Computationally efficient method for retrieval of atmospherically distorted astronomical images

TL;DR

This paper introduces a computationally efficient tomographic speckle imaging method using Radon transform for retrieving atmospherically distorted astronomical images, offering comparable quality to existing techniques but with reduced computational resources.

Contribution

The paper presents a novel Radon transform-based tomographic speckle masking algorithm that significantly improves computational efficiency over direct bispectrum methods.

Findings

Comparable image recovery quality to direct bispectrum method

Reduced computational time and memory requirements

Validated with simulations and real astronomical data

Abstract

Speckle Imaging based on triple correlation is a very efficient image reconstruction technique which is used to retrieve Fourier phase information of the object in presence of atmospheric turbulence. We have developed both Direct Bispectrum and Radon transform based Tomographic speckle masking algorithms to retrieve atmospherically distorted astronomical images. The latter is a much computationally efficient technique because it works with one dimensional image projections. Tomographic speckle imaging provides good image recovery like direct bispectrum but with a large improvement in computational time and memory requirements. The algorithms were compared with speckle simulations of aperture masking interferometry with 17 sub-apertures using different objects. The results of the computationally efficient tomographic technique with laboratory and real astronomical speckle images are also discussed.