FireVision – Real-Time Fire Detection using Computer Vision

Project Overview

FireVision is a real-time fire detection system that uses computer vision techniques to monitor live video streams and detect fire-like regions. The system processes frames from a webcam using OpenCV and identifies fire based on color detection in the HSV color space. When potential fire pixels exceed a predefined threshold, an alarm is triggered to alert users.

This system can be used for early fire detection in surveillance environments such as warehouses, offices, or public areas.

Features

• Real-time fire detection using webcam feed
• OpenCV-based video processing
• Fire detection using HSV color thresholding
• Automatic alarm alert when fire is detected
• Lightweight and fast detection pipeline

System Workflow

1. Capture live video feed using OpenCV.
2. Process each frame from the video stream.
3. Convert frame to HSV color space.
4. Detect fire-like colors using HSV thresholds.
5. Count fire-colored pixels in the frame.
6. Trigger alarm if pixel count exceeds threshold.

Technologies Used

• Python
• OpenCV
• NumPy
• Pygame (for alarm sound)
• Visual Studio Code

Installation

1. Clone the repository

git clone https://github.com/shreya100504/firevision.git

2. Navigate to the project folder

cd firevision

3. Install required libraries

pip install opencv-python numpy pygame

Running the Project

Run the Python script:

python fire-detector.py

The system will activate the webcam and start monitoring for fire-like regions. If fire is detected, an alarm sound will be triggered.

Applications

• Fire monitoring in industrial environments
• Warehouse safety systems
• Surveillance camera monitoring
• Early fire warning systems