Detection of Unathorized Vechile in BRT Lane

The increasing number of unauthorized vehicles in the Bus Rapid Transit (BRT) lanes creates a significant challenge to urban transportation. Due to this efficiency and quality of safety decreases. Our project aims to develop an automated system that detects unauthorized vehicles that come into BRT lanes using deep learning techniques like YOLOv8. The system employs combination video surveillance, image processing, and deep learning models to accurately identify and track vehicles that violate the BRT regulations. The proposed solution enhances traffic management and reduces the incidence of unauthorized lane usage by integrating the real- time data processing and alert mechanisms. The system employs a dataset from Roboflow, extended with a self-made dataset containing images of various vehicles in BRT lanes. We utilize YOLOv8 for object detection to accurately identify and classify vehicles. Also, Easy-OCR is used to recognize text from license plates for validation and verification against authorized vehicle list. Our preliminary results demonstrate that the system achieves 80% accuracy in detecting unauthorized vehicles. Our solution is urban traffic management by reducing illegal lane usage and ensuring easier public transport operations.