A Bidirectional Deep Neural Network for Accurate Silicon Color Design

Abstract

Silicon nanostructure color has achieved unprecedented high printing resolution and larger color gamut than sRGB. The exact color is determined by localized magnetic and electric dipole resonance of nanostructures, which are sensitive to their geometric changes. Usually, the design of specific colors and iterative optimization of geometric parameters are computationally costly, and obtaining millions of different structural colors is challenging. Here, a deep neural network is trained, which can accurately predict the color generated by random silicon nanostructures in the forward modeling process and solve the nonuniqueness problem in the inverse design process that can accurately output the device geometries for at least one million different colors. The key results suggest deep learning is a powerful tool to minimize the computation cost and maximize the design efficiency for nanophotonics, which can guide silicon color manufacturing with high accuracy.

A cyclical deep learning based framework for simultaneous inverse and forward design of nanophotonic metasurfaces

“…However, the development of optimized metasurfaces for various functionalities requires one to go beyond the limitations of physical intuition. Recently, deep learning (DL) has been used for inverse design and optimization of metasurface-based nanostructures for directed functionality [8][9][10][11][12][13] . With the use of multiple models based on DL, computational expense has been significantly reduced, and design and optimization processes have become highly efficient.…”

Multi-dimensional manipulation of optical field withmetasurfaces and its optimization based on machine learning

“…(h) 不规则编码结构和线偏光入射时反射场的三维全波段模拟结果 [119] . (i) 原图(左)与结构色重现图样(右) 的对比图 [120] . (j) 高数值孔径的离散消色差透镜的结构示意图与电场分布图样 [112] Figure 8 (Color online) Manipulating optical field by machine learning.…”

“…(h) Geometry of the coding metasurface and the 3D full-wave simulation result of the left-handed circular polarization and righthanded circular polarization component [119] . (i) The designed colors of the painting (left) and the inverse-designed structural colors (right) [120] . (j) Optimal design of the near-unity-numerical aperture achromatic metalens and the normalized field-intensity profiles [112] 1300 nm的出射光.…”

Enhanced light–matter interactions in dielectric nanostructures via machine-learning approach

“…40 Recently, deep learning approaches, based on the artificial neural networks (ANNs), have emerged as a revolutionary and robust methodology in nanophotonics. [41][42][43][44][45][46][47][48][49][50][51][52][53][54][55] Indeed, applying the deep learning algorithms to the nanophotonic inverse design can introduce remarkable design flexibility that can go far beyond that of the conventional methods. The inverse design approach works based one the training process, that enables fast prediction of complex optical properties of nanostructures with intricate architectures.…”