Technology Category
Search and Optimization: Learning to SearchComputer Vision: Object Detection & CategorizationMachine Learning: Deep Neural Architectures and Foundation Models
Robustness analysis of Deep Sky Objects detection models on HPC
🤖 AI Summary
Detecting faint deep-sky objects (galaxies, nebulae, star clusters) in astronomical images remains challenging due to low signal-to-noise ratios and complex, heterogeneous backgrounds.
Method: This paper proposes an automated detection and evaluation framework specifically tailored to astronomical image characteristics. It systematically benchmarks YOLO and RET-DETR on large-scale deep-sky imagery using high-performance computing (HPC), enabling massively parallel model training and evaluation. Key adaptations include astronomy-aware preprocessing, customized loss functions, and domain-specific evaluation metrics.
Results: Experiments demonstrate that the framework achieves significantly improved detection accuracy and robustness—maintaining high recall and precise localization even under severe background clutter and extremely low SNR. The results validate the effectiveness, robustness, and scalability of deep learning models for complex astronomical imaging. Moreover, the framework provides a reproducible, generalizable technical pathway for intelligent processing of large-scale sky-survey data.
Technology Category
Application Category
📝 Abstract
Astronomical surveys and the growing involvement of amateur astronomers are producing more sky images than ever before, and this calls for automated processing methods that are accurate and robust. Detecting Deep Sky Objects -- such as galaxies, nebulae, and star clusters -- remains challenging because of their faint signals and complex backgrounds. Advances in Computer Vision and Deep Learning now make it possible to improve and automate this process. In this paper, we present the training and comparison of different detection models (YOLO, RET-DETR) on smart telescope images, using High-Performance Computing (HPC) to parallelise computations, in particular for robustness testing.
Problem
Research questions and friction points this paper is trying to address.
Automating detection of faint Deep Sky Objects
Improving robustness in astronomical image processing
Comparing models for object detection using HPC
Innovation
Methods, ideas, or system contributions that make the work stand out.
Uses YOLO and RET-DETR detection models
Leverages High-Performance Computing (HPC)
Parallelizes computations for robustness testing
🔎 Similar Papers
No similar papers found.
