🔥💨 Fire Smoke Detection & Telegram Alert System

📖 Overview

This project is a real-time fire and smoke detection system that leverages YOLOv11 for object detection and OpenCV for video stream processing. The system continuously analyzes video streams (from a webcam or saved footage), detects fire and smoke, draws bounding boxes with labels, and sends annotated snapshots with alerts to a Telegram chat. It is designed to provide early warnings for fire hazards, ensuring quick response and safety.

---

Demo_Fire01.mp4

---

Demo_Smoke01.mp4

---

🚀 Features

• Real-Time Detection: Detects fire and smoke in video streams with high accuracy using YOLOv11.
• Motion Detection: Filters frames based on motion detection to reduce false positives.
• Alert System: Sends real-time alerts with annotated images to a Telegram chat.
• Configurable Sensitivity: Adjust detection confidence threshold and notification frequency.
• Visualization: Displays the detection results with bounding boxes and labels on the video feed.

---

🛠 Technologies & Libraries

• Model & Inference:
  • YOLOv11 (Ultralytics)
  • PyTorch, OpenCV
• Notification:
  • TelegramBot API
• Data & Annotation:
  • Roboflow for dataset labeling
• Environment:
  • Python 3.8+
  • Google Colab (training)

---

⚙️ Installation & Setup

1️⃣ Training the Model (Google Colab)

• Open the Colab Notebook & Run the Fire_detection_Model.ipynb file

• Download the best.pt file to your local machine for deployment.

2️⃣ Running the Detection Script Locally

• Setup Local Environment: Ensure you have Python 3.8+ installed.

• Install required libraries:

pip install python-dotenv opencv-python

• Prepare Environment Variables

  • Create a .env file in the project root directory with the following content:

TELEGRAM_TOKEN=your_bot_token
TELEGRAM_CHAT_ID=your_chat_id

• Run the Detection Script

python FireSmokeDetection.py

• Make sure the VIDEO_PATH in the script points to the correct video file.

3️⃣ System Workflow

• Video Input: Reads video input from a file or webcam.

• Detection: YOLOv11 detects fire and smoke in the frames, draws bounding boxes, and adds labels.

• Alert System: Sends annotated frames to Telegram if fire or smoke is detected.

• User Control: Send /stop to the bot to terminate detection.

4️⃣ Notes

• GPU in Colab: Ensure you enable GPU acceleration (Runtime > Change runtime type > GPU).

---

👤 Author

Developed by THUC TU

---

Contact

For any questions or feedback, please contact:

• Email: tuthucdz@gmail.com
• GitHub: Ne4nf