KAFE-SPN: attention fusion enhanced-based single-pixel phase-shifting holography.

Area of Science:

• Optics and Photonics
• Computational Imaging
• Digital Holography

Background:

• Traditional digital holography faces limitations in single-pixel imaging, particularly regarding data acquisition and resolution.
• Existing methods often require mechanical phase shifters or multiple data captures, increasing complexity and data volume.

Purpose of the Study:

• To propose a novel single-pixel phase-shifting holographic imaging method.
• To overcome the limitations of traditional digital holography in single-pixel imaging applications.
• To develop a high-precision 3D reconstruction approach for holographic imaging.

Main Methods:

• A Kolmogorov-Arnold attention fusion-enhanced based single-pixel phase-shifting network (KAFE-SPN) was developed.
• A 1D compressed data-to-multi-phase-shifted hologram mapping framework was established.
• Four holograms with phase shifts of 0, π/2, π, and 3π/2 were generated without mechanical components or multiple acquisitions.

Main Results:

• The KAFE-SPN method successfully generated high-quality holograms at a 20% sampling rate.
• Stable performance was maintained under extremely low sampling rates and high noise conditions.
• High resolution and depth information were retained despite data reduction.

Conclusions:

• The proposed method significantly reduces imaging data requirements while preserving essential holographic information.
• KAFE-SPN offers a promising high-precision 3D reconstruction solution for holographic imaging.
• This approach enhances the efficiency and robustness of single-pixel holographic imaging.