🚗 AI-Powered Advanced Driver Assistance System (ADAS)

An intelligent Computer Vision based ADAS platform developed using YOLOv8 and OpenCV to improve road safety through real-time vehicle monitoring and hazard detection.

---

📌 Overview

This project implements multiple Advanced Driver Assistance System (ADAS) features using Computer Vision and Deep Learning techniques.

The system analyzes video streams in real time and provides safety alerts for:

• Forward Collision Warning (FCW)
• Blind Spot Detection
• Pothole Detection
• Vehicle Monitoring
• Real-Time Hazard Awareness

The objective is to assist drivers in making safer driving decisions and reducing accident risks.

---

🎯 Problem Statement

Road accidents are often caused by:

• Driver distraction
• Blind spot visibility issues
• Poor road conditions
• Delayed reaction time

This system uses AI-powered object detection and distance estimation to identify potential hazards before they become critical.

---

🏗 System Architecture

Input Video ↓ YOLOv8 Object Detection ↓ Vehicle Detection & Tracking ↓ Distance Estimation ↓ FCW Module ↓ Blind Spot Detection ↓ Pothole Detection ↓ Driver Alerts & Visualization

---

⚙ Features

🚨 Forward Collision Warning (FCW)

• Detects vehicles ahead
• Calculates Time-To-Collision (TTC)
• Generates collision warnings

🔍 Blind Spot Detection

• Monitors left and right blind zones
• Detects hidden vehicles
• Provides visual alerts

🛣 Pothole Detection

• Custom YOLOv8 pothole model
• Detects road surface defects
• Helps prevent vehicle damage

📊 Real-Time Analytics

• Live object tracking
• Vehicle counting
• Detection visualization

---

🧠 Technologies Used

Programming

• Python

Computer Vision

• OpenCV

Deep Learning

• YOLOv8
• Ultralytics

Data Processing

• NumPy
• Pandas

Visualization

• Matplotlib

---

📈 Results

Vehicle Detection

• Real-time vehicle recognition
• Multi-object tracking

Collision Prediction

• TTC based warning generation

Blind Spot Monitoring

• Continuous zone surveillance

Road Safety

• Pothole identification and alerts

---

📂 Project Structure

ADAS-ComputerVision/ │ ├── models/ ├── datasets/ ├── outputs/ ├── screenshots/ ├── videos/ ├── app.py ├── requirements.txt └── README.md

---

🚀 Installation

git clone https://github.com/mohesh05/ADAS_ComputerVision.git

cd ADAS_ComputerVision

pip install -r requirements.txt

python app.py

---

🎓 Applications

• Smart Vehicles
• Driver Assistance Systems
• Autonomous Driving Research
• Automotive Safety Solutions
• Transportation Analytics

---

👨‍💻 Author

Mohesh V K

AI & ML Engineer | Computer Vision Developer

Christ University

Hosur, Tamil Nadu, India

---

⭐ If you found this project useful, consider starring the repository.