Learning a Probabilistic Strategy for Computational Imaging Sensor Selection
- Creators
- Sun, He
-
Dalca, Adrian V.
-
Bouman, Katherine L.
Abstract
Optimized sensing is important for computational imaging in low-resource environments, when images must be recovered from severely limited measurements. In this paper, we propose a physics-constrained, fully differentiable, autoencoder that learns a probabilistic sensor-sampling strategy for optimized sensor design. The proposed method learns a system's preferred sampling distribution that characterizes the correlations between different sensor selections as a binary, fully-connected Ising model. The learned probabilistic model is achieved by using a Gibbs sampling inspired network architecture, and is trained end-to-end with a reconstruction network for efficient co-design. The proposed framework is applicable to sensor selection problems in a variety of computational imaging applications. In this paper, we demonstrate the approach in the context of a very-long-baseline-interferometry (VLBI) array design task, where sensor correlations and atmospheric noise present unique challenges. We demonstrate results broadly consistent with expectation, and draw attention to particular structures preferred in the telescope array geometry that can be leveraged to plan future observations and design array expansions.
Additional Information
© 2020 IEEE. The authors would like to thank Lindy Blackburn, Alexander Raymond, Michael Johnson, and Sheperd Doeleman for helpful discussions on the constraints of a next-generation EHT array, and Michael Kellman for helpful discussions on Fourier ptychography. This work was supported by NSF award 1935980: "Next Generation Event Horizon Telescope Design," and Beyond Limits.Attached Files
Accepted Version - 2003.10424.pdf
Files
Name | Size | Download all |
---|---|---|
md5:61e51e2673885e4c05ccd36aab53f7f0
|
8.6 MB | Preview Download |
Additional details
- Eprint ID
- 103735
- DOI
- 10.1109/iccp48838.2020.9105133
- Resolver ID
- CaltechAUTHORS:20200605-134947553
- NSF
- AST-1935980
- Created
-
2020-06-05Created from EPrint's datestamp field
- Updated
-
2021-06-03Created from EPrint's last_modified field
- Caltech groups
- Astronomy Department