Turning everyday cameras into intelligent safety systems.

DETECTION PROOF

TELEGRAM ALERT PROOF

SentinelCV is a real-time computer vision platform that transforms any CCTV, IP camera, webcam, or video stream into an AI-powered safety monitor.

It starts with its first module — real-time fall detection — and is built to grow into a full modular vision-intelligence platform covering industrial safety, surveillance, healthcare, and smart infrastructure.

---

✨ Features

• 🚶 Real-time human fall detection
• 📹 Works with CCTV, RTSP streams, USB webcams, and recorded video files
• ⚡ YOLOv8-powered detection pipeline
• 🔔 Instant Telegram alerts the moment a fall is detected
• 🖥️ Lightweight, real-time inference
• 🔧 Built to extend — fall detection is module one, not the whole product

---

🎯 Use Cases

Elderly Care	Hospitals	Smart Homes
Warehouses	Factories	Construction Sites
Assisted Living	Public Surveillance

---

🏗️ How It Works

Camera Feed (CCTV / RTSP / Webcam / Video File)
        │
        ▼
  Frame Capture
        │
        ▼
  YOLOv8 Detection
        │
        ▼
  Fall Analysis
        │
        ▼
  Alert Engine
        │
        ├── Telegram Notification
        ├── Dashboard (Planned)
        └── Logging

---

🚀 Quick Start

git clone https://github.com/sreerevanth/SentinelCV.git
cd SentinelCV

Install the core dependencies (OpenCV, Ultralytics, NumPy, python-telegram-bot):

pip install opencv-python ultralytics numpy python-telegram-bot

Run the monitor:

python complete_monitor.py

📌 Set up your Telegram bot token and chat ID first — see telegram bot details in this repo for the setup steps used by notification_alert_test.py.

---

📂 Project Structure

SentinelCV/
├── complete_monitor.py         # Main fall detection + monitoring pipeline
├── video_test.py                # Run detection against a video file
├── test.py                      # Detection pipeline testing
├── download_dataset.py          # Fetch/prepare training data
├── notification_alert_test.py   # Telegram alert integration test
├── telegram bot details         # Telegram bot setup notes
└── .gitignore

(A requirements.txt, trained model weights, and a proper src/ layout are on the way — see Roadmap.)

---

🧠 Tech Stack

• Python
• OpenCV — video capture & frame processing
• Ultralytics YOLOv8 — object/pose detection
• NumPy
• Telegram Bot API — real-time alerting

---

🛣️ Roadmap

• Real-time fall detection
• requirements.txt + packaged CLI entry point
• Fire & smoke detection
• PPE (personal protective equipment) detection
• Violence detection
• Weapon detection
• Intrusion detection
• Crowd analytics
• Web dashboard
• Multi-camera support
• Edge deployment (Jetson / Raspberry Pi)

---

🤝 Contributing

Contributions are welcome! Please open an issue before submitting major changes so we can align on direction first.

---

📄 License

MIT License

---

SentinelCV — the first module of a bigger vision intelligence platform.