tech2 News StaffJul 08, 2019 12:21:57 IST
Galaxy clusters are some of the largest, yet hard-to-spot structures in the cosmos. Astronomers are now turning to artificial intelligence for help. Specifically, a deep learning technology called "Deep-CEE" (Deep Learning for Galaxy Cluster Extraction and Evaluation) that speeds up the hunt for galaxy clusters.
The study and software were presented by Matthew Chan, a doctoral researcher at Lancaster University, at the Royal Astronomical Society's National Astronomy meeting on 4 July.
Galaxy clusters are quite rare in the universe. Most galaxies reside in low-density environments called "the field", or in small groups, like our Milky Way and the neighbouring Andromeda do. While rarer, galaxy clusters do represent the most extreme environments for star systems to live in. Studying them could be a window into some of the most elusive theories about the universe — dark matter and dark energy.
A state-of-the-art artificial intelligence model has been trained as a virtual astronomer to "look" at colour images and specifically pick out galaxy clusters. It uses neural networks, which mimic the way the human brain identifies objects — using specific neurons that can visualise patterns and colours.
In a pilot study of the AI's capabilities, it managed to identify and classify galaxy clusters in images that contain many other astronomical objects. The results were encouraging, as per a statement. This new automation of the discovery process allows scientists to quickly scan sets of images and return precise predictions with minimal human interaction. This will be essential for analysing data in the future, according to a statement.
"We have successfully applied Deep-CEE to the Sloan Digital Sky Survey," says Chan, "Ultimately, we will run our model on revolutionary surveys such as the Large Synoptic Survey Telescope (LSST) that will probe wider and deeper into regions of the Universe never before explored," he added.
The LSST sky survey, due to go online in 2021, is a mega-project to image the entire patch of sky visible from the southern hemisphere. The estimated 15 TB of data from the project every night is sure to add a lot of new discoveries to the growing field of astronomy.
Efforts like Deep-CEE are adding power and skill to the discovery process itself and help ease the load on astronomers by automating some of the most challenging and mundane parts of the process.
Find our entire collection of stories, in-depth analysis, live updates, videos & more on Chandrayaan 2 Moon Mission on our dedicated #Chandrayaan2TheMoon domain.