AI detects and classifies galaxies in astronomical photographs

US, WASHINGTON (ORDO NEWS) — Researchers at the University of California at Santa Cruz have developed a powerful new computer program called Morpheus, which can analyze data on astronomical images pixel by pixel to identify and classify all galaxies and stars in large amounts of data from astronomical research.

Morpheus is a deep learning environment that incorporates various artificial intelligence technologies designed for applications such as image and speech recognition. Brant Robertson, a professor of astronomy and astrophysics who leads a research team in computational astrophysics at the University of California at Santa Cruz, said the rapidly growing size of astronomical data made it necessary to automate some of the tasks traditionally performed by astronomers.

“There are some things that we simply won’t be able to do as people, so we must find ways to use computers to process the huge amount of data that will come over the next few years from major astronomical research projects,” he said.

Robertson worked with Ryan Hausen, a graduate student in computer science at the Baskin School of Engineering at UCSC, who has developed and tested Morpheus for the past two years. After publishing their results on May 12 in a series of additions to the Astrophysical Journal, Hausen and Robertson also publicly publish the Morpheus code and provide online demos.

The morphology of galaxies, from rotating disk galaxies like our Milky Way, to amorphous elliptical and spheroidal galaxies, can tell astronomers how galaxies form and develop over time. Large-scale research will generate huge amounts of image data, and Robertson was involved in planning how to use this data to understand the formation and evolution of galaxies.

During space exploration at the Vera Rubin Observatory, which is currently under construction in Chile, more than 800 panoramic images will be taken every night using a 3.2 billion pixel camera, recording all the visible sky twice a week.

“Imagine that you turned to astronomers and asked them to classify billions of objects – how could they do this? Now we will be able to automatically classify these objects and use this information to study the evolution of galaxies, ”Robertson said.


Contact us: [email protected]

Our Standards, Terms of Use: Standard Terms And Conditions.