TRAFFIC SIGNALVIOLATION DETECTION SYSTEM

Abstract

This abstract proposes the rise in urbanization has led to a significant increase in vehicular traffic, which in turn has caused a surge in traffic rule violations, particularly signal violations at intersections which have become a growing concern globally. These infractions contribute to road accidents, property damage, and loss of life, highlighting the urgent need for automated monitoring systems. To address this critical issue and help prevent such consequences, traffic violation detection systems are essential. The proposed system utilizes YOLOv3(You Only Look Once, Version 3), an advanced object detection algorithm, to identify violations such as signal jumping, speeding, and vehicle count monitoring. YOLOv3's single-stage detection approach, which divides input images into grids and predicts bounding boxes along with class probabilities, enables efficient and accurate detection of multiple objects simultaneously. Its multi-scale detection capability allows it to identify vehicles of varying sizes under diverse traffic conditions. The proposed system offers a scalable, automated solution for enhancing road safety and enforcing traffic regulations effectively.