University of Leeds researchers have introduced a neural network that can quickly and accurately chart large Antarctic icebergs in satellite images in a mere 0.01 seconds, making it 10,000 times faster than human efforts. Icebergs captured by the Sentinel-1 satellite appear as bright objects against the darker ocean and sea-ice background.
This neural network excels in mapping icebergs even in challenging conditions like when smaller iceberg fragments make them easily grouped together with the main iceberg by mistake. The neural networks’ ability to understand intricate non-linear relationships and take the whole image context into account make this a non-issue. Plus, it was trained using Sentinel-1 images exhibiting giant icebergs in different settings, with manually-derived outlines serving as the target.
Sale
NATIONAL GEOGRAPHIC Dual LED Kids Microscope – 50+ pc Science Kit with 10 Prepared Slides & 10 DIY Blank…
- TWO MICROSCOPES IN ONE! Use the lower LED lights to view biological specimens on slides or switch over to the upper lights to examine 3D objects in…
- OVER 50 ACCESSORIES INCLUDED – Explore a curated set of 10 prepared slides with a range of biological specimens, then create your own with the…
- POWERFUL LEARNING TOOL – This microscope features two sets of optical glass lenses providing 20x and 50x magnification, and is easy to operate making…
Satellites are, of course, essential for monitoring changes and understanding processes occurring far from civilization. This new neural network automates what would otherwise be a manual and labor-intensive task of locating and reporting iceberg extent,” said Mark Drinkwater, ESA’s GOCE Mission Scientist.
[Source]
