Analog Optical Spatial Differentiators Based on Dielectric Metasurfaces. Issue 4 (11th December 2019)

• Record Type: Journal Article
• Authors: Zhou, Yi; Wu, Wenhui; Chen, Rui; Chen, Wenjie; Chen, Ruipin; Ma, Yungui
• Is Part Of: Advanced optical materials. Volume 8:Issue 4(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-12-11
• Subjects: edge detection -- magnetic resonance -- metasurfaces -- optical spatial differentiation -- optics in computing; Optical materials -- Periodicals; Photonics -- Periodicals; 620.11295
• DOI: 10.1002/adom.201901523
• Languages: English
• ISSNs: 2195-1071

Abstract

Abstract: Analog spatial differentiator is an important optical computational device that can be potentially used in the field of high‐speed edge detection and optical image processing. In the current stage, a general method is still required to robustly design compact devices for various spectrum or complex situation. In this work, a dielectric metasurface method is proposed and experimentally demonstrated to build optical spatial differentiators. This is physically realized by a high‐quality magnetic resonance mode that is hybridized with the classic bounded surface wave via grating coupling. Experimentally, a well‐defined first‐order differentiation effect is observed at oblique light incidence. The application potential of this device is evaluated by inputting various figures. It is shown that it can perform an optical processor to trace the profiles of the figure content with a resolution better than 30.8 µm. This work may pave a robust way to build ultracompact optical computation devices for real‐time image processors, which can be freely extended to various frequencies. Abstract : A first‐order analog optical spatial differentiation based on a metasurface is presented here. The hybrid coupling of a localized magnetic resonant mode with the classic bounded surface wave is utilized to produce the required transfer Green's functions. The experimental results verify the promising feature of the device in optical edge detection.