Side view of Moltke crater taken from Apollo 10. Credit: Public Domain
A team of researchers affiliated with several institutions in China, one in Italy and one in Iceland used a machine learning artificial intelligence application to count and note the location of more than 100,000 craters on the moon. In their paper published in the journal Communications about nature, the group describes the programming of their crater recognition system, training it with data collected by Chinese lunar orbiters.
Previous work to identify and map the craters on the moon tended to be a slow process – usually done manually, with researchers studying photos and transferring these observations to monthly maps or globes. In this new effort, researchers have found a way to dramatically speed up the process by teaching a computer to identify craters and then count them.
Teaching a computer to recognize craters on the moon has been a difficult process because of the many shapes that craters can take. Not all are round and have different ages, which means that the definition of characteristics has eroded over long periods of time. Scientists would like to harass all the craters on the moon and meet each of them – this could provide a unique way to study the history of the solar system.
The new approach of the team working in China involved the training of a machine learning application on the basics of craters. He was then instructed to see the craters with a broader perspective, with data from the Chinese lunar orbiters Chang’e-1 and Chang’e-2. Once the system found out what to look for, the researchers used it to analyze data from the Chang’e 5 lander, which was part of the Chinese mission that recovered rocks from the moon’s surface. The AI application used this data to identify and count craters in the middle and low latitude regions of the moon. The new system counted 109,956 craters – far more than ever before. He also kept track of the location of each of the craters he found and placed each in a predefined geological time period depending on how much he eroded the crater.
Study suggests much more water on the moon than previously thought (Update)
Chen Yang et al. Identification of the monthly impact crater and estimation of age with Chang’E data through deep learning and transfer, Communications about nature (2020). DOI: 10.1038 / s41467-020-20215-y
Provided by Science X
© 2020 Science X Network
Citation: Using AI to count and map craters on the moon (2020, December 23) retrieved on December 24, 2020 from https://phys.org/news/2020-12-ai-craters-moon.html
This document is subject to copyright. Apart from any fair transaction for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.