Generative-adversarial-network-based dimensional measurement of optical waveguides

We propose a high-throughput and precise waveguide-dimensional-measurement method consisting of a generative adversarial network (GAN) and curve-fitting-based dimensional calculator using sidewall functions. The GAN can learn the differences between low-magnification (LM) and high-magnification (HM) optical microscope images taken with different objective lenses at different magnifications over the same area. The LM and HM images of the waveguides are captured using an optical microscope at magnifications of 500x and 2000x, respectively. We obtained a standard deviation of the waveguide widths of approximately 0.8 pixels (similar to 42 nm), and confirmed precise width measurement using super-resolution images at the same imaging throughput as with an LM microscope. (C) 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

Automated summary

We propose a high-throughput and precise waveguide-dimensional-measurement method using a generative adversarial network (GAN) and curve-fitting-based dimensional calculator. The GAN can learn and differentiate optical microscope images taken with different objective lenses and magnifications, and precise measurement is achieved using super-resolution images.