< Back

Marine Litter Detection via Satellites

Closing date is

14 Nov 2023

Create an AI computer vision system to identify litter in marine locations from satellite data.

Marine Litter Detection via Satellites

Client:

Organiser:

Mi4People

Project Status

We just kicked-off this project with a handful of volunteers. We expect that we have enough people to create an MVP ML-model. However, we are already looking for people who would like to help develop this project beyond the MVP.

What's needed

When we will go beyond MVP, we expect that we will need volunteers who are able to:
• Create a data pipeline for satellite images (publicly available data from Sentinel-2 satellites)
• Create a labeling pipeline to label more training data. Thereby we plan to use the MVP ML-model to help identify interesting regions on satellite images
• Build a web app to visualize results of the model prediction
• Setup and manage cloud infrastructure to perform real-time calculations/update (it will be a lot of calculations 😊)
If you have experience in one of these domains and have motivation to help fight pollution of marine environment, you are welcome! 😊
Depending on your task, it would be great if you could commit roughly 5-10 hours a week.

Why it's important

Marine litter is a major ecological threat to marine ecosystems. It can trap and kill marine life, smother its habitat, act as a hazard to navigation, and ultimately lead to loss of biodiversity. Because of plastic decomposition into the microplastic, tiny toxic particles can also be ingested by aquatic life causing the toxic particles move up the food chain, and finally threaten the health of humans.
Therefore, cleaning the seas and preventing further pollution of the oceans is extremely important for our planet, our health, and even our economy. To enable this, it is essential to know where marine litter is located.
However, currently location of marine litter is mainly found via local surveys and on-ground/sea observations. This approach is expensive, and time-consuming, not scalable, and leads to relatively sparse data resulting in only rough estimations of overall amount of marine litter.

What you'll learn

• Improve your skills in remote sensing/satellite image processing
• Improve your skills in computer vision
• Learn about AI-supported labeling
• Improve your skill at or learn how to run large complex calculations in the cloud (ideally in a cost-saving manner)
• Learn to think out of the box applying high-tech to help solve fundamental ecological problems

How to apply?

Marine Litter Detection via Satellites