Single-mask sphere-packing with implicit neural representation reconstruction for ultrahigh-speed imaging.

Area of Science:

• Optical imaging
• High-speed photography
• Scientific instrumentation

Background:

• Single-shot, high-speed 2D optical imaging is crucial for capturing transient phenomena.
• Compressed optical-streaking ultra-high-speed photography (COSUP) achieves high frame rates but suffers from reconstruction artifacts due to random coded apertures.
• Non-multiplexing coded apertures offer simpler reconstruction for longer video recovery.

Purpose of the Study:

• To design a non-multiplexing coded aperture for COSUP.
• To improve the reconstruction process and reduce artifacts in ultra-high-speed videos.
• To enable longer video recovery from single-shot measurements.

Main Methods:

• Designed a non-multiplexing coded aperture for COSUP utilizing congruent sphere packing (SP).
• Developed an implicit neural representation for self-training from single measurements.
• Integrated galvanometer linear scanning with optimal SP encoding patterns for uniform space-time sampling.

Main Results:

• The new aperture design simplifies reconstruction algorithms for COSUP.
• The implicit neural representation significantly reduces training time and eliminates the need for training datasets.
• Reconstructed videos showed improved clarity and length compared to previous methods.

Conclusions:

• The proposed non-multiplexing coded aperture and implicit neural representation scheme enhance COSUP capabilities.
• This approach allows for more efficient and artifact-free ultra-high-speed video reconstruction.
• Validated through simulations and experimental results, demonstrating practical applicability.