Design method and device of polarized beam splitting metasurface and beam splitter

Abstract

The invention provides a polarized beam splitting metasurface design method and device and a beam splitter, and belongs to the technical field of optics. The method comprises the following steps: acquiring a far-field intensity distribution image; performing feature extraction and reconstruction on the far-field intensity distribution image to obtain a right-handed circular polarization phase distribution diagram and a left-handed circular polarization phase distribution diagram of the far-field intensity distribution image; inputting the right-handed circular polarization phase distribution diagram and the left-handed circular polarization phase distribution diagram into a preset angular spectrum propagation model so as to recover and obtain metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model; and carrying out polarized light beam splitting metasurface design based on the metasurface phase distribution. According to the technical scheme, the design efficiency of the polarized beam splitting metasurface can be improved.

Description

Technical Field

[0001] The present application relates to the field of optical technologies, and particularly to a design method, a device, and a beam splitter for a polarization beam splitting metasurface. Background Art

[0002] The basic building blocks in advanced photonics systems can manipulate light according to the polarization state of light. Although components such as bulk optical elements can also achieve the function of manipulating light according to the polarization state of light, with the increasing demand for miniaturization, integration, and multifunctionality in modern photonics devices, the research on optical elements has shifted to the development of more compact and multifunctional alternatives, such as metasurface polarization beam splitters, etc.

[0003] In related technologies, as a two-dimensional subwavelength optical nanoresonator array, the metasurface provides a new platform for manipulating light at the nanoscale. However, the current metasurface design still has the technical problem of low design efficiency in the miniaturization and integration of optical elements. Summary of the Invention

[0004] The main objective of the embodiments of the present application is to propose a design method, a device, and a beam splitter for a polarization beam splitting metasurface, aiming to improve the design efficiency of the polarization beam splitting metasurface.

[0005] To achieve the above objective, a first aspect of the embodiments of the present application proposes a design method for a polarization beam splitting metasurface, the method comprising:

[0006] Obtain a far-field intensity distribution image;

[0007] Extract features and reconstruct the far-field intensity distribution image to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image;

[0008] Input the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover a metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model;

[0009] Design a polarization beam splitting metasurface based on the metasurface phase distribution.

[0010] In some embodiments, the step of inputting the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover a metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model includes:

[0011] Input the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map as model input data into a preset angular spectrum propagation model;

[0012] Based on the angular spectrum propagation model, perform far-field intensity prediction calculation with the model input data to obtain the predicted far-field intensity;

[0013] Based on a preset intensity-phase gradient relationship, perform backpropagation optimization on the mean square error between the predicted far-field intensity and the input far-field intensity to obtain the metasurface phase distribution corresponding to the far-field intensity distribution image; the input far-field intensity is the far-field intensity of the far-field intensity distribution image.

[0014] In some embodiments, the method further includes:

[0015] Convert a preset angular spectrum propagation model into a differentiable neural network layer, and set the angular spectrum propagation model as a partial embedded training loop architecture to construct a physics-informed deep learning framework;

[0016] Wherein, the angular spectrum propagation model is the physics-informed layer in the deep learning framework; the deep learning framework is used to perform neural network training with the far-field intensity distribution image as the input data, and output the metasurface phase distribution corresponding to the far-field intensity distribution image recovered based on the angular spectrum propagation model.

[0017] In some embodiments, the feature extraction and reconstruction of the far-field intensity distribution image to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image includes:

[0018] Input the far-field intensity distribution image into a preset convolutional neural network architecture; the convolutional neural network architecture includes a shared encoder and dual decoder branches;

[0019] Based on the shared encoder and the dual decoder branches, perform feature extraction and reconstruction on the far-field intensity distribution image to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0020] In some embodiments, the based on the shared encoder and the dual decoder branches, perform feature extraction and reconstruction on the far-field intensity distribution image to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image includes:

[0021] Based on the shared encoder, perform feature extraction on the far-field intensity distribution image to obtain hierarchical features;

[0022] Based on the first branch of the dual decoder branch, reconstruct the right-handed circular polarization phase distribution with the hierarchical features as the input to obtain the right-handed circular polarization phase distribution map of the far-field intensity distribution image, and, based on the second branch of the dual decoder branch, reconstruct the left-handed circular polarization phase distribution with the hierarchical features as the input to obtain the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0023] In some embodiments, the design of the polarization beam splitting metasurface based on the metasurface phase distribution includes:

[0024] Convert the metasurface phase distribution into nano-pillar structure parameters, and design a nano-pillar superlattice based on the nano-pillar structure parameters to construct a polarization beam splitting metasurface.

