New AI tool can generate faster, accurate and sharper cosmic images

“Slight differences in shape can tell us about gravity in the universe. These differences are already difficult to detect. If you look at an image from a ground-based telescope, a shape might be warped. It’s hard to know if that’s because of a gravitational effect or the atmosphere,” said Emma Alexander, the study’s senior author, in a statement.

The team was able to produce blur-free, high-resolution images of the universe by incorporating this AI algorithm. This pre-existing computer algorithm is commonly used to sharpen photographs.

The AI tool also enables faster and more accurate image processing by utilizing computational power to sift through image data. As a result, better measurements of cosmic observations are provided. 

“Photography’s goal is often to get a pretty, nice-looking image. But astronomical images are used for science. By cleaning up images in the right way, we can get more accurate data. The algorithm removes the atmosphere computationally, enabling physicists to obtain better scientific measurements. At the end of the day, the images do look better as well,” added Alexander. 

The tool produced images with approximately 38 percent less error than traditional blur-removal methods and approximately 7 percent less error than current modern methods. 

They also trained it to match imaging parameters for the Vera C. Rubin Observatory in north-central Chile, which is set to open soon.

This AI tool, along with coding and tutorial guidelines, is available online.

They have published the results in the journal Monthly Notices of the Royal Astronomical Society. 

Study abstract:

Removing optical and atmospheric blur from galaxy images significantly improves galaxy shape measurements for weak gravitational lensing and galaxy evolution studies. This ill-posed linear inverse problem is usually solved with deconvolution algorithms enhanced by regularisation priors or deep learning. We introduce a so-called ’physics-informed deep learning’ approach to the Point Spread Function (PSF) deconvolution problem in galaxy surveys. We apply algorithm unrolling and the Plug-and-Play technique to the Alternating Direction Method of Multipliers (ADMM), in which a neural network learns appropriate hyperparameters and denoising priors from simulated galaxy images. We characterize the time-performance trade-off of several methods for galaxies of differing brightness levels, as well as our method’s robustness to systematic PSF errors and network ablations. We show an improvement in reduced shear ellipticity error of 38.6 per cent (SNR=20)/45.0 per cent (SNR=200) compared to classic methods and 7.4 per cent (SNR=20)/33.2 per cent (SNR = 200) compared to modern methods.