Multi-CCD modelling of the point spread function

Liaudat, Tobias; Bonnin, J. G.; Starck, Jean-Luc; Schmitz,; Guinot, Axel; Kilbinger, M.; Gwyn, Stephen

doi:10.1051/0004-6361/202039584

Cited by 14 publications

(29 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data-driven PSF models Classical data-driven PSF models only rely on the stars to build the model in pixel space and are blind to the physics of the inverse problem. They mostly differ in the way they handle the spatial variations and the super-resolution [3,[6][7][8][9]. They have difficulties in modeling complex PSF shapes such as those from space missions which are close to the diffraction limit.…”

Section: Related Workmentioning

confidence: 99%

“…In this paper we illustrate how some new technologies brought about by the Deep Learning (DL) revolution can be leveraged to rethink the way we model the instrument response or point spread function (PSF). To this end, we follow the approach in [4], and propose a paradigm shift in the form of a novel framework that includes a differentiable optical forward model. This allows to shift the usual data-driven modeling space from the pixels to the wavefront and translate much of the modeling complexity into the forward model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Rethinking the modeling of the instrumental response of telescopes with a differentiable optical model

Liaudat¹,

Starck²,

Kilbinger³

et al. 2021

Preprint

View full text Add to dashboard Cite

We propose a paradigm shift in the data-driven modeling of the instrumental response field of telescopes. By adding a differentiable optical forward model into the modeling framework, we change the data-driven modeling space from the pixels to the wavefront. This allows to transfer a great deal of complexity from the instrumental response into the forward model while being able to adapt to the observations, remaining data-driven. Our framework allows a way forward to building powerful models that are physically motivated, interpretable, and that do not require special calibration data. We show that for a simplified setting of a space telescope, this framework represents a real performance breakthrough compared to existing data-driven approaches with reconstruction errors decreasing 5 fold at observation resolution and more than 10 fold for a 3x super-resolution. We successfully model chromatic variations of the instrument's response only using noisy broad-band in-focus observations. * https://tobias-liaudat.github.io Fourth Workshop on Machine Learning and the Physical Sciences (NeurIPS 2021).

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Rethinking the modeling of the instrumental response of telescopes with a differentiable optical model

Liaudat¹,

Starck²,

Kilbinger³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…• MCCD [30] is a state-of-the-art data-driven method that was originally designed for the ground-based Canada-France Imaging Survey (CFIS) ¶ on the Canada-France-Hawaii Telescope (CFHT). The MCCD model extends the notions from the two methods described above and builds a single PSF model for the entire CCD mosaic in the focal plane.…”

Section: Related Workmentioning

confidence: 99%

“…We next focus on a reliable generalization capability of the PSF to target positions, and the model's adaptation to observed data. For this reason we propose to use a weighted sum of wavefront features (or eigenWFE if we draw a parallel with the notion of eigenPSF [15,16,30]). A wavefront feature is distributed across the FOV and, therefore, shared by all the PSFs at any FOV position.…”

Section: Wavefront Error Psf Modelmentioning

confidence: 99%

See 1 more Smart Citation

Rethinking data-driven point spread function modeling with a differentiable optical model

Liaudat¹,

Starck²,

Kilbinger³

et al. 2022

Preprint

View full text Add to dashboard Cite

In astronomy, upcoming space telescopes with wide-field optical instruments have a spatially varying point spread function (PSF). Certain scientific goals require a highfidelity estimation of the PSF at target positions where no direct measurement of the PSF is provided. Even though observations of the PSF are available at some positions of the field of view (FOV), they are undersampled, noisy, and integrated in wavelength in the instrument's passband. PSF modeling represents a challenging ill-posed problem, as it requires building a model from these observations that can infer a super-resolved PSF at any wavelength and any position in the FOV. Current data-driven PSF models can tackle spatial variations and super-resolution. However, they are not capable of capturing PSF chromatic variations.Our model, coined WaveDiff, proposes a paradigm shift in the data-driven modeling of the point spread function field of telescopes. By adding a differentiable optical forward model into the modeling framework, we change the data-driven modeling space from the pixels to the wavefront. This change allows the transfer of a great deal of complexity from the instrumental response into the forward model. The proposed model relies on efficient automatic differentiation technology as well as modern stochastic first-order optimization techniques recently developed by the thriving machine-learning community. Our framework paves the way to building powerful models that are physically motivated and do not require special calibration data.This paper demonstrates the WaveDiff model on a simplified setting of a space telescope. The proposed framework represents a performance breakthrough with respect to the existing state-of-the-art data-driven approach. The pixel reconstruction errors decrease 6-fold at observation resolution and 44-fold for a 3x super-resolution. The ellipticity errors are reduced by a factor of at least 20 and the size error by a factor of more than 250. By only using noisy broad-band in-focus observations, we successfully capture the PSF chromatic variations due to diffraction. WaveDiff source code and examples associated with this paper are available at this link .

show abstract

Combined strong and weak gravitational lensing mass measurements in clusters of galaxies

Abriola

2022

Proc. IAU

View full text Add to dashboard Cite

Gravitational lensing in clusters of galaxies is one of the most powerful methods to probe the dark matter mass distribution inside such systems, after mapping the baryonic component, and test the currently accepted ɅCDM cosmological paradigm. With the advent of new high-resolution facilities such as the JWST, strong lensing (SL) is capable of providing extremely accurate mass measurements in the densest regions of such structures. Weak lensing (WL) provides complementary information by measuring the total mass distribution in the outskirts of galaxy clusters, where no multiple images of background sources are produced.In my talk, I will present updates on the WL mass reconstruction of the Frontier Fields galaxy cluster Abell 2744 (z = 0.308) based on Subaru, Magellan, and JWST data, and I will show how the results obtained, combined with accurate SL modeling of this lens, provide a consistent picture of the cluster total mass distribution. I will discuss the pipeline used for this work, the extensive checks performed on the different datasets and the scientific results obtained. Being composed of several substructures undergoing a merging process, the complex geometry of this cluster makes it an ideal laboratory to verify the consistency and reliability of the results obtained with the two methods. I will compare the predictions of the SL models extrapolated in the outer regions of the cluster with the non-parametric WL mass reconstruction to look for potential systematic effects affecting the SL analysis.

show abstract

Multi-CCD modelling of the point spread function

Cited by 14 publications

References 50 publications

Rethinking the modeling of the instrumental response of telescopes with a differentiable optical model

Rethinking the modeling of the instrumental response of telescopes with a differentiable optical model

Rethinking data-driven point spread function modeling with a differentiable optical model

Combined strong and weak gravitational lensing mass measurements in clusters of galaxies

Contact Info

Product

Resources

About