9–11 Dec 2026
Indian Institute of Science Education and Research (IISER), Tirupati
Asia/Kolkata timezone

ASTRONOMICAL IMAGE RECONSTRUCTION: DEPLOY­ING A PHYSICS-INFORMED DEEP LEARNING AP­PROACH TO PREDICT MISSING SPECTRAL BANDS

Not scheduled
20m
Indian Institute of Science Education and Research (IISER), Tirupati

Indian Institute of Science Education and Research (IISER), Tirupati

Srinivasapuram Venkatagiri Road, Jangalapalli G, P), Yerpedu, Mandal, Panguru, Andhra Pradesh 517619
Poster Presentation Astronomy & Astrophysics

Speaker

Kshitij Chandrakar (UPES Dehradun)

Description

The study of the physical nature of galaxies involves their images in different visible bands. However, with SOTA telescopes deployed to image these galaxies in multiple bands face several restrictions on available hardware deployed leading to some bands missing from these images. This makes the physical nature of the galaxies difficult to study. The proposed model will learn the physical rules between these different observed wavelengths and help us make an educated guess about the bands that it has not been observed in yet.This will help us create a unified library of images in common bands that can help us understand the unobserved underlying physics of the galaxies. This will enable cosmologists to apply a common model on all galaxy images in order to study their properties and have a centralised dataset of these galaxies in all the possible combinations of observable bands.

Nature of work Computational

Authors

Ms Anchal Negi (UPES Dehradun) Mr Chirag Garg (UPES Dehradun) Kshitij Chandrakar (UPES Dehradun) Dr Pranshu Mandal (UPES Dehradun) Mr Vedant Kedia (UPES Dehradun)

Presentation materials

There are no materials yet.