[0025] To achieve the above object, a second aspect of the embodiments of the present application proposes a design device for a polarization beam splitting metasurface, the device includes:

[0026] An acquisition module, configured to acquire a far-field intensity distribution image;

[0027] An image processing module, configured to perform feature extraction and reconstruction on the far-field intensity distribution image to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image;

[0028] An intensity-phase mapping module, configured to input the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover the metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model;

[0029] A metasurface design module, configured to perform polarization beam splitting metasurface design based on the metasurface phase distribution.

[0030] To achieve the above object, a third aspect of the embodiments of the present application proposes a metasurface beam splitter, and the metasurface of the metasurface beam splitter is constructed based on the design method of the polarization beam splitting metasurface described in the first aspect above.

[0031] To achieve the above object, a fourth aspect of the embodiments of the present application proposes an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the method described in the first aspect above is implemented.

[0032] To achieve the above object, a fifth aspect of the embodiments of the present application proposes a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect above is implemented.

[0033] To achieve the above object, a sixth aspect of the embodiments of the present application proposes a computer program product. The computer program product stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect above is implemented.

[0034] The design method, device, metasurface beam splitter, electronic device, computer-readable storage medium, and computer program product of the polarization beam splitting metasurface proposed in the embodiments of the present application include: obtaining a far-field intensity distribution image; performing feature extraction and reconstruction on the far-field intensity distribution image to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image; inputting the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover the metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model; and designing the polarization beam splitting metasurface based on the metasurface phase distribution.

[0035] In this way, after obtaining the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image, the embodiments of the present application determine the metasurface phase distribution corresponding to the far-field intensity distribution image through the angular spectrum propagation model based on the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map. This can not only achieve an accurate mapping from the far-field intensity to the phase distribution, but also effectively reduce the computing resources required for metasurface design, thereby improving the design efficiency of the polarization beam splitting metasurface. Description of the Drawings

[0036] Figure 1 It is a schematic flowchart of the steps of the design method of the polarization beam splitting metasurface provided by the embodiments of the present application in some embodiments;

[0037] Figure 2 For Figure 1 it is a schematic flowchart of the refined steps of step S102 in

[0038] Figure 3 For Figure 2 it is a schematic flowchart of the refined steps of step S202 in

[0039] Figure 4 It is a schematic diagram of the U-Net architecture of the convolutional neural network based on deep learning involved in the design method of the polarization beam splitting metasurface provided by the embodiments of the present application in some embodiments;

[0040] Figure 5 For Figure 1 it is a schematic flowchart of the refined steps of step S103 in

[0041] Figure 6Schematic diagram of the physics-informed feedback framework based on deep learning involved in the design method of the polarization beam splitting metasurface proposed in the embodiments of the present application;

[0042] Figure 7 Schematic diagram of constructing a metasurface by realizing complete Jones matrix decoupling through geometric parameter engineering and Pancharatnam-Berry (PB) phase modulation involved in the embodiments of the present application

[0043] Figure 8 Schematic diagram of the structure of the design device of the polarization beam splitting metasurface provided by the embodiments of the present application;

[0044] Figure 9 Schematic diagram of the hardware structure of the electronic device provided by the embodiments of the present application. Detailed implementation manners

[0045] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.

[0046] It should be noted that although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order from the module division in the device or the order in the flowchart. Terms such as "first" and "second" in the specification, claims and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0048] First, the overall concept of the embodiments of the present application will be described.

[0049] Elements that can manipulate light according to the polarization state of light are the basic building blocks in advanced photonics systems. Although traditional bulk optical elements can achieve these functions, with the increasing demand for miniaturization, integration and multifunctionality in modern photonic devices, the research has turned to more compact and multifunctional alternatives. As a two-dimensional subwavelength optical nanoresonator array, metasurfaces provide a new platform for manipulating light at the nanoscale. However, it is still challenging to design a metasurface polarization beam splitter that simultaneously achieves high efficiency, large extinction ratio and arbitrary beam directivity.

[0050] The embodiments of the present application propose a design method, device, and beam splitter for a polarization beam splitting metasurface. Based on a physics-informed deep learning method, it designs a metasurface beam splitter with arbitrary polarization multiplexing, and can achieve complete Jones matrix decoupling through the design of a silicon nanocolumn superlattice, thereby being able to independently control the orthogonal polarization states, and further realizing the polarization beam splitting function with high efficiency, large extinction ratio, and arbitrary beam directivity.

[0051] The design method of the polarization beam splitting metasurface proposed in the embodiments of the present application is based on a physics-informed deep learning framework to design a metasurface beam splitter with arbitrary polarization multiplexing. Through three core technical components: an improved U-Net architecture, an angular harmonic propagation model, and a silicon nanocolumn superlattice design, it effectively solves the problems existing in the miniaturization and integration of traditional optical elements, such as low efficiency, poor extinction ratio, and difficult directivity control.

[0052] The core technical means of the design method of the polarization beam splitting metasurface provided in the embodiments of the present application include: (1) a U-Net architecture optimized for the characteristics of the optical field, which realizes efficient feature extraction and reconstruction through an encoder-decoder structure and skip connections; (2) an angular harmonic propagation model that embeds physical information into the deep learning framework to achieve an accurate mapping from far-field intensity to phase distribution; (3) a silicon nanocolumn superlattice design based on geometric parameter engineering to achieve precise control of the Pancharatnam-Berry phase (a phase concept in quantum mechanics that describes the phase rotation of light in an optical device).

[0053] Through the above technical means, the embodiments of the present application can achieve the following technical effects: (1) The design efficiency is significantly improved, and the computational resource requirements are reduced by more than 50%; (2) A high extinction ratio of 34.11 dB is achieved, which is about 15% higher than the related technologies; (3) A high efficiency of 63.91% is achieved; (4) It supports precise control of arbitrary polarization states, and the applicable range is extended to multiple application fields such as quantum optics and biological imaging.

[0054] Next, based on the overall concept of the above embodiments of the present application, specific embodiments of the design method, device, metasurface beam splitter, electronic device, computer-readable storage medium, and computer program product of the polarization beam splitting metasurface provided in the embodiments of the present application are proposed. First, each specific embodiment of the design method of the polarization beam splitting metasurface in the embodiments of the present application is described in detail.

[0055] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is a theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0056] Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. Artificial intelligence software technologies mainly include several major directions such as computer vision technology, robotics, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0057] It should be noted that in each specific embodiment of the present application, when relevant processing needs to be performed based on data related to the user's identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first. Moreover, the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the user's separate permission or separate consent will be obtained through methods such as pop-up windows or redirecting to a confirmation page. After clearly obtaining the user's separate permission or separate consent, the necessary user-related data for the normal operation of the embodiments of the present application will be obtained.

[0058] In addition, the design method of the polarization beam splitting metasurface provided by the embodiments of the present application can be applied to terminals, can also be applied to server sides, or can be software running on terminals or server sides. In some embodiments, the terminal can be a terminal device such as a smart phone, a tablet computer, a laptop computer, or a desktop computer; the server side can be configured as an independent physical server, can also be configured as a server cluster or distributed system composed of multiple physical servers, or can be configured as a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the design method of the polarization beam splitting metasurface, etc., but is not limited to the above forms.

[0059] Alternatively, the embodiments of the present application can also be used in numerous general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer computer devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0060] For the convenience of understanding and elaboration, in the following text, the design method of the polarization beam splitting metasurface provided by the embodiments of the present application applied to a terminal device will be taken as an example for detailed description. The implementation of the design method of the polarization beam splitting metasurface provided by the embodiments of the present application for any of the above forms of the theme can refer to the process of the design method of the polarization beam splitting metasurface applied to a terminal device described later.

[0061] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of the steps in some embodiments of the design method of the polarization beam splitting metasurface provided by the embodiments of the present application. It should be understood that although Figure 1 shows the execution order of some method steps, based on different design requirements in actual applications, the design method of the polarization beam splitting metasurface provided by the embodiments of the present application can of course adopt an execution order different from that shown in the figure. That is, Figure 1 the order of the method steps shown does not constitute a limitation on the execution logic order of the design method of the polarization beam splitting metasurface provided by the embodiments of the present application, and any reasonable changes based on Figure 1 the order of the method steps shown should be included in the protection scope of the design method of the polarization beam splitting metasurface provided by the embodiments of the present application.

[0062] As Figure 1 shown, in some embodiments, the design method of the polarization beam splitting metasurface provided by the embodiments of the present application may include but is not limited to steps S101 to S104.

[0063] Step S101: Obtain a far-field intensity distribution image.

[0064] During the process of designing a polarization beam splitting metasurface by a terminal device, the far-field intensity distribution image is first obtained as the input data for the entire process of designing the polarization beam splitting metasurface.

[0065] In some embodiments, the terminal device can collect a target far-field intensity distribution image including information such as the beam splitting point position, beam splitting intensity ratio, polarization state, etc. as the input data.

[0066] Step S102: Extract features and reconstruct the far-field intensity distribution image to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0067] After the terminal device obtains the far-field intensity distribution image as the input data, through a preset U-Net architecture, based on the encoder-decoder structure and skip connections, the far-field intensity distribution image is subjected to feature extraction and reconstruction, so as to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0068] It should be noted that the U-Net architecture has a shared encoder and dual decoder branches, so as to be able to optimize the phase distributions of right-handed circularly polarized light (RCP) and left-handed circularly polarized light (LCP) simultaneously. Among them, the phase distribution of RCP is the right-handed circular polarization phase distribution map, and the phase distribution of LCP is the left-handed circular polarization phase distribution map.

[0069] Step S103: Input the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover the metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model.

[0070] After the terminal device obtains the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image, it further inputs the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model, so as to perform an accurate mapping from the far-field intensity to the phase distribution based on the angular spectrum propagation model with the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map, and to recover the metasurface phase distribution corresponding to the far-field intensity distribution image.

[0071] Step S104: Design a polarization beam splitting metasurface based on the metasurface phase distribution.

[0072] After the terminal device further obtains the metasurface phase distribution corresponding to the far-field intensity distribution image, it can design a polarization beam splitting metasurface based on the metasurface phase distribution, so as to obtain the designed metasurface beam splitter.

[0073] In the embodiments of the present application, a far-field intensity distribution image is obtained by a terminal device as input data for the entire design process of a polarization beam splitting metasurface. Then, through a preset U-Net architecture, based on an encoder-decoder structure and skip connections, feature extraction and reconstruction are performed on the far-field intensity distribution image, so as to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image. After that, the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map are further input into a preset angular spectrum propagation model, so as to perform an accurate mapping from the far-field intensity to the phase distribution based on the angular spectrum propagation model, and to recover the metasurface phase distribution corresponding to the far-field intensity distribution image. Finally, based on the metasurface phase distribution, a polarization beam splitting metasurface is designed, so as to obtain a designed metasurface beam splitter.

[0074] In this way, after obtaining the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image in the embodiments of the present application, the angular spectrum propagation model is used to determine the metasurface phase distribution corresponding to the far-field intensity distribution image based on the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map, which can not only achieve an accurate mapping from the far-field intensity to the phase distribution, but also effectively reduce the computing resources required for metasurface design, thereby improving the design efficiency of the polarization beam splitting metasurface.

[0075] Please refer to Figure 2 , Figure 2 For Figure 1 a schematic diagram of the detailed step flow of step S102 in

[0076] In some embodiments, as Figure 2 shown, the above step S102: performing feature extraction and reconstruction on the far-field intensity distribution image to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image may include, but is not limited to, steps S201 and S202 shown below.

[0077] Step S201: inputting the far-field intensity distribution image into a preset convolutional neural network architecture; the convolutional neural network architecture includes a shared encoder and dual decoder branches.

[0078] During the process of the terminal device performing feature extraction and reconstruction on the far-field intensity distribution image, the far-field intensity distribution image is input into a preset convolutional neural network architecture, so as to extract the deep features of the image through the encoder of the convolutional neural network architecture. Among them, each encoding block contains two 3×3 convolutional layers and one max-pooling layer.

[0079] It should be noted that the preset convolutional neural network architecture is the above-mentioned U-Net architecture. The U-Net architecture is obtained by pre-improving the U-Net network. This U-Net architecture has a shared encoder and dual decoder branches, and can optimize the phase distributions of right-handed circularly polarized light (RCP) and left-handed circularly polarized light (LCP) simultaneously.

[0080] Step S202: Based on the shared encoder and the dual decoder branches, perform feature extraction and reconstruction on the far-field intensity distribution image to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0081] After the terminal device inputs the far-field intensity distribution image into the U-Net architecture, the U-Net architecture performs feature extraction and reconstruction on the far-field intensity distribution image through the shared encoder and the dual decoder branches, so as to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image by optimizing the phase distributions of right-handed circularly polarized light (RCP) and left-handed circularly polarized light (LCP) simultaneously.

[0082] Please refer to Figure 3 , Figure 3 is Figure 2 the schematic flow diagram of the refinement steps of step S202 in

[0083] In some embodiments, as Figure 3 shown, the above-mentioned step S202 may include, but is not limited to, the following step S301 and step S302.

[0084] Step S301: Based on the shared encoder, perform feature extraction on the far-field intensity distribution image to obtain hierarchical features.

[0085] When the terminal device performs feature extraction and reconstruction on the far-field intensity distribution image based on the U-Net architecture, first, through the shared encoder in the U-Net architecture, perform deep image feature extraction on the far-field intensity distribution image to obtain the hierarchical features of the far-field intensity distribution image.

[0086] Step S302: Based on the first branch in the dual decoder branches, use the hierarchical features as input to reconstruct the right-handed circular polarization phase distribution to obtain the right-handed circular polarization phase distribution map of the far-field intensity distribution image, and based on the second branch in the dual decoder branches, use the hierarchical features as input to reconstruct the left-handed circular polarization phase distribution to obtain the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0087] After the terminal device extracts the hierarchical features of the far-field intensity distribution image through the shared encoder in the U-Net architecture, it further passes through the first branch of the dual decoder branches of the U-Net architecture, and reