collision-warning

Python script for vehicle detection and tracking using YOLOv8 with dynamic road extent visualization. The model adapts to various camera angles, accurately determining road height by detecting road, sky, and mid lines. Includes TTC calculations, cut-in warnings, and visual overlays.

This repository offers code to repeat the experiments and reuse the method in the paper "Minimising missed and false alarms: a vehicle spacing based approach to conflict detection".

Forward Collision Warning (FCW) prototype using OpenCV and YOLOv8 for lane detection, vehicle detection, lane-based vehicle filtering, and collision warning generation.

EV Collision Warning System using HB100 Radar, ToF Sensing, ESP32, Raspberry Pi 5, FFT Speed Estimation and Random Forest Risk Classification.

C++ in-engine vehicle perception ADAS for OpenGL racing simulator — distance, TTC, risk HUD

Vision-based ADAS prototype using YOLO for pothole, crack, manhole, vehicle and person detection, with explainable forward-collision risk warnings and a browser demo.

AI-powered Advanced Driver Assistance System (ADAS) using YOLOv8 and OpenCV for Forward Collision Warning, Blind Spot Monitoring, and Pothole Detection with real-time video analytics.

Real-time ADAS using MiDaS depth estimation and YOLO object detection for collision alerts, lane departure warnings, and intuitive visual/audio feedback.