Team Optimus| Spot That Fire V3.0
Spot That Fire V3.0
Firebox
Summary
Firebox is a web application that detect, predict and assess the economical impacts of wildfires and make impact to minimize losses.
How We Addressed This Challenge
We have developed a web application named Firebox.
Due to wildfires huge amount of biomass are burning, many people are dying, produce various air pollutants and often report that the ambient concentration of particulate matters (PM) increases substantially which causes higher health risks. All this problems are ultimately damaging human life and climate. So minimizing the losses are so much important.
Firebox will detects wildfires, the area containing the fire and predict expected loss in next 24 hours and notified this information to local authorities. It contains location tracker through which it will notify the users if they have the probability to affected by the wildfires. It also assess the actual losses after the wildfire.
So, the decision makers and local authorities can take their decisions in better way to solve the challenge of wildfires. It will notifying the users if there any probability of affecting them through wildfires; So they can save their lives by going to safe places. It also assessing actual losses due to it; Which can be helpful for research purposes.
How We Developed This Project
We have developed this application with Django. Django is a Python-based free and open-source web framework that follows the model-template-views architectural pattern. we made an asynchronous task (with Celery, it is an open source asynchronous task queue or job queue which is based on distributed message passing. While it supports scheduling, its focus is on operations in real time) which will update database with 24 hour MCD14DL data from FIRMS. it will segment location based on longitude,latitude and use current weather data from weather Api and use it in our fire spread prediction system and estimate fire spread for each coordinate for next 24 hour.
For modeling fire spread we used scikit learn (Scikit-learn is a free software machine learning library for the Python programming language) which is based on average temperature, humidity and wind speed of that day. when we have predicted these data, we will send a sms to local authority about fire.
If any user wants to check his/her current state they can use location picker (leaflet js) which will show information about current location based on fire location distance and air quality data from weather Api. we have used Folium for fire data visualization. For checking vegetation changing we have use selenium for automate downloading satellite photo of 3 days range and segment color with Opencv and used it's data to show vegetation change. Opencv will segment vegetation color area based on color range. Now for getting fire damage data we have used online news portal data, which is scraped with python web scraping tool and used spaCy for language processing.
How We Used Space Agency Data in This Project
Fire Information for Resource Management System (FIRMS) gives us active fire information of 24 hour/ 48 hour / 7 days . We have used fire location of 100% confidence for filtering active fire location and calculated estimated burning area. It also gives us temperature around these area and date. NASA EARTH SCIENCE DATA gives realtime satellite image we used MODIS Terra Corrected Reflectance Bands721 image to compare historic vegetation change.