Image Processing Using High‐Index Dielectric Metasurfaces Based on Fano Resonance

Abstract

AbstractThis paper presents the design and analysis of high‐index dielectric metasurfaces based on Fano resonance for image processing applications. With the aim of optical edge detection, two different metasurface structures are proposed to create first‐ and second‐order derivative operators. The metasurfaces designed for second‐order derivative is a square lattice structure consisting of silicon cylinders surrounded by a silicon dioxide layer. Moreover, another metasurface with a square lattice structure composed of three different silicon cylinders embedded in a silicon dioxide layer is proposed to perform first‐order derivation. To achieve the best performance of the proposed metasurfaces in edge detection, the physical dimensions of the proposed structures are optimized by conducting parametric studies to reach the maximum available numerical aperture . Consequently, the second‐order derivative metasurface attains a numerical aperture with a magnitude , while the first‐order derivative metasurface achieves a numerical aperture of . Finally, the applicability of the proposed structures in optical edge detection is verified in the analog domain. The proposed metasurfaces provide high spatial resolution, high efficiency, and low dependence on the horizon angle, which pave the way for high‐quality image processing applications.