Abstract

Forward prediction of directional scattering from all-dielectric nanostructures by a two-level nested U-shaped convolutional neural network (U²-Net) is investigated. Compared with the traditional U-Net method, the U²-Net model with lower model height outperforms for the case of a smaller image size. For the input image size of 40 × 40, the prediction performance of the U²-Net model with the height of three is enhanced by almost an order of magnitude, which can be attributed to the more excellent capacity in extracting richer multi-scale features. Since it is the common problem in nanophotonics that the model height is limited by the smaller image size, our findings can promote the nested U-shaped network as a powerful tool applied to various tasks concerning nanostructures.

© 2022 Optica Publishing Group

Prediction of metasurface spectral response based on a deep neural network

Ying Chen, Zhixin Ding, JianKun Wang, Jian Zhou, and Min Zhang
Opt. Lett. 47(19) 5092-5095 (2022)

Performance of a U-Net-based neural network for predictive adaptive optics

Justin G. Chen, Vinay Shah, and Lulu Liu
Opt. Lett. 46(10) 2513-2516 (2021)

Circular dichroism of a tilted U-shaped nanostructure

TianKun Wang, Tong Fu, Yuyan Chen, and Zhongyue Zhang
Opt. Lett. 42(14) 2842-2845 (2017)

Convolutional neural networks used for random structure SPP gratings spectral response prediction

Tianle Qu, Liping Zhu, and Zhenghua An
Opt. Lett. 48(2) 448-451 (2023)

Thermo-optical reshaping of second-harmonic emission from dimer all-dielectric nanoresonators

Olesia Pashina, Kristina Frizyuk, George Zograf, and Mihail Petrov
Opt. Lett. 47(8) 1992-1995 (2022)