Inferno Watch

Inspiration

Wildfires are becoming more frequent and devastating due to climate change. We wanted to create a system that uses real-time environmental data and AI to predict fire risks early, helping to prevent disasters before they spread.

What It Does

Inferno Watch collects temperature, humidity, and timestamps from local sensors, cross-references them with OpenWeather API data, and uses machine learning to predict wildfire risks. It then provides early warnings to help mitigate fire threats.

How We Built It

We used FastAPI for the backend, integrating OpenWeather API for real-time weather data. Machine learning models analyze sensor inputs to detect fire risk patterns. The system is deployed via DigitalOcean, ensuring scalability and reliability.

Challenges We Ran Into

• Data Accuracy: Ensuring reliable sensor readings and aligning them with external weather data.
  
• Model Training: Finding the best ML approach for fire prediction with limited real-world fire data.
  
• Integration Issues: Combining APIs, sensor data, and AI predictions into a seamless workflow.
  

Accomplishments That We're Proud Of

• Successfully integrating real-time sensor data with weather APIs.
  
• Developing a machine learning model that detects early wildfire risks.
  
• Deploying the system and making it accessible for real-world testing.
  

What We Learned

• The importance of data preprocessing and feature selection for accurate fire prediction.
  
• How to efficiently integrate sensor networks with AI-based forecasting.
  
• The challenges of deploying machine learning models in a real-time monitoring system.
  

What’s Next for Inferno Watch

• Expanding data sources: Incorporating satellite and drone imagery for enhanced predictions.
  
• Improving AI accuracy: Training models with more wildfire case studies.
  
• Community integration: Partnering with local agencies to deploy in high-risk areas.
  

Built With

• digitalocean
• esp32
• fastapi
• hugginface
• numpy
• python