traffic-detection

Detecting traffic using OpenCV and YOLO and tracking the vehicles for counting using Sort

Discover the future of urban mobility with the City Sense which is a UIT Data Science Traffic Application for Smart Cities. Our cutting-edge solution revolutionizes the way cities manage traffic, enhancing the quality of life for residents and fostering sustainable urban development.

A deep learning-based traffic object detection system using YOLOv8. The model detects vehicles and traffic signs such as cars, trucks, buses, traffic lights, and stop signs, providing bounding boxes and confidence scores. Trained on a filtered dataset and evaluated on real-world images.

AI-powered traffic detection and vehicle classification system using YOLO11 for Bangladesh highway surveillance. Built with Ultralytics, OpenCV, and Python.

Successfully developed an object detection model using Faster R-CNN to detect vehicles and traffic-related objects in real-time road scenes, supporting smart traffic monitoring and surveillance applications.

Traffic Car Colour & Object Detection System built with React, TypeScript and Express.

Real-time traffic violation detector using YOLOv8. Detects helmet violations, wrong-side driving and signal jumps. Sends SMS/Email alerts with auto-generated fines. UP Hackathon 2025.

This project was developed as part of the PIDEV – 4th Year Engineering Program (TWIN) at Esprit School of Engineering – Tunisia (Academic Year 2025–2026).

Interactive traffic analytics dashboard powered by YOLOv8 vehicle detection, built with Streamlit and Plotly

This project demonstrates a simple yet powerful application of the YOLOv8 (You Only Look Once) object detection model for identifying various traffic-related objects.

Deteção de veiculos, tracking de veículo e estimador de velocidade

CPU-focused edge traffic and road-object detection with PyTorch, ONNX Runtime, reproducible evaluation, robustness testing, Docker deployment, and benchmarking.

AI-based Smart Traffic Violation Detection using YOLOv8 for real-time helmet and signal violation detection

Real-time drone traffic object detection and tracking using YOLOv8m, ByteTrack, OpenCV and VisDrone dataset.

Application for the control of a system of multiple robots interacting and coordinating tasks, traffic conflicts and navigation in a mining environment

Training detection models (RetinaNet and SSD) to detect road objects, then applying a model to real world traffic video from Moscow.

This project was developed at Esprit School of Engineering – Tunisia as part of the PIDEV program (Academic Year 2025–2026). Technologies: React, Spring Boot, AI, YOLOv8.

Real-time AI traffic violation detection using CNN + OpenCV — 89% accuracy, 15 FPS, pilot selected for city deployment

REST API based on YOLOv8m for vehicle and traffic light signal detection (9 classes) in video — red-light violation monitoring. FastAPI + OpenCV. mAP50 63.2%, traffic lights >95%. Trained in Google Colab on a Roboflow dataset.