YOLO-CIANNA: Galaxy detection with deep learning in radio data. I. A new YOLO-inspired source detection method applied to the SKAO SDC1

TL;DR

YOLO-CIANNA is a deep learning-based galaxy detection method tailored for large radio datasets, significantly improving detection accuracy and speed over previous approaches, and demonstrating state-of-the-art results on simulated SKA data.

Contribution

The paper introduces YOLO-CIANNA, a customized deep learning detector for astronomical images, achieving superior detection performance and efficiency on SKA simulation data.

Findings

Improved challenge-winning score by +139% on SDC1 dataset.

Detected 40-60% more sources than previous top methods.

Achieved 94% detection purity with strong characterization accuracy.

Abstract

The upcoming Square Kilometer Array (SKA) will set a new standard regarding data volume generated by an astronomical instrument, which is likely to challenge widely adopted data-analysis tools that scale inadequately with the data size. The aim of this study is to develop a new source detection and characterization method for massive radio astronomical datasets based on modern deep-learning object detection techniques. For this, we seek to identify the specific strengths and weaknesses of this type of approach when applied to astronomical data. We introduce YOLO-CIANNA, a highly customized deep-learning object detector designed specifically for astronomical datasets. In this paper, we present the method and describe all the elements introduced to address the specific challenges of radio astronomical images. We then demonstrate the capabilities of this method by applying it to simulated 2D continuum images from the SKA observatory Science Data Challenge 1 (SDC1) dataset. Using the SDC1 metric, we improve the challenge-winning score by +139% and the score of the only other post-challenge participation by +61%. Our catalog has a detection purity of 94% while detecting 40 to 60% more sources than previous top-score results, and exhibits strong characterization accuracy. The trained model can also be forced to reach 99% purity in post-process and still detect 10 to 30% more sources than the other top-score methods. It is also computationally efficient, with a peak prediction speed of 500 images of 512x512 pixels per second on a single GPU. YOLO-CIANNA achieves state-of-the-art detection and characterization results on the simulated SDC1 dataset and is expected to transfer well to observational data from SKA precursors.