Detecting streaks in smart telescopes images with Deep Learning
Olivier Parisot, Mahmoud Jaziri · 2025
Read Detecting streaks in smart telescopes images with Deep Learning by Olivier Parisot, Mahmoud Jaziri (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-17540.txt.
Introduction
Satellites crossing the night sky leave bright streaks across telescope images, spoiling both casual astrophotography and professional data. Olivier Parisot and Mahmoud Jaziri study how deep learning can spot those streaks automatically in raw images from smart telescopes, the compact automated scopes now popular with amateurs. Finding the streaks is the first step before any cleanup or repair of the picture can begin.
Parisot and Jaziri test and adapt several deep learning approaches on images captured between March 2022 and February 2023 with such telescopes. Their work checks which designs cope with the faint, thin, and varied trails that satellites draw across real observing data. The results guide anyone building streak cleanup tools for the growing flood of smart telescope pictures.
Catalog record
| CREATOR | Olivier Parisot, Mahmoud Jaziri |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-17540
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Detecting streaks in smart telescopes images with Deep Learning Olivier Parisot, Mahmoud Jaziri 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. 1 Introduction Observational astronomy and astrophotography are challenging, whether professionally or as a hobby, and one of them is the following the proliferation of mega-constellation satellites such as SpaceX’s Starlink, OneWeb and Amazon’s Project Kuiper in Low-Earth Orbit (LEO) [ 11 , 19 ] . While these satellite networks have started to revolutionize the global internet communications, they also raise growing concerns, for multiple aspects (environment, defence, culture, etc.) [ 39 ] . In particular, the impact of these mega-constellations on astronomy and astrophotography has become a key issue [ 40 , 38 ] , calling into question the possibility of observing the night sky without disturbance. This problem is satellite light pollution, which occurs when orbiting satellites reflect the sun’s light unto the Earth [ 12 ] . It can make astronomical observations and astrophotography considerably more difficult [ 20 ] , and diminishes the quality of images captured by both amateur and professional instruments: • Loss of quality: mega-constellation satellites can create light trails as they pass in view of a telescope or camera lens during long exposure photography (Figure 1 ). These streaks can compromise image quality by leaving unwanted lines of light across astronomical images from an aesthetic point of view, but also in terms of data loss. • Increased sky brightness: the sun’s reflection of satellite surfaces can contribute to a general increase in the brightness of the night sky. This makes it more difficult to observe and capture faint, distant celestial objects, such as galaxies, nebulae and faint stars. • Reduced contrast: the presence of moving satellites can reduce the contrast between celestial objects and the sky background. Subtle details in astronomical pictures, which depend on a dark, uniformly illuminated night sky, can be compromised by bright streaks and scattered spots. • Complications for image calibration: satellite light steaks can disrupt the process by introducing non-stellar elements into the images, making treatment difficult or even impossible. These problems require advanced post-processing to solve, such as technical adjustments and specialized algorithms like detection, denoising and inpainting to mitigate the undesirable effects caused by satellites. Figure 1: Observation of NGC6482 by the authors with a smart telescope: a streak has completely damaged the image during acquisition . Streaks are also for professional ground-based observatories [ 10 ] , making it imperative to set up a process to estimate the concrete impact on the quality of large digital sky surveys [ 22 ] , avoid disturbances and correct data if possible [ 36 ] . The effect is significant: as recent studies have shown [ 20 , 2 ] , the increased traffic in LEO will inevitably lead to a loss of astronomical data and therefore reduce the possibility of discoveries, as weak astrophysical signals are increasingly lost in the noise. The International Astronomical Union (IAU) recently published a ’Call to Protect the Dark and Quiet Sky from Harmful Interference by Satellite Constellations’ 1 1 1 https://cps.iau.org/documents/49/techdoc102.pdf . Furthermore, it’s a hot topic for space-based observatories like Hubble [ 17 ] , which adds a number of constraints that are difficult to resolve, especially given the cost of operating such facilities in space. Satellites have had and will continue to have a significant influence on this particular phenomenon of light pollution. However, it is also important to mention that other fast-moving Near-Earth Objects (NEO), such as meteors and large space debris, or even cosmic rays, have the potential to leave linear features on astronomical images [ 24 ] . In this article, we propose the study of a dataset containing raw astronomical images captured over a year with a smart telescope, in conditions accessible to amateurs, and we evaluate the quantity of images effectively impacted by streaks. The rest of the paper is structured as follows. In Section 2 , we present a brief review of the state of the art concerning the detection of streaks in astronomical images. Section 3 describes the dataset of images with smart telescopes between March 2022 and February 2023, while Section 4 proposes a study of this dataset using different methods. We discuss the results in Section 5 , before concluding and proposing some perspectives in Section 6 . 2 Related works For many years now, the scientific community has proposed many techniques to detect and track satellites, and to deal with the trails they cause in astronomical images [ 24 , 3 , 14 ] . Specialized software for astronomical images processing (like SharpCap 2 2 2 https://www.sharpcap.co.uk ) provide basic features to manage this problem. In the Python software ecosystem, we can mention these tools: • ASTRiDE aims to detect streaks in astronomical images [ 16 ] with boundary-tracing and morphological parameters. ASTRiDE can detect not only long streaks, but also relatively faint, short or curved ones. As we will see later in this article, this is also a problem because it tends to confuse real streaks with tracking problems – so it requires a fine configuration like in [ 8 ] . • Authors of [ 7 ] proposed a Python pipeline for lightweight streak detection, identification and initial orbit determination from FITS raw files captured by amateur-grade telescopes – but it was tested on only a few images captured around the celestial equator (85). FITS (Flexible Image Transport System) is a file format most commonly used into store astronomical data. • A research team applied the probabilistic Hough transform through a Python scripts using well-known open source libraries like openCV and scikit-images, and by using GPU-specific computation to detect streaks in FITS files captured by the Tomo-e Gozen camera at Kiso Observatory in Japan [ 5 ] . Unfortunately, the source code is not available. • pyradon is a Python package based on Fast Radon Transform (FRT) to find streaks in 2D astronomical images [ 24 ] . Some existing approaches are based on Deep Learning. As an example, an approach based on YOLO ( You Only Look Once , version 2) was proposed by [ 37 ] to detect streaks in images captured by a multi-camera system with a wide field of view. An other YOLO model (version 5) was proposed by [ 3 ] to detect streaks in synthetic raw images. Furthermore, a recent work compared two techniques based on Deep Convolutional Neural Networks: an extended feature pyramid network (EFPN) with faster region-based CNNs (Faster R-CNN) and a feature pyramid network (FPN) with Faster R-CNN [ 9 ] . These approaches are elaborated but they were only tested on synthetic data. The authors of [