🌊 Project MARS – Maritime Activity Recognition System

Automatically detect, track, and analyze maritime vessels from satellite videos using YOLOv8 + ByteTrack.

---

🧠 About

Project MARS is an AI-powered maritime monitoring system that processes satellite video feeds to:

• Detect vessels using YOLOv8
• Track them across frames with ByteTrack
• Output an annotated video
• Generate CSV logs of detections

Built with:

• Flask backend
• Clean HTML/CSS/JS frontend
• Deep learning + computer vision

---

✨ Features

Feature	Description
📅 Video Upload	Upload any .mp4 satellite video via the web interface
🛰️ Vessel Detection	YOLOv8 detects maritime vessels per frame
🎯 ByteTrack Tracking	Assigns unique IDs to vessels over time
🎥 Annotated Output Video	Bounding boxes + track IDs overlaid in output .mp4
📃 CSV Export	Frame-by-frame detection logs as vessel_tracks.csv
⚡ Fallback Logic	Falls back to YOLOv8 COCO model if custom weights don’t detect vessels

---

📂 Project Structure

project-mars/
├── app.py              # Flask API server
├── tracker.py          # Inference & tracking logic
├── models/
│   └── best.pt         # Custom YOLOv8 model
├── jobs/               # Per-job outputs (videos, CSVs)
├── uploads/            # Uploaded videos
├── index.html          # Web UI
├── script.js           # Frontend logic
├── style.css           # CSS styling
└── requirements.txt

---

🚀 Getting Started

✅ Prerequisites

• Python 3.12+
• pip
• GPU (optional, for faster inference)

➜ Clone & Install

git clone https://github.com/your-username/project-mars.git
cd project-mars
pip install -r requirements.txt

🌐 Run Flask API

python app.py

Runs on: http://127.0.0.1:5000

🌐 Open Frontend

Just open index.html in your browser to access the control panel.

---

📄 CSV Format

frame,id,class,conf,x1,y1,x2,y2
3,1,0,0.92,100,50,140,90

---

🚧 API Endpoints

Route	Method	Description
/track	POST	Upload video + start analysis
/status/<job_id>	GET	Poll job status
/jobs/<job_id>/<file>	GET	Download CSV or video result

---

💼 Deployment

• ✅ Works locally with Python 3
• 🚧 Docker version planned
• ☁️ Ideal for deployment on AWS EC2, Azure, or Render

---

📆 Roadmap

• Add Docker support
• Real-time map visualization
• Geo-tagged detection overlay

---

📚 License

MIT License — free to use and modify with attribution.

---

Enjoy tracking ships like a satellite intelligence analyst! 🚀🚢