A comparative study of convolutional neural networks for the detection of strong gravitational lensing

As we enter the era of large-scale imaging surveys with the upcoming telescopes such as the Large Synoptic Survey Telescope (LSST) and the Square Kilometre Array (SKA), it is envisaged that the number of known strong gravitational lensing systems will increase dramatically. However, these events are still very rare and require the efficient processing of millions of images. In order to tackle this image processing problem, we present machine learning techniques and apply them to the gravitational lens finding challenge. The convolutional neural networks (CNNs) presented here have been reimplemented within a new, modular, and extendable framework, Lens EXtrActor CaTania University of Malta (LEXACTUM). We report an area under the curve (AUC) of 0.9343 and 0.9870, and an execution time of 0.0061 and 0.0594 s per image, for the Space and Ground data sets, respectively, showing that the results obtained by CNNs are very competitive with conventional methods (such as visual inspection and arc finders) for detecting gravitational lenses.

Automated summary

With the era of large-scale imaging surveys approaching, the number of known gravitational lensing systems is expected to increase dramatically. Machine learning techniques, specifically convolutional neural networks, have shown competitive results in detecting gravitational lenses.