Real-Time Fire Detector

A real-time fire detection system using Convolutional Neural Networks (CNNs).
The system instantly detects fires from live camera feeds (default: laptop webcam), triggers audible alerts, and automatically logs fire events.

---

Key Features

• Real-Time Detection: Detects fire instantly from webcam or IP camera feeds.
• Cross-Platform Support: Works with laptop webcams (default) and optional Android phone cameras.
• Event Logging: Automatically logs fire detection events in fire_alerts.log.
• Audio Alerts: Plays a simple beep alarm when fire is detected.
• Lightweight Model: Trained CNN model optimized for fast inference.

---

Tech Stack

• — Core programming language
• — CNN model training & inference
• — High-level model building
• — Video capture & image preprocessing
• — Numerical operations
• — Plotting training results
• — Evaluation metrics

---

Project Structure

RealTime-Fire-Detector/
│
├── dataset_sample/           # Small sample images (fire / non-fire)
├── models/
│   └── forest_fire_cnn.h5    # Trained CNN model (from models.rar)
├── notebooks/
│   └── train_fire_model.ipynb
├── fire_detection.py          # Real-time detection code
├── requirements.txt           # Python dependencies
├── README.md
└── LICENSE

Installation & Requirements

Python 3.10+ is required along with the following packages:

tensorflow>=2.12.0
keras>=2.12.0
opencv-python
numpy
matplotlib
scikit-learn
jupyter
ipykernel

Install dependencies via pip:

pip install -r requirements.txt

Or with Conda:

conda create -n ds_env python=3.10
conda activate ds_env
pip install -r requirements.txt

---

How to Use

Laptop Webcam (Default)

Simply run:

python fire_detection.py

The system will automatically use your laptop’s built-in webcam (device index 0). Fire detection results will appear on screen, and a beep alert will sound if fire is detected. Press q to quit the program.

---

Android Phone Camera (Optional, via IP Webcam)

1. Install the IP Webcam app on your phone.
2. Connect your PC and phone to the same Wi-Fi network.
3. Start IP Webcam → choose MJPEG / Browser stream.
4. Update the URL in fire_detection.py:

url = "http://<your-phone-ip>:8080/video"

5. Run the detection script:

python fire_detection.py

---

Download the Trained Model

The trained CNN model for real-time fire detection can be downloaded here:

Download models.rar

Extract the .rar file to get forest_fire_cnn.h5 before running fire_detection.py.

---

Notes & Tips

• Any small flame (candle, lighter, or fire video) can be used for testing.
• No need to demonstrate an actual forest fire.
• Ensure good lighting and camera clarity for accurate detection.
• Recommended for educational, safety simulation, or prototyping purposes.

---

Contributing

Contributions are welcome! If you have improvements, optimizations, or bug fixes, feel free to submit a pull request or open an issue.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

About

This project demonstrates the power of computer vision and deep learning for real-world safety applications. It showcases practical experience with TensorFlow, Keras, and OpenCV, as well as deploying models for real-time live video analysis.