ai-cctv-system-2.0

AI-powered CCTV system with face recognition, behavior detection, and Telegram alerts using

Comment

🛡️ Sentinel AI — Smart Security System

An AI-powered real-time security system using YOLOv8 for person detection, Telegram alerts, live video streaming, and automatic video recording.

✨ Features

Feature Description
🤖 AI Detection YOLOv8 detects people in real-time
🏃 Behavior Analysis Classifies movement as NORMAL / SUSPICIOUS / RUNNING
📱 Telegram Alerts Sends instant alerts with screenshot to your phone
🎥 Auto Recording Records 6-second video clip on every alert
🌐 Live Stream Browser-accessible MJPEG stream at http://localhost:5000/stream
🔔 Sound Alert Plays beep when intruder is detected
👤 Face Recognition Optional — identify known vs unknown faces
🔒 Privacy Safe All credentials stored in .env file (never in code)

🚀 Quick Start

1. Clone the repo

git clone https://github.com/YOUR_USERNAME/sentinel-ai.git
cd sentinel-ai

2. Install dependencies

pip install -r requirements.txt

3. Download YOLOv8 model

# Auto-downloads on first run, or manually:
wget https://github.com/ultralytics/assets/releases/download/v0.0.0/yolov8n.pt

4. Set up your credentials

cp .env.example .env

Edit .env and fill in your values:

TELEGRAM_TOKEN=your_bot_token_here
TELEGRAM_CHAT_ID=your_chat_id_here
CAMERA_INDEX=0
MODEL_PATH=yolov8n.pt

5. Run

python main.py

📱 Telegram Setup

  1. Open Telegram → search @BotFather
  2. Send /newbot → follow steps → copy the token
  3. Search @userinfobot → send any message → copy your chat ID
  4. Paste both into your .env file

📁 Project Structure

sentinel-ai/
├── main.py              # Main application
├── requirements.txt     # Python dependencies
├── .env.example         # Environment variable template
├── .env                 # Your credentials (NOT committed)
├── .gitignore           # Git ignore rules
├── yolov8n.pt           # AI model (NOT committed, download separately)
├── known_faces/         # Add photos of known people here
├── unknown_faces/       # Unknown face captures saved here
└── evidence/            # Alert screenshots & videos saved here

🌐 API Endpoints

Endpoint Method Description
/status GET Returns current alert status as JSON
/stream GET Live MJPEG video stream

👤 Optional: Face Recognition

To enable face recognition:

  1. Install extra dependencies:

    pip install cmake dlib face_recognition

  2. Add photos of known people to known_faces/:

    • Name each file as the person's name: john.jpg, alice.png

  3. Uncomment the face recognition blocks in main.py

⚙️ Configuration

All settings are in .env:

Variable Default Description
TELEGRAM_TOKEN — Your Telegram bot token
TELEGRAM_CHAT_ID — Your Telegram chat ID
CAMERA_INDEX 0 Webcam index (0 = default)
MODEL_PATH yolov8n.pt Path to YOLO model

📋 Requirements

  • Python 3.8+
  • Webcam
  • Windows / Linux / macOS

⚠️ Privacy Notice

  • Evidence footage is stored locally only in evidence/
  • Telegram alerts contain screenshots — only sent to your own Telegram chat
  • Never commit .env — it contains your private credentials
  • Known faces data is excluded from git by .gitignore

📄 License

MIT License — free to use and modify.

🙏 Credits

Built With

Share this project:

Updates