Rapid single-photon color imaging of moving objects.

Area of Science:

• Photonics and Computational Imaging
• Development of advanced imaging techniques for low-light conditions.

Background:

• High-speed color imaging of moving objects in low-photon environments is challenging.
• Efficient spectral separation and fast, computationally inexpensive reconstruction algorithms are needed.

Purpose of the Study:

• To demonstrate a Bayesian image reconstruction approach for rapid single-photon color imaging.
• To address challenges in low-photon flux environments and computationally intensive algorithms.

Main Methods:

• Utilized a Single-Photon Avalanche Diode (SPAD) detector array with an integrated color filter array (CFA).
• Developed a bespoke Bayesian algorithm for high-speed image reconstruction.
• Validated on simulated data and experiments with multi-colored, moving targets under white-light illumination.

Main Results:

• Achieved high-fidelity color videos from single-photon data.
• Demonstrated per-frame reconstruction rates hundreds of times faster than previous methods.
• Obtained uncertainty measures for system optimization and decision-making.

Conclusions:

• The demonstrated techniques enable rapid video-rate single-photon color imaging.
• The Bayesian algorithm and SPAD technology pave the way for live 3D color videography in extremely low-light conditions.