Fire & Smoke Detection

Fire accidents can lead to severe damage to life, property, and the environment. Early detection of fire and smoke is essential for preventing disasters and ensuring safety. In this project, we build an AI-powered Fire & Smoke Detection System using PictoBlox and Machine Learning.

The system uses a camera to continuously monitor the surroundings and analyze live video frames. A trained AI model detects the presence of fire or smoke in real time. Whenever fire or smoke is identified, the system immediately generates an alert through messages, LEDs, or buzzers, allowing users to take quick action.

This project demonstrates the practical application of Artificial Intelligence, Machine Learning, and Computer Vision in creating smart safety solutions. It can be used in homes, schools, offices, factories, warehouses, and other public places to improve safety and reduce fire-related risks.

1. Pictoblox Software: Download PictoBlox | Windows, MacOS, Linux, Chromebook, Android & iOS
2. Laptop/PC
3. Inbuilt or External Camera Setup
4. Speaker or Headphones
5. Quarky

Connect the Quarky board to your computer using a USB cable.

Open PictoBlox on your computer.

Create a new project.

Connect Quarky in Upload or Live mode.

Ensure the webcam is connected and working properly.

Add the Machine Learning Environment extension in PictoBlox.

Save the project with the name "Fire and Smoke Detection System".

Open the project in PictoBlox.

Add the Image Classifier extension.

Turn on the webcam using the turn on video block.

Load the trained model containing:

1. Fire
2. Smoke
3. No Fire & Smoke

Display the camera feed on the stage.

Run the project and check whether the model can recognize the three classes correctly.

Test with sample fire, smoke, and normal images.

The camera starts capturing live video, and the AI model continuously predicts whether the scene contains:

1. 🔥 Fire
2. 🌫️ Smoke
3. ✅ No Fire & Smoke

Use a forever loop to continuously monitor the camera.

Run the Image Classifier on the live video feed.

Check the predicted label.

If the label is Fire, display "Fire Detected!".

If the label is Smoke, display "Smoke Detected!".

If the label is No Fire & Smoke, display "Area is Safe".

If Fire is detected:

1. Turn the Quarky LED Red.
2. Play a warning buzzer sound.
3. Display "🔥 Fire Detected! Evacuate Immediately!"

If Smoke is detected:

1. Turn the Quarky LED Yellow/Orange.
2. Play a warning buzzer sound.
3. Display "🌫️ Smoke Detected! Check the Area!"

If No Fire & Smoke is detected:

1. Turn the Quarky LED Green.
2. Stop the buzzer.
3. Display "No fire"

Homes and Apartments

Schools and Colleges

Shopping Malls

Forest Monitoring

The Fire & Smoke Detection System successfully uses Artificial Intelligence and Computer Vision to identify fire and smoke in real time. By continuously monitoring the surroundings through a camera, the system can quickly detect potential fire hazards and provide immediate alerts using LEDs, buzzers, and warning messages. This helps improve safety and enables faster emergency response.