Bayesian and Convolutional Networks for Hierarchical Morphological Classification of Galaxies

TL;DR

This paper introduces BCNN, a hierarchical galaxy classification method combining CNNs and Bayesian networks, which improves accuracy by respecting the galaxy class hierarchy using probabilistic inference.

Contribution

The novel BCNN approach integrates CNNs with Bayesian networks to enhance hierarchical galaxy classification accuracy.

Findings

BCNN outperforms several CNNs in accuracy and hierarchical F-measure.

Achieved 67% exact match and 78% accuracy on Hubble galaxy images.

Effective hierarchical constraint enforcement improves classification consistency.

Abstract

This work is focused on the morphological classification of galaxies following the Hubble sequence in which the different classes are arranged in a hierarchy. The proposed method, BCNN, is composed of two main modules. First, a convolutional neural network (CNN) is trained with images of the different classes of galaxies (image augmentation is carried out to balance some classes); the CNN outputs the probability for each class of the hierarchy, and its outputs/predictions feed the second module. The second module consists of a Bayesian network that represents the hierarchy and helps to improve the prediction accuracy by combining the predictions of the first phase while maintaining the hierarchical constraint (in a hierarchy, an instance associated with a node must be associated to all its ancestors), through probabilistic inference over the Bayesian network so that a consistent prediction is obtained. Different images from the Hubble telescope have been collected and labeled by experts, which are used to perform the experiments. The results show that BCNN performed better than several CNNs in multiple evaluation measures, reaching the next scores: 67% in exact match, 78% in accuracy, and 83% in hierarchical F-measure.