Automated Zebra Crossing using Artificial Intelligence

Abstract

Pedestrian safety remains a critical concern, particularly at zebra crossings where accidents often occur due to poor visibility and a lack of driver awareness. This project presents an Automated Zebra Crossing System that integrates Artificial Intelligence (AI) with Raspberry Pi technology to enhance pedestrian safety. The system utilises a webcam and a YOLO-based object detection model to identify pedestrians in real-time. When a pedestrian is detected, the Raspberry Pi triggers a flashing light and buzzer alarm to alert both pedestrians and drivers, ensuring a safer crossing experience. To improve efficiency and sustainability, the system is powered by a 12V lithium-ion battery, recharged via a solar panel, making it energy-efficient and environmentally friendly. The hardware components, including the microprocessor, relay module, and output devices, are enclosed in a waterproof electrical box to withstand various environmental conditions. Additionally, the implementation of the COCO dataset in the YOLO algorithm ensures high detection accuracy, with a confidence score threshold set above 70% to minimise false positives. The results demonstrate that the system effectively detects pedestrians and activates the warning signals in real-time, enhancing road safety. With its high commercial potential, this system can be implemented in industrial areas and later expanded to urban roads with further enhancements. Future improvements include upgrading to more advanced microprocessors, optimising image processing, and increasing public awareness through educational campaigns. Overall, this project contributes to the development of smart transportation systems, promoting safer and more efficient urban mobility.