Super-resolving Herschel - a deep learning based deconvolution and denoising technique

TL;DR

This paper introduces a deep learning super-resolution method using a transformer neural network to enhance Herschel far-infrared images, significantly improving resolution, speed, and source extraction accuracy for dusty star-forming galaxies.

Contribution

The work presents a novel transformer-based deep learning approach for super-resolution of Herschel images, outperforming traditional methods in speed and accuracy without extensive fine-tuning.

Findings

Achieves 4.5x resolution enhancement of Herschel images.

Flux accuracy within 5% for sources >8 mJy.

Inference time of 1s per square degree on consumer GPUs.

Abstract

Dusty star-forming galaxies (DSFGs) dominate the far-infrared and sub-millimetre number counts, but single-dish surveys suffer from poor angular resolution, complicating mult-wavelength counterpart identification. Prior-driven deblending techniques require extensive fine-tuning and struggle to process large fields. This work aims to develop a fast, reliable deep-learning based deconvolution and denoising super-resolution (SR) technique. We employ a transformer neural network to improve the resolution of Herschel/SPIRE 500 $\mu$m observations by a factor 4.5, using Spitzer/MIPS 24$\mu$m and Herschel/SPIRE 250, 350, 500$\mu$m images. Trained on SIDES and SHARK simulations, we injected instrumental noise into the input simulated images, while keeping the target images noise-free to enhance de-noising capabilities of our method. We evaluated the performance on simulated test sets and real JCMT/SCUBA-2 450 $\mu$m observations in the COSMOS field which have superior resolution compared to Herschel. Our SR method achieves an inference time of $1s/deg^2$ on consumer GPUs, much faster than traditional deblending techniques. Using the simulation test sets, we show that fluxes of the extracted sources from the super-resolved image are accurate to within 5% for sources with an intrinsic flux $\gtrsim$ 8 mJy, which is a substantial improvement compared to blind extraction on the native images. Astrometric error is low ($\lesssim$ 1" vs 12" pixel scale). Reliability is $\gtrsim$ 90% for sources $>$3 mJy and $>$90% of sources with intrinsic fluxes $\gtrsim5$ mJy are recovered. Applied to real 500 $\mu$m observations, fluxes of the extracted sources from the super-resolved map agree well with SCUBA-2 measured fluxes for sources $\geq$10 mJy. Our technique enables SR over hundreds of $deg^2$ without the need for fine-tuning, facilitating statistical analysis of DSFGs.