Single-pixel imaging pattern sets and their implications on scene reconstruction

Kallepalli, Akhil and Innes, John and Gibson, Graham M. and Padgett, Miles J.; Padgett, Miles J. and Bongs, Kai and Fedrizzi, Alessandro and Politi, Alberto, eds. (2021) Single-pixel imaging pattern sets and their implications on scene reconstruction. In: Quantum Technology. Proceedings of SPIE . SPIE, GBR. ISBN 9781510646070 (https://doi.org/10.1117/12.2601228)

[thumbnail of Kallepalli-etal-SPIE-2021-Single-pixel-imaging-pattern-sets]
Preview
 Text. Filename: Kallepalli-etal-SPIE-2021-Single-pixel-imaging-pattern-sets.pdf
Final Published Version
License: Strathprints license 1.0
Download (25MB)| Preview

Abstract

Alternatives to focal-plane detector arrays have emerged in recent years and are now being widely investigated. One of the most promising of these are single-pixel imaging techniques. Single-pixel techniques recreate a scene using the knowledge of projected patterns and the measured backscattered signals. This research investigates the implications of using idealised patterns, Fourier-transformed patterns and camera-captured patterns generated by plane wave decomposition methods. Hadamard patterns are projected into the far-field of the phased-array modulator source and used for robust reconstruction in a reflective arrangement. The choice of the optimal pattern sets from these sources are used for single-pixel imaging reconstruction and compressed sensing. The technique is robust to poor signal-to-noise conditions and is applicable in cases where a limited number of measurements are possible. Our technique and methodology can be further applied to any region of the electromagnetic spectrum where phased-array sources are available, such as in the radar regime.

CORE (COnnecting REpositories)