ShapeNet: Shape Constraint for Galaxy Image Deconvolution

Abstract

Deep Learning (DL) has shown remarkable results in solving inverse problems in various domains. In particular, the Tikhonet approach is very powerful to deconvolve optical astronomical images (Sureau et al. 2020). Yet, this approach only uses the 2 loss, which does not guarantee the preservation of physical information (e.g. flux and shape) of the object reconstructed in the image. In Nammour et al. ( 2021), a new loss function was proposed in the framework of sparse deconvolution, which better preserves the shape of galaxies and reduces the pixel error. In this paper, we extend Tikhonet to take into account this shape constraint, and apply our new DL method, called ShapeNet, to optical and radio-interferometry simulated data set. The originality of the paper relies on i) the shape constraint we use in the neural network framework, ii) the application of deep learning to radio-interferometry image deconvolution for the first time, and iii) the generation of a simulated radio data set that we make available for the community. A range of examples illustrates the results.

Deblurring galaxy images with Tikhonov regularization on magnitude domain

“…The PSF deconvolution is a technique that restores images degraded by a PSF and significantly improves the spatial resolution. There has been much research investigating deconvolution methods (Starck et al 2002;Anconelli et al 2007;Rydbeck 2008;Chung et al 2021;Shibuya et al 2022), including blind deconvolution (Shi et al 2017;Fétick et al 2020;Hope et al 2022) and machinelearning based methods (Schawinski et al 2017;Sureau et al 2020;Gan et al 2021;Nammour et al 2022). The most famous and classical method is the Richardson-Lucy algorithm (RL; Richardson 1972, Lucy 1974, which assumes the pixel values to follow a Poisson distribution and would provide a maximum likelihood solution.…”