Classification of Planetary Nebulae through Deep Transfer Learning

Dayang N F Awang Iskandar,Gary A Fuller,Iain Mcdonald,Johari Abdullah,Ahmad H Fauzi,Rosni Abdullah,Albert A Zijlstra

doi:10.3390/galaxies8040088

Dayang N F Awang Iskandar, Gary A Fuller + Show 5 more

Open Access

https://doi.org/10.3390/galaxies8040088

Copy DOI

Abstract

This study investigate the effectiveness of using Deep Learning (DL) for the classification of planetary nebulae (PNe). It focusses on distinguishing PNe from other types of objects, as well as their morphological classification. We adopted the deep transfer learning approach using three ImageNet pre-trained algorithms. This study was conducted using images from the Hong Kong/Australian Astronomical Observatory/Strasbourg Observatory H-alpha Planetary Nebula research platform database (HASH DB) and the Panoramic Survey Telescope and Rapid Response System (Pan-STARRS). We found that the algorithm has high success in distinguishing True PNe from other types of objects even without any parameter tuning. The Matthews correlation coefficient is 0.9. Our analysis shows that DenseNet201 is the most effective DL algorithm. For the morphological classification, we found for three classes, Bipolar, Elliptical and Round, half of objects are correctly classified. Further improvement may require more data and/or training. We discuss the trade-offs and potential avenues for future work and conclude that deep transfer learning can be utilized to classify wide-field astronomical images.

Highlights

A planetary nebulae (PN) forms when a sun-like star ejects its envelope at the end of its life.The ejected envelope forms an expanding nebula around the remnant core of the star which ionizes it.After some 104 years, the PN fades from view, both because of the expansion and dilution of the nebula and because of the fading of the ionizing star
We investigate the efficacy of Deep Learning (DL) for deciding whether an object is a PN and for determining its morphological classification
Frozen parameters are parameter values that are not changed or updated. As this is an initial study on how DL can be used to classify planetary nebulae (PNe), we focus the evaluation on the effectiveness of different DL models

Summary

Introduction

A planetary nebulae (PN) forms when a sun-like star ejects its envelope at the end of its life.The ejected envelope forms an expanding nebula around the remnant core of the star which ionizes it.After some 104 years, the PN fades from view, both because of the expansion and dilution of the nebula and because of the fading of the ionizing star. A planetary nebulae (PN) forms when a sun-like star ejects its envelope at the end of its life. The ejected envelope forms an expanding nebula around the remnant core of the star which ionizes it. PNe show up as compact nebulosity on images of the sky, with typical spectra that are dominated by emission lines. They are commonly identified by comparing images taken at different wavelengths. They can be confused with other types of astronomical objects: confirmation that a nebula is a PN requires follow-up spectroscopy. An overview of PNe discovery surveys can be found in Parker [1]

Methods

Results

Conclusion