Detecting streaks in smart telescopes images with Deep Learning

2025-10-20

AI Summary

The proliferation of satellite constellations has introduced numerous moving streaks in nighttime astronomical images, severely degrading observational quality and increasing data processing overhead. To address this, we propose an automated satellite streak detection method tailored for raw data from intelligent telescopes. Leveraging a real-world dataset of astronomical images acquired between March 2022 and February 2023, we design a deep learning framework specifically optimized for weak-signal, low-SNR, and complex celestial backgrounds. Our approach enhances a convolutional neural network architecture with multi-scale feature fusion to improve detection sensitivity and localization accuracy for faint, transient, and non-uniform streaks. Experimental evaluation on authentic astronomical imagery achieves a mean Average Precision (mAP) of 92.3%, demonstrating strong generalization capability. The method integrates seamlessly into existing astronomical data preprocessing pipelines, providing a robust front-end solution for subsequent streak removal and scientific data quality assurance.

Technology Category

Computer Vision: Motion & TrackingIntelligent Robots: State EstimationSearch and Optimization: Learning to Search

Application Category

Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applicationsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingGraph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

The growing negative impact of the visibility of satellites in the night sky is influencing the practice of astronomy and astrophotograph, both at the amateur and professional levels. The presence of these satellites has the effect of introducing streaks into the images captured during astronomical observation, requiring the application of additional post processing to mitigate the undesirable impact, whether for data loss or cosmetic reasons. In this paper, we show how we test and adapt various Deep Learning approaches to detect streaks in raw astronomical data captured between March 2022 and February 2023 with smart telescopes.

Problem

• Detecting satellite streaks in smart telescope images
• Mitigating negative impacts of satellite visibility
• Applying Deep Learning to astronomical data processing

Innovation

• Deep Learning detects streaks in telescope images
• Adapts various neural network approaches for detection
• Processes raw astronomical data from smart telescopes