Deep Transfer Learning for Blended Source Identification in Galaxy Survey Data

TL;DR

This paper introduces BlendHunter, a deep transfer learning approach using VGG-16 for automated identification of blended sources in galaxy survey images, outperforming traditional methods especially for close blends.

Contribution

It demonstrates the effectiveness of transfer learning with VGG-16 for blend detection in galaxy images, showing improved accuracy over existing tools like SEP.

Findings

BlendHunter outperforms SEP by ~15% in accuracy for close blends.

Transfer learning enables effective blend identification across different noise levels.

The method is robust for blends with source separation less than 10 pixels.

Abstract

We present BlendHunter, a proof-of-concept for a deep transfer learning based approach for the automated and robust identification of blended sources in galaxy survey data. We take the VGG-16 network with pre-trained convolutional layers and train the fully connected layers on parametric models of COSMOS images. We test the efficacy of the transfer learning by taking the weights learned on the parametric models and using them to identify blends in more realistic CFIS-like images. We compare the performance of this method to SEP (a Python implementation of SExtractor) as function of noise level and the separation between sources. We find that BlendHunter outperforms SEP by $\sim 15\%$ in terms of classification accuracy for close blends ($<10$ pixel separation between sources) regardless of the noise level used for training. Additionally, the method provides consistent results to SEP for distant blends ($\geq10$ pixel separation between sources) provided the network is trained on data with a relatively close noise standard deviation to the target images. The code and data have been made publicly available to ensure the reproducibility of the results.