Although astrophotographs have been improved for accuracy in the past, advances in artificial intelligence are literally transforming the field in an amazing way.

When light reaches our Earth from distant stars, planets and galaxies, it first passes through the atmosphere and is already undergoing significant changes. Not only does the atmosphere block certain wavelengths of light, but it also distorts the light that reaches Earth. Even the clearest night sky has moving air that affects the light. This is also why the largest ground-based telescopes are installed at high places, so that the atmosphere that interferes with the recordings is as thin as possible.

A colleague from Northwestern University explained it all analogously: “It’s a bit like looking up from the bottom of a swimming pool. Water pushes and distorts light. Atmosphere is of course much less dense, but it’s still a similar concept.”

See also  The mystery of the Eye of the Desert may have been solved

Blurring becomes a problem when astrophysicists analyze the images to extract cosmic data. By studying the apparent shape of galaxies, scientists can detect the gravitational effects of large cosmic structures that bend light on its way to our planet. This can make an elliptical galaxy, for example, appear rounder than it actually is. However, blurring in the atmosphere distorts the image in a way that distorts the shape of the galaxy. Removing the blur allows scientists to gather more accurate data, explains A Northwest Engineering.

Even small differences in shape can reveal the gravity of the universe, but these small differences are hard to notice. When viewing an image from a ground-based telescope, the figure may be distorted. It is difficult to say whether this is due to the effect of gravity or due to the atmosphere.

For this reason, researchers from Northwestern University and Peking Tsinghua University decided to call artificial intelligence.

©




They combined an optimization algorithm with a deep learning network trained on astronomical images. The resulting device produced images with 38.6 percent fewer errors compared to traditional error removal methods and 7.4 percent fewer errors compared to modern methods. The image below shows how to remove noise from an image during processing.

The Vera C. Rubin Observatory telescopes, which opens next year, will begin a decades-long deep survey of a vast swath of the night sky. Since the researchers also trained the new instrument on data designed to simulate the observatory’s upcoming images, it will be able to analyze long-awaited astronomical data more accurately.

For astronomers interested in the instrument, the code is open source, easy to use, and accompanying tutorials Available online.

If you want to know similar things other times, like it HVG Tech department’s Facebook page.




HVG


The number of editorial offices independent of the authorities is constantly decreasing, and the ones that still exist are trying to stay afloat in a headwind that is getting stronger every day. At HVG, we persevere, do not give in to pressure, and bring local and international news every day.

That’s why we ask you, our readers, to support us! We promise to keep doing the best we can!