Detection of Lane and Speed Breaker Warning System for Vehicles Using Machine Learning

Abstract

With the rapid advancement of vehicle technologies, ensuring the safety of these vehicles on roads has become a paramount concern. One of the critical aspects of safe driving is the accurate detection of lanes and potential road hazards, such as speed breakers. In this study, we propose a Lane and Speed Breaker Warning System (LSBWS) that employs machine learning algorithms to enhance the perception capabilities of vehicles. The LSBWS utilizes a combination of computer vision and machine learning techniques to detect and analyze road lanes, speed breakers in real-time and also a real time object detection on road. The system utilizes a camera sensor to capture the road scene ahead and then employs image processing algorithms to identify lane markings and speed breakers, objects on road. Random Sample Consensus Algorithm is used for the lane detection and tracking for speed breaker detection YOLOv4 is employed to accurately detect and classify these features within the captured images and for the object detection YOLOv5 is used for detecting the real time objects and classify them.