Golden Iris| Automated Detection of Hazards

Automated Detection of Hazards

Countless phenomena such as floods, fires, and algae blooms routinely impact ecosystems, economies, and human safety. Your challenge is to use satellite data to create a machine learning model that detects a specific phenomenon and build an interface that not only displays the detected phenomenon, but also layers it alongside ancillary data to help researchers and decision-makers better understand its impacts and scope.

HestIA - Automated Wildfire Detection

Summary

"Houston, we have a problem."Detection of wildfires in raw satellite data is a huge task, especially because we are dealing with petabytes of data. Our approach to solving this issue is to analyze images with AI. We minimize as much as time as possible in detecting real fire outbreaks for quick responsive actions.The firefighting brigades are the most affected in combating fire outbreaks with outdated information. We use triangulated coordinates of images to deliver righteous information, such as wind speed, land moisture, and vegetation. We can eliminate false/positive fire points with multi-source combined data.

How We Addressed This Challenge

Fires are becoming more frequent and intense year by year, preventing vegetation from regenerating. Directly impacting the local fauna and flora, the economy, society and the soot derived from the great fires reach up to travel thousands of kilometres polluting and affecting other regions.


The real image detection and rapid fire outbreaks, directly minimizes the negative impacts of fires. Helping the work of the fire brigades to reach the initial fires faster.

The information on detection of fire outbreaks right at the beginning, also helps in the prevention of environmental crimes, by the Public Ministry to be able to act with actions with the competent bodies after the identification of criminal outbreaks, punishing those responsible.


 We use Artificial Intelligence technologies to analyze several images, over time (days and hours), identifying points with smoke and that migrate geographically, reporting as a real fire spot or false / positive points, which always appear static. The information will be made available to users in the form of alerts.

How We Developed This Project

Our solution

We use available satellite information to analyse possible wildfires. Through available APIs we gather images from the sky and run those images in a customized Artificial Intelligence model. We can analyse information from hundreds of hotpots, and our AI model reports exactly if we are dealing with real fire locations or false positives.




Tools

Languages: Python, Javascript, PHP and SQL

Auxiliary Libs: TensorFlow, keras, numpy, OpenCV, Flask

Dev Resources: Docker, Visual Studio, Github

Neural Network: Unet

How We Used Space Agency Data in This Project

We use images provided by the Aqua - MODIS satellites, Terra - MODIS, NOOA - VIIRS, S-NPP - VIIRS.

Project Demo
Project Code
Data & Resources

Weblinks:

https://2019.spaceappschallenge.org/challenges/living-our-world/smash-your-sdgs/teams/massa/project 

https://wiki.earthdata.nasa.gov/display/GIBS/GIBS+API+for+Developers

https://wiki.openstreetmap.org/wiki/Slippy_map_tilenames#Lon..2Flat._to_tile_numbers_2

https://xview2.org/download-links

https://earthdata.nasa.gov/earth-observation-data/near-real-time/firms

https://earthdata.nasa.gov/earth-observation-data/near-real-time/hazards-and-disasters/fires

https://images.nasa.gov/search-results?q=Fires&page=1&media=image&yearStart=1920&yearEnd=2020

https://images-assets.nasa.gov/image/PIA17384/PIA17384~small.jpg

http://sigma.cptec.inpe.br/acervohd/pdf/FAQ_001_NPP_acervoHD_v1-1.pdf 

http://www.digitalglobe.com/ecosystem/open-data

https://worldview.earthdata.nasa.gov/

https://images.nasa.gov/docs/images.nasa.gov_api_docs.pdf

https://earthdata.nasa.gov/about/esdis-project

https://www.bbc.com/portuguese/geral-54202546

https://www.youtube.com/watch?v=xoJFj6vedjg 

https://www.nytimes.com/2020/09/16/climate/wildfires-globally.html

https://ec.europa.eu/jrc/en/news/more-countries-ever-hit-forest-fires-2018

https://www.google.com/search?q=fires+NASA&rlz=1C1CHZN_pt-BRBR915BR915&sxsrf=ALeKk01o1E7LTzIFy_tnGNzHevNs4Hbunw:1601820129408&source=lnms&tbm=isch&sa=X&ved=2ahUKEwjDobr-jJvsAhVQIrkGHUKfAh0Q_AUoAnoECAwQBA&biw=1366&bih=625 

https://www.nytimes.com/2020/09/04/world/americas/brazil-wetlands-fires-pantanal.html 

https://www.careourearth.com/overview-of-major-wildfires-around-the-world-in-2019/

https://firms.modaps.eosdis.nasa.gov/map/#d:2020-10-03..2020-10-04;l:Suomi_NPP_Orbit_Dsc,NOAA-20_Orbit_Dsc,Aqua_Orbit_Dsc,Terra_Orbit_Dsc;@0.0,0.0,3z

https://www.sciencemag.org/news/2020/08/illegal-deforestation-brazil-soars-amid-climate-impunity

Tags
#Al #wildfires #fire #socialimpact #forest
Judging
This project was submitted for consideration during the Space Apps Judging process.