fbpx AI for Good Challenge: Detecting wildfire outbreaks to avoid deforestation

Detecting forest fire outbreaks to avoid disastrous deforestation

If you are an (aspiring) data scientist, AI engineer, or an enthusiast just starting your AI journey, apply for the following two-month challenge to boost your skills on real data while helping to find a solution for one of the grandest challenges on the planet. The challenge starts on October 8th.

The Problem

Deforestation and agriculture damages contribute 17% to climate change. Wild Fires are responsible for 8 billion tons of CO2 emissions, which is one-third of all global CO2 emissions!

The challenge partner Sintecsys, a mission-driven Brazilian startup, built an automatic-early-stage-exact-location-fire-outbreak-detection-solution, which already reduced losses by 90%!

The challenge goal

You will help to improve the current solution by building machine learning models that detect fires in real-time (or near real-time) using landscape images generated by cameras on top of communication towers. The focus will lie on,

1. identifying smoke in the footage or images during daytime as being a fire outbreak.

2. identifying lightness in nighttime images as being a fire outbreak.

The datasets

• Thousands of high-resolution images for every condition (fire, non-fire, day, night, sunrise, nightfall, etc.)
• Data for transition conditions (from day to night) will require labeling effort

Apply and be one of 40 collaborators to make a real-world impact.

Impact
Our Partners

Sintecsys is a mission-driven company based in Sao Paulo and has designed a solution to detect fire outbreaks in forests and plantations in an average time of 5 minutes.

We welcome (aspiring) data scientists, AI engineers, and enthusiasts who just started their AI journey but are eager to solve a real-world problem.
Requirements:
  • Good English
  • A good/very good grasp in computer science and/or mathematics
  • Student, (aspiring) data scientist, AI engineer, data engineer
  • Programming experience with C/C++, C#, Java, Python, Javascript or similar
  • Understanding of ML and Deep learning algorithms.
  • Availability of around 15-20 hours a week.
Your benefits:
  • Working with world-class mentors and domain experts
  • Making international friends in a fast-growing supportive community of collaborators
  • Boosting your technical skills, problem-solving capabilities, and international collaboration
  • Building your personal brand and publishing your own articles on our website and blog
  • Certificates of participation and references

Community Collaborator (Forest Fire)