Radio Galaxy Morphology Generation Using DNN Autoencoder and Gaussian Mixture Models
This work addresses the need for simulated radio galaxy images to study supermassive black hole evolution, but it is incremental as it applies existing methods to a specific domain.
The authors tackled the problem of generating realistic radio galaxy morphologies for FRI and FRII types by proposing a framework using a deep neural network autoencoder and Gaussian mixture models, achieving high efficiency and performance in simulations.
The morphology of a radio galaxy is highly affected by its central active galactic nuclei (AGN), which is studied to reveal the evolution of the super massive black hole (SMBH). In this work, we propose a morphology generation framework for two typical radio galaxies namely Fanaroff-Riley type-I (FRI) and type-II (FRII) with deep neural network based autoencoder (DNNAE) and Gaussian mixture models (GMMs). The encoder and decoder subnets in the DNNAE are symmetric aside a fully-connected layer namely code layer hosting the extracted feature vectors. By randomly generating the feature vectors later with a three-component Gaussian Mixture models, new FRI or FRII radio galaxy morphologies are simulated. Experiments were demonstrated on real radio galaxy images, where we discussed the length of feature vectors, selection of lost functions, and made comparisons on batch normalization and dropout techniques for training the network. The results suggest a high efficiency and performance of our morphology generation framework. Code is available at: https://github.com/myinxd/dnnae-gmm.