First Full-Event Reconstruction from Imaging Atmospheric Cherenkov Telescope Real Data with Deep Learning

Jacquemont, Mikaël; Vuillaume, Thomas; Benoit, Alexandre; Maurin, Gilles; Lambert, Patrick; Lamanna, Giovanni

doi:10.48550/arxiv.2105.14927

Cited by 5 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, this task has been attracting attention of experiments in gamma-astronomy for quite a while now [9][10][11][12][13][14][15][16][17][18]. In this case, the main instruments are Cherenkov telescopes (Imaging Atmosphere Cherenkov Telescopes, IACTs), which register tracks produced by cascades initiated by gamma rays or by hadrons.…”

Section: Energy Estimationmentioning

confidence: 99%

Machine Learning Techniques for the EUSO-SPB2 Fluorescence Telescope

Filippatos,

Zotov

2023

Proceedings of 38th International Cosmic Ray Conference — PoS(ICRC2023)

View full text Add to dashboard Cite

The Extreme Universe Space Observatory on a Super Pressure Balloon 2 (EUSO-SPB2) is the most advanced balloon mission undertaken by the JEM-EUSO collaboration. EUSO-SPB2 is built on the experience of previous stratosphere missions, EUSO-Balloon and EUSO-SPB, and of the Mini-EUSO space mission currently active onboard the International Space Station. EUSO-SPB2 is equipped with two instruments: a fluorescence telescope aimed at registering ultra-high energy cosmic rays (UHECRs) with an energy above 2 EeV and a Cherenkov telescope built to measure direct Cherenkov emission from cosmic rays with energies above 1 PeV. The EUSO-SPB2 mission will provide pioneering observations on the path towards a space-based multi-messenger observatory. As such, a special attention was paid to the development of triggers and other software aimed at comprehensive data analysis. A whole number of methods based on machine learning (ML) and neural networks was developed during the construction of the experiment and a few others are under active development. Here we provide a brief review of the ML-based methods already implemented in the instrument and the ground software and report preliminary results on the ML-based reconstruction of UHECR parameters for the fluorescence telescope.

show abstract

Section: Energy Estimationmentioning

confidence: 99%

Machine Learning Techniques for the EUSO-SPB2 Fluorescence Telescope

Filippatos,

Zotov

2023

Proceedings of 38th International Cosmic Ray Conference — PoS(ICRC2023)

View full text Add to dashboard Cite

show abstract

“…The CNN we developed, 𝛾-PhysNet DA, is presented in depth in Ref. [16,17]. It is a multi-task network, so contrary to the classical approach and to previous works using CNNs, a single algorithm is trained to perform the full event reconstruction (particle classification, energy reconstruction and direction reconstruction).…”

Section: Ii-b 𝛾-Physnet -A Convolutional Neural Network To Analyse Ls...mentioning

confidence: 99%

Analysis of the Cherenkov Telescope Array first Large-Sized Telescope real data using convolutional neural networks

Vuillaume¹,

Jacquemont²,

Lavergne³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The Cherenkov Telescope Array (CTA) is the future ground-based gamma-ray observatory and will be composed of two arrays of imaging atmospheric Cherenkov telescopes (IACTs) located in the Northern and Southern hemispheres respectively. The first CTA prototype telescope built on-site, the Large-Sized Telescope (LST-1), is under commissioning in La Palma and has already taken data on numerous known sources. IACTs detect the faint flash of Cherenkov light indirectly produced after a very energetic gamma-ray photon has interacted with the atmosphere and generated an atmospheric shower. Reconstruction of the characteristics of the primary photons is usually done using a parameterization up to the third order of the light distribution of the images. In order to go beyond this classical method, new approaches are being developed using state-of-the-art methods based on convolutional neural networks (CNN) to reconstruct the properties of each event (incoming direction, energy and particle type) directly from the telescope images. While promising, these methods are notoriously difficult to apply to real data due to differences (such as different levels of night sky background) between Monte Carlo (MC) data used to train the network and real data. The GammaLearn project, based on these CNN approaches, has already shown an increase in sensitivity on MC simulations for LST-1 as well as a lower energy threshold. This work applies the GammaLearn network to real data acquired by LST-1 and compares the results to the classical approach that uses random forests trained on extracted image parameters. The improvements on the background rejection, event direction, and energy reconstruction are discussed in this contribution.

show abstract

“…The use of ML algorithms, which are very effective for image or time series analysis, would better exploit all image features contrary to the standard Hillas parametrization. For this reason, two analysis pipelines [9,13] based on convolutional neural networks are being tested for the image reconstruction. In addition other methods are been investigated, like look-up tables or semianalytical formula [4,10,14,16].…”

Section: Simulation and Analysis For The Advanced Cameramentioning

confidence: 99%

Development of an advanced SiPM camera for the Large Size Telescope of the Cherenkov Telescope Array

Heller,

Armstrong,

Bellato

et al. 2021

Preprint

View full text Add to dashboard Cite

First Full-Event Reconstruction from Imaging Atmospheric Cherenkov Telescope Real Data with Deep Learning

Cited by 5 publications

References 16 publications

Machine Learning Techniques for the EUSO-SPB2 Fluorescence Telescope

Machine Learning Techniques for the EUSO-SPB2 Fluorescence Telescope

Analysis of the Cherenkov Telescope Array first Large-Sized Telescope real data using convolutional neural networks

Development of an advanced SiPM camera for the Large Size Telescope of the Cherenkov Telescope Array

Contact Info

Product

Resources

About