Transformer-Based Neural Network for Transient Detection without Image Subtraction
This work addresses the efficiency and accuracy of supernova detection in large-scale astronomical surveys, representing an incremental improvement over existing CNN methods.
The paper tackles the problem of classifying real and bogus transient detections in astronomical images by introducing a transformer-based neural network that eliminates the need for computationally-expensive difference imaging, achieving 97.4% accuracy on the autoScan dataset from the Dark Energy Survey.
We introduce a transformer-based neural network for the accurate classification of real and bogus transient detections in astronomical images. This network advances beyond the conventional convolutional neural network (CNN) methods, widely used in image processing tasks, by adopting an architecture better suited for detailed pixel-by-pixel comparison. The architecture enables efficient analysis of search and template images only, thus removing the necessity for computationally-expensive difference imaging, while maintaining high performance. Our primary evaluation was conducted using the autoScan dataset from the Dark Energy Survey (DES), where the network achieved a classification accuracy of 97.4% and diminishing performance utility for difference image as the size of the training set grew. Further experiments with DES data confirmed that the network can operate at a similar level even when the input images are not centered on the supernova candidate. These findings highlight the network's effectiveness in enhancing both accuracy and efficiency of supernova detection in large-scale astronomical surveys.