Direct field-to-pattern monolithic design of holographic metasurface via residual encoderdecoder convolutional neural network

作     者：Ruichao Zhu Jiafu Wang Tianshuo Qiu Dingkang Yang Bo Feng Zuntian Chu Tonghao Liu Yajuan Han Hongya Chen Shaobo Qu Ruichao Zhu;Jiafu Wang;Tianshuo Qiu;Dingkang Yang;Bo Feng;Zuntian Chu;Tonghao Liu;Yajuan Han;Hongya Chen;Shaobo Qu

作者机构：Shaanxi Key Laboratory of Artificially-Structured Functional Materials and DevicesAir Force Engineering UniversityXi'an 710051China The Academy for Engineering&TechnologyFudan UniversityShanghai 200433China 

出 版 物：《Opto-Electronic Advances》 (光电进展（英文）)

年 卷 期：2023年第6卷第8期

页      面：29-38页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 080203[工学-机械设计及理论] 0835[工学-软件工程] 0802[工学-机械工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supports from the National Natural Science Foundation of China under Grant Nos.61971435,62101588,62101589 Natural Science Basic Research Program of Shaanxi Province(Grant No:2022JM-352,2022JQ-335,2023-JC-YB-069) the National Key Research and Development Program of China(Grant No.:SQ2017YFA0700201) 

主　　题：metasurface holography complex amplitude deep learning monolithic design 

摘      要：Complex-amplitude holographic metasurfaces(CAHMs)with the flexibility in modulating phase and amplitude profiles have been used to manipulate the propagation of wavefront with an unprecedented level,leading to higher image-reconstruction quality compared with their natural ***,prevailing design methods of CAHMs are based on Huygens-Fresnel theory,meta-atom optimization,numerical simulation and experimental verification,which results in a consumption of computing ***,we applied residual encoder-decoder convolutional neural network to directly map the electric field distributions and input images for monolithic metasurface design.A pretrained network is firstly trained by the electric field distributions calculated by diffraction theory,which is subsequently migrated as transfer learning framework to map the simulated electric field distributions and input *** training results show that the normalized mean pixel error is about 3%on *** verification,the metasurface prototypes are fabricated,simulated and *** reconstructed electric field of reverse-engineered metasurface exhibits high similarity to the target electric field,which demonstrates the effectiveness of our ***,this work provides a monolithic field-to-pattern design method for CAHMs,which paves a new route for the direct reconstruction of metasurfaces.