An international team of scientists developed a deep learning algorithm to more deeply explore permanently shadowed regions (PSRs) of the Moon, and to image extremely small geologic features.
The researchers trained the algorithm on over 70,000 images of PSRs, coupled with data about the camera's temperature and orbital position, in order to identify and screen out camera noise.
They then fed the algorithm millions of sunlit lunar photos paired with simulated versions in shadow, to address residual noise.
The researchers used the algorithm to analyze the size and distribution of craters and boulders in several PSRs that might be explored by the U.S. National Aeronautics and Space Administration's Artemis lunar program.
From Scientific American
View Full Article
Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA