Home
- Lectures
- Institute of Astronomy and Astrophysics
- Location
R1203 of the Astronomy-Mathematics Building, National Taiwan University
- Speaker Name
Prakruth Adari, Stony Brook University
- State
Definitive
- Url
2025-07-31 14:20 - 15:20
Abstract:
Unrecognized blends are a class of objects that are mistakenly identified as a single object instead of multiple. For ground based surveys with large PSFs and seeing, unrecognized blends are an unavoidable problem that prevent precise measurements. It is expected that 20% of objects with 23 < i < 24.5 mags for the Vera Rubin Observatory will be an unrecognized blend. One of the best ways to identify unrecognized blends is through the use of space based data in deep fields. Using the COSMOS field we are able to label ground based measurements as unrecognized blends and train several machine learning algorithms to see if we are able to probabilistically label blends based on catalog values alone. I will also discuss extensions to this work and the on-going validation.
Paper: https://arxiv.org/abs/2503.16680
