposed a 3D reconstruction method for pothole detection using stereo images captured by cameras. By analyzing the depth and surface irregularities, the system could identify potholes with high accuracy.
This system uses a camera mounted on a vehicle to capture live video of the road ahead, processes the images using advanced image processing techniques, and machine learning models (YOLO ,CNN) to detect potholes. The system aims to automatically detect and classify road conditions, differentiating between normal road surfaces and potholes.
[6] Asutosh saha , Gaurav sharma(2021), “Smart implementation of computer vision and machine learning for pothole detection” Smart implementation of computer vision and machine
Once a pothole is detected, the system can trigger alerts to the driver, allowing them to take immediate action to avoid potential damage. Additionally, the system can map
© 2024, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 812