bonheurne07/trafic_light

Traffic Light Project: Implementing real-time traffic density calculation and object detection using YOLOv3-tiny. Seeking to enhance car detection accuracy with a curated dataset from Zenodo.

Traffic Object Detection with YOLO

This project implements real-time object detection using the YOLO (You Only Look Once) algorithm. It detects various objects in traffic scenes captured by CCTV cameras at traffic junctions and calculates traffic density based on the detected objects.

Features

• Real-time object detection using YOLOv3-tiny model
• Detects various objects such as cars, trucks, buses, etc.
• Draws bounding boxes around detected objects
• Counts the number of objects detected
• Saves annotated images with bounding boxes
• Generates a JSON file with object names and counts
• Supports customization and further development

Installation

1. Clone the repository:

git clone https://github.com/bonheurNE07/trafic_light.git