-
-
Fire detection simulator
Inspiration
Wildfires can grow rapidly, so detecting smoke as early as possible can make a major difference. We wanted to create an affordable system that uses existing cameras and drones to detect potential wildfires before they become disasters.
## What it does
Our system uses Python, OpenCV, and Pyronear to process live camera and drone footage. It analyzes frames for signs of smoke or fire, filters potential detections, and can trigger an alert when a threat is identified.
## How we built it
We built a pipeline connecting OpenCV’s video processing capabilities with Pyronear’s wildfire detection technology. Python handles the pipeline, from capturing and preprocessing frames to running detection and triggering alerts.
## Challenges we ran into
We had to deal with false positives from clouds, fog, lighting, and other objects that can look like smoke. We also had to make the pipeline efficient enough to process video continuously while keeping detection fast.
## Accomplishments that we're proud of
We successfully connected camera/video input, OpenCV processing, and Pyronear detection into a single pipeline. Our system demonstrates how existing cameras and drones could become part of a scalable early-warning network.
## What we learned
We learned how to build a real-time computer vision pipeline and integrate different technologies into one system. We also learned that reliable wildfire detection requires balancing speed, accuracy, and false positives.
## What's next for Fire Watch
We want to improve detection accuracy, add stronger temporal filtering, support more camera and drone feeds, and build a dashboard for real-time alerts and monitoring. Our long-term goal is a scalable edge-based wildfire detection network.