Low-Energy Electron-Track Imaging for a Liquid Argon Time-Projection-Chamber Telescope Concept using Probabilistic Deep Learning by M. Buuck et al. on Monday 21 November
The GammaTPC is an MeV-scale single-phase liquid argon
time-projection-chamber gamma-ray telescope concept with a novel dual-scale
pixel-based charge-readout system. It promises to enable a significant
improvement in sensitivity to MeV-scale gamma-rays over previous telescopes.
The novel pixel-based charge readout allows for imaging of the tracks of
electrons scattered by Compton interactions of incident gamma-rays. The two
primary contributors to the accuracy of a Compton telescope in reconstructing
an incident gamma-ray's original direction are its energy and position
resolution. In this work, we focus on using deep learning to optimize the
reconstruction of the initial position and direction of electrons scattered in
Compton interactions, including using probabilistic models to estimate
predictive uncertainty. We show that the deep learning models are able to
predict locations of Compton scatters of MeV-scale gamma-rays from simulated
pixel-based data to better than 0.6 mm RMS error, and are sensitive to the
initial direction of the scattered electron. We compare and contrast different
deep learning uncertainty estimation algorithms for reconstruction
applications. Additionally, we show that event-by-event estimates of the
uncertainty of the locations of the Compton scatters can be used to select
those events that were reconstructed most accurately, leading to improvement in
locating the origin of gamma-ray sources on the sky.
arXiv: http://arxiv.org/abs/http://arxiv.org/abs/2207.07805v2
All content for Astro arXiv | all categories is the property of Corentin Cadiou and is served directly from their servers
with no modification, redirects, or rehosting. The podcast is not affiliated with or endorsed by Podjoint in any way.
Low-Energy Electron-Track Imaging for a Liquid Argon Time-Projection-Chamber Telescope Concept using Probabilistic Deep Learning by M. Buuck et al. on Monday 21 November
The GammaTPC is an MeV-scale single-phase liquid argon
time-projection-chamber gamma-ray telescope concept with a novel dual-scale
pixel-based charge-readout system. It promises to enable a significant
improvement in sensitivity to MeV-scale gamma-rays over previous telescopes.
The novel pixel-based charge readout allows for imaging of the tracks of
electrons scattered by Compton interactions of incident gamma-rays. The two
primary contributors to the accuracy of a Compton telescope in reconstructing
an incident gamma-ray's original direction are its energy and position
resolution. In this work, we focus on using deep learning to optimize the
reconstruction of the initial position and direction of electrons scattered in
Compton interactions, including using probabilistic models to estimate
predictive uncertainty. We show that the deep learning models are able to
predict locations of Compton scatters of MeV-scale gamma-rays from simulated
pixel-based data to better than 0.6 mm RMS error, and are sensitive to the
initial direction of the scattered electron. We compare and contrast different
deep learning uncertainty estimation algorithms for reconstruction
applications. Additionally, we show that event-by-event estimates of the
uncertainty of the locations of the Compton scatters can be used to select
those events that were reconstructed most accurately, leading to improvement in
locating the origin of gamma-ray sources on the sky.
arXiv: http://arxiv.org/abs/http://arxiv.org/abs/2207.07805v2
Estimating effective wind speed from Gemini Planet Imager's adaptive optics data using covariance maps
Astro arXiv | all categories
1 minute
2 years ago
Estimating effective wind speed from Gemini Planet Imager's adaptive optics data using covariance maps
Estimating effective wind speed from Gemini Planet Imager's adaptive optics data using covariance maps by Daniel M. Levinstein et al. on Wednesday 30 November
The Earth's turbulent atmosphere results in speckled and blurred images of
astronomical objects when observed by ground based visible and near-infrared
telescopes. Adaptive optics (AO) systems are employed to reduce these
atmospheric effects by using wavefront sensors (WFS) and deformable mirrors.
Some AO systems are not fast enough to correct for strong, fast, high
turbulence wind layers leading to the wind butterfly effect, or wind-driven
halo, reducing contrast capabilities in coronagraphic images. Estimating the
effective wind speed of the atmosphere allows us to calculate the atmospheric
coherence time. This is not only an important parameter to understand for site
characterization but could be used to help remove the wind butterfly in post
processing. Here we present a method for estimating the atmospheric effective
wind speed from spatio-temporal covariance maps generated from pseudo open-loop
(POL) WFS data. POL WFS data is used as it aims to reconstruct the full
wavefront information when operating in closed-loop. The covariance maps show
how different atmospheric turbulent layers traverse the telescope. Our method
successfully recovered the effective wind speed from simulated WFS data
generated with the soapy python library. The simulated atmospheric turbulence
profiles consist of two turbulent layers of ranging strengths and velocities.
The method has also been applied to Gemini Planet Imager (GPI) AO WFS data.
This gives insight into how the effective wind speed can affect the wind-driven
halo seen in the AO image point spread function. In this paper, we will present
results from simulated and GPI WFS data.
arXiv: http://arxiv.org/abs/http://arxiv.org/abs/2211.16441v1
Astro arXiv | all categories
Low-Energy Electron-Track Imaging for a Liquid Argon Time-Projection-Chamber Telescope Concept using Probabilistic Deep Learning by M. Buuck et al. on Monday 21 November
The GammaTPC is an MeV-scale single-phase liquid argon
time-projection-chamber gamma-ray telescope concept with a novel dual-scale
pixel-based charge-readout system. It promises to enable a significant
improvement in sensitivity to MeV-scale gamma-rays over previous telescopes.
The novel pixel-based charge readout allows for imaging of the tracks of
electrons scattered by Compton interactions of incident gamma-rays. The two
primary contributors to the accuracy of a Compton telescope in reconstructing
an incident gamma-ray's original direction are its energy and position
resolution. In this work, we focus on using deep learning to optimize the
reconstruction of the initial position and direction of electrons scattered in
Compton interactions, including using probabilistic models to estimate
predictive uncertainty. We show that the deep learning models are able to
predict locations of Compton scatters of MeV-scale gamma-rays from simulated
pixel-based data to better than 0.6 mm RMS error, and are sensitive to the
initial direction of the scattered electron. We compare and contrast different
deep learning uncertainty estimation algorithms for reconstruction
applications. Additionally, we show that event-by-event estimates of the
uncertainty of the locations of the Compton scatters can be used to select
those events that were reconstructed most accurately, leading to improvement in
locating the origin of gamma-ray sources on the sky.
arXiv: http://arxiv.org/abs/http://arxiv.org/abs/2207.07805v2
