Abstract The discovery of pulsars is of great significance in the field of physics and astronomy. As the astronomical equipment produces a large number of pulsar data, an algorithm for automatically identifying pulsars becomes urgent. We propose a deep learning framework for pulsar recognition. In response to the extreme imbalance between positive and negative examples and the hard negative sample issue presented in the High Time Resolution Universe Medlat Training Data, there are two coping strategies in our framework: the smart under-sampling and the improved loss function. We also apply the early-fusion strategy to integrate features obtained from different attributes before classification to improve the performance. To our best knowledge, this is the first study that integrates these strategies and techniques in pulsar recognition. The experiment results show that our framework outperforms previous works with respect to either the training time or F1 score. We can not only speed up the training time by 10× compared with the state-ofthe-art work, but also get a competitive result in terms of F1 score.
Keywords methods: data analysis — (stars:) pulsars: general — techniques: image processing
It accepts original submissions from all over the world and is internationally published and distributed by IOP