Deep learning-enabled inverse design of polarization-selective structural color based on coding metasurface
Haolin Yang(杨昊霖)1, Bo Ni(倪波)1, Junhong Guo(郭俊宏)2, Hua Zhou(周华)1,†, and Jianhua Chang(常建华)1
1 Jiangsu Key Laboratory of Meteorological Observation and Information Processing, School of Electronic and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China; 2 College of Electronic and Optical Engineering and College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications, Nanjing 210023, China
Abstract Structural colors based on metasurfaces have very promising applications in areas such as optical image encryption and color printing. Herein, we propose a deep learning-enabled reverse design of polarization-selective structural color based on coding metasurface. In this study, the long short-term memory (LSTM) neural network is presented to enable the forward and inverse mapping between coding metasurface structure and corresponding color. The results show that the method can achieve 98% accuracy for the forward prediction of color and 93% accuracy for the inverse design of the structure. Moreover, a cascaded architecture is adopted to train the inverse neural network model, which can solve the non-uniqueness problem of the polarization-selective color reverse design. This study provides a new path for the application and development of structural colors.
Haolin Yang(杨昊霖), Bo Ni(倪波), Junhong Guo(郭俊宏), Hua Zhou(周华), and Jianhua Chang(常建华) Deep learning-enabled inverse design of polarization-selective structural color based on coding metasurface 2025 Chin. Phys. B 34 050702
[1] Yu N F, Genevet P, Kats M A, Aieta F, Tetienne J P, Capasso F and Gaburro Z 2011 Science 334 333 [2] Xie X, Pu M B, Jin J J, Xu M F, Guo Y H, Li X, Gao P, Ma X L and Luo X G 2021 Phys. Rev. Lett. 126 183902 [3] Kumar K, Duan H G, Hegde R S, Koh S C W, Wei J N and Yang J K W 2012 Nat. Nanotechnol. 7 557 [4] Roberts A S, Pors A, Albrektsen O and Bozhevolnyi S I 2014 Nano Lett. 14 783 [5] Wu B Y, Wang M J, Sun Y S, Wu F, Shi Z X and Wu X H 2022 Adv. Compos. Hybrid Mater. 5 2527 [6] Yan Z D, Tang C J, Wu G H, Tang Y M, Gu P, Chen J, Liu Z Q and Huang Z 2021 Nanomaterials 11 63 [7] Pei Y, Sang T, Mi Q, Wang J C and Wang Y K 2022 J. Opt. 24 024001 [8] Reza M R 2021 Photon. Nanostruct. 43 100883 [9] Wang X Y, Chen M, Zhao W L, Li R J, Shi X Y, Han W H, Liu J B, Teng C X, Deng S J, Cheng Y and Yuan L B 2024 Opt. Laser Technol. 170 110190 [10] Liu Y Q, Chen W Q, Du X M, Shu Y C, Wu L J, Ren Z R, Yin H C, Sun J H, Qi K A, Che Y X and Li L S 2023 Result Phys. 47 106366 [11] Cui T J, Qi M Q, Wan X, Zhao J and Cheng Q 2014 Light Sci. Appl. 3 e218 [12] Wei Y X, Zhao M and Yang Z Y 2022 Opt. Lett. 47 5344 [13] Wu Z P, Zhang Z Q, Xu Y J, Zhai Y S, Zhang C R, Wang B Z and Wang Q L 2022 Opt. Lett. 47 4548 [14] Li H X, Long B, Wang T, Zhou F, Xu Y L and Zhang Z P 2024 Opt. Commun. 554 130205 [15] Cheng T, Ma Y K, Zhao H H, Fei T H, Liu L H and Yang J Y 2023 Nanophotonics 12 3121 [16] Zhou Y T, Wang Q Y, Ji Z Q and Zeng P 2022 Photonics 9 402 [17] Wen Y, Lin J, Chen K L, Lin Y S and Yang B R 2022 Opt. Laser Technol. 150 108004 [18] Gu J T, Liu Y, Meng N N, Sahmuganathan V, Tan S C, Sudijono, J, Tang J C, Venkatasubramanian E, Mallick A, Tjiptoharson F, Rezaei S D, Teo S L, Zhu Q, Chen Y J, Lin M, Dong Z G and Loh K P 2023 Adv. Opt. Mater. 11 2202826 [19] Wang L, Wang T, Yan R Q, Yue X Z, Wang H M, Wang Y D, Zhang J Y, Yuan X Y, Zeng J W and Wang J 2023 Nano Lett. 23 5581 [20] Li D Y, Wang W, Chu L Y and Deng N N 2023 Nano Lett. 23 9657 [21] Yang B, Liu W W, Li Z C, Cheng H, Choi D Y, Chen S Q and Tian J G 2019 Nano Lett. 19 4221 [22] Namhwa K and Seok E H 2023 J. Korean Phys. Soc. 82 166 [23] Liu X, Huang Z and Zang J F 2020 Nano Lett. 20 8739 [24] Zhang R, Guo X, Qiu H, Liu X, Han M, Jia T and Cheng H 2021 JETP Lett. 114 371 [25] Li X Z, Shu J, Gu W H and Gao L 2019 Opt. Mater. Express 9 3857 [26] Liu Z C, Zhu D Y, Rodrigues S P, Lee K T and Cai W S 2018 Nano Lett. 18 6570 [27] Liu Z C, Zhu D Y, Lee K T, Kim A S, Raju L and Cai W S 2020 Adv. Mater. 32 1904790 [28] Zhu L, Zhang C, Dong L, Rong M X, Gong J Y and Meng F Y 2023 Phys. Scr. 98 105501 [29] Ding Z P, Su W, Luo Y L, Ye L P, Li W L, Zhou Y H, Zou J F, Tang B and Yao H B 2024 Nanoscale 16 1384 [30] Ghorbani F, Shabanpour J, Beyraghi S, Soleimani H, Oraizi H and Soleimani M 2021 Appl. Phys. A 127 869 [31] Sajedian I, Kim J and Rho J 2019 Microsyst. Nanoeng. 5 27 [32] Wang J H, Lin Z C, Fan Y, Mei L Y, Deng W Q, Lv J W and Xu Z J 2022 Materials 15 7008 [33] Ma T G, Wang H Z and Guo L J 2024 Opto-Electron. Adv. 7 240062 [34] Palik E D 1985 Handbook of Optical Constants (Academic Press) [35] Duan X Y, Kamin S and Liu N 2018 Nat. Commun. 8 14606 [36] Xu Y L, Li H X, Zhang X, Liu W J, Zhang Z P, Qin S J and Liu J T 2021 Opt. Express 29 32491 [37] M. Quinten 2001 Appl. Phys. B 73 317 [38] Costa T L, Hamer R D, Nagy B V, BarboniMT S, Gualtieri M, Boggio P S and Ventura D F 2015 Exp. Brain Res. 233 1213
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
