Data-Efficient Classification of Radio Galaxies

The continuum emission from radio galaxies can be generally classified into different classes like FRI, FRII, Bent, or Compact. In this paper, we explore the task of radio galaxy classification based on morphology using deep learning methods with a focus on using a small scale dataset (~ 2000 samples).We apply few-shot learning techniques based on Siamese Networks and transfer learning techniques using a pre-trained DenseNet model with advanced techniques like cyclical learning rate, discriminative learning to train the model rapidly. We achieve a classification accuracy of over 92% using our best performing model with the biggest source of confusion being between Bent and FRII type galaxies. Our results show that focusing on a small but curated dataset along with the use of best practices to train the neural network can lead to good results. Automated classification techniques will be crucial for upcoming surveys with next generation radio telescopes which are expected to detect thousands of new radio galaxies in the future.

无线电星系的数据有效分类

射电星系的连续发射通常可以分为FRI，FRII，Bent或Compact等不同类别。在本文中，我们使用深度学习方法探索基于形态学的射电星系分类任务，重点是使用小规模数据集（约2000个样本）。.. 我们应用基于暹罗网络的快速学习技术，并使用经过预先训练的DenseNet模型和高级技术（例如循环学习率，判别式学习）来转移学习技术，以快速训练模型。我们使用性能最佳的模型实现了超过92％的分类精度，其中最大的混淆源是本特和FRII型星系之间。我们的结果表明，仅关注少量但经过精心挑选的数据集，以及使用最佳实践来训练神经网络，都可以得出良好的结果。自动分类技术对于即将进行的下一代射电望远镜的勘测至关重要，该望远镜有望在未来探测成千上万个新的射电星系。 （阅读更多）