Using algorithms that can strip away atmospheric interference, the researchers have made it possible for Earth-bound telescopes to produce some of the deepest, clearest images of distant stars, galaxies, and other cosmic elements needed to study the universe's origins and structure.
"By sharpening our view of the sky, we can see farther, fainter targets and push the threshold of what's detectable," said Tamas Budavari, an astronomer and mathematician at Johns Hopkins University who led the research. "That will give us prettier pictures of the night sky, but not just for academic amusement. It will open new opportunities to improve cosmology research and revolutionize how we process and understand astronomical observations."
Even the most powerful ground-based telescopes struggle to observe the sky because Earth's atmosphere is in constant flux. Variations in temperature, pressure, and other air conditions cause subtle but important distortions in how light refracts as it passes through the atmosphere-especially for faint sources of light.
Traditional techniques to process these atmospheric distortions have historically struggled to produce high-quality images because they either blur fine details or introduce grainy artifacts, Budavari said. The new solution, called ImageMM, improves telescope images by modeling how light from celestial objects travels and how changing conditions across different layers of the atmosphere influence those light waves.
"Think of the atmosphere as a restless sheer curtain, constantly shifting and shimmering, so the scene behind it always looks blurred," said Yashil Sukurdeep, a Johns Hopkins mathematician who developed the algorithm. "Our algorithms learn to see past that curtain, reconstructing the still, sharp image hidden behind it. Using advanced mathematical techniques, we are able to produce the clearest possible view-revealing the night sky in stunning clarity. We dubbed our algorithm ImageMM, because at its core, it relies on the Majorization-Minimization (MM) method-an elegant mathematical technique that we've adapted in a new way for exploring the cosmos."
Early tests of the new method restored blurry and noisy images from the Subaru Telescope, one of the world's largest, in a matter of seconds. The reprocessed images revealed details such as the intricate structure of spiral galaxies in unprecedented clarity. They were specifically acquired by the Subaru Telescope at Mauna Kea summit in Hawaii to test for exposures of similar quality as future captures by the Vera C. Rubin Observatory, a state-of-the-art facility in Chile that will begin operating this year.
"Astronomers already have very sophisticated tools to analyze imaging data from telescopes, but they don't remove all the noise, they don't remove all the blur, and they don't deal very well with missing pixel values," Sukurdeep said. "Our framework can recover a near-perfect image from a series of imperfect observations. We'll never have ground truth, but we think this is as close as it currently gets to perfect."
While the algorithms can process new and old data from various observatories, they are being tailored for the Rubin Observatory's upcoming sky survey to acquire vast amounts of data about dark energy and dark matter. These two mysterious components of the cosmos are believed to be responsible for accelerating the universe's expansion and holding galaxies together.
"It's critical for astronomers to accurately measure the shapes of objects, not just to analyze the morphology of individual galaxies but to statistically analyze their distortions that come from dark matter and other gravitational effects," Budavari said. "When it comes to billion-dollar ground-based observatories, gaining even just a small degree of depth and quality improvement from those observations can be huge."
Although space telescopes have a superior ability to capture extremely deep, high-resolution images, they can only manage to capture a tiny portion of the observable sky, Budavari said. Over its 34-year lifetime, the Hubble Space Telescope has only photographed about 0.1% of the sky, according to NASA. Instead, ground-based facilities such as the Rubin Observatory will image the entire visible sky every few days.
"With our technique, that's hundreds of observations that we can turn into images almost comparable to what we could only obtain with a space telescope before," Budavari said. "Of course that's idealistic, but that's really the goal here: removing the atmosphere."
Research Report:ImageMM: Joint Multi-frame Image Restoration and Super-resolution
Related Links
Johns Hopkins University
Stellar Chemistry, The Universe And All Within It
| Subscribe Free To Our Daily Newsletters |
| Subscribe Free To Our Daily Newsletters |
