AI Model Morpheus Prepares to Analyze James Webb Telescope Data
Scientists are deploying a machine learning model called Morpheus to process the massive influx of images from the James Webb Space Telescope. The AI tool will classify hundreds of thousands of galaxies to help map the earliest structures in the universe.
The James Webb Space Telescope releases its first highly anticipated images on 12 July, marking a new era for space observation as the largest and most powerful observatory ever built. To handle the massive influx of cosmic data, scientists rely on artificial intelligence to quickly analyze and classify what the telescope sees. Machine learning models provide a hands-off approach to processing this information based on specific, predefined metrics.
A professor of astronomy and astrophysics at UC Santa Cruz, Brant Robertson, leads the development of Morpheus, a deep learning model designed to detect and classify galaxies at the pixel level. Originally utilized to categorize images from the Hubble Space Telescope, Morpheus now undergoes upgrades to tackle the advanced data from Webb. Tech company Nvidia supplies powerful GPUs to accelerate the AI model across multiple computing platforms.
Robertson and a team of nearly 50 researchers apply Morpheus to the COSMOS-Webb project, an ambitious program to map the earliest structures in the universe. Over 200 hours of observation time, the telescope surveys a massive patch of sky to image half a million galaxies in near-infrared and 32,000 galaxies in mid-infrared. This survey stands as the largest contiguous area survey planned for the telescope in the foreseeable future.