Double-phase metasurface operators for all-optical image processing

Area of Science:

• Optics and Photonics
• Nanotechnology
• Computational Science

Background:

• Conventional electronic systems for image processing are limited by speed and energy consumption.
• Traditional optical processors often require bulky components, hindering miniaturization and integration.
• There is a need for compact, efficient platforms for analog optical computing.

Purpose of the Study:

• To develop a compact metasurface-based platform for analog optical computing.
• To demonstrate arbitrary image transformations and complex computational operations using a single passive nanophotonic device.
• To showcase applications in high-resolution complex holography and volumetric wavefront control.

Main Methods:

• Utilizing double-phase encoding and polarization multiplexing on a metasurface.
• Implementing arbitrary image transformations within a single passive nanophotonic device.
• Experimentally demonstrating differentiation, cross-correlation, vertex detection, and Laplacian differentiation.

Main Results:

• Achieved arbitrary image transformations with a compact metasurface.
• Successfully performed key analog optical computations including differentiation and cross-correlation.
• Demonstrated high-resolution complex holography with subwavelength-scale volumetric wavefront control for depth-resolved reconstructions.

Conclusions:

• Established a scalable and versatile platform for computational optics.
• The metasurface approach eliminates the need for complex optical setups and digital post-processing.
• Potential applications include real-time image processing, energy-efficient computing, biomedical imaging, and holographic displays.