Thank you for liking
You have already liked this page, you can only like it once!
Successfully mitigating the damage caused by marine litter requires large-scale, accurate mapping of ocean borne plastics across the globe. Quantifying where and how much plastic debris enters the marine environment is critical in formulating solutions. Satellite-based measurements can help survey remote locations where in-situ observations are sparse, costly, and difficult.
Plastics commonly found as ocean litter have spectral features that can be detected with multispectral assets such as the Copernicus Sentinel-2 satellite, but the sensor’s reduced ability to distinguish between plastic and other materials makes it difficult to create reliable detection maps. On the other hand, hyperspectral sensors, such as the upcoming CHIME satellite, have better spectral resolution that allows for more successful plastic detection. Unfortunately there are not currently many of these hyperspectral sensors available.
The objective of this project was to develop and assess the feasibility of an enhanced multispectral plastic detection algorithm, leveraging the latest in machine learning approaches. MDA developed a novel neural network called Joint Plastic Spectral Embedding (JPSEmbed) that learns to associate multispectral data with the more capable hyperspectral data in a way that allows for better detection of plastic than using multispectral input alone. Multispectral sensors in orbit can then better contribute to creating plastic detection and concentration maps along with hyperspectral ones as they become available.
Find out more here.
The project was funded through the Discovery element of ESA’s Basic Activities. It was selected through an Open Space Innovation Platform (OSIP) call for ideas on remote sensing of plastic marine litter.
Access the other Final presentations of ESA Discovery element activities videos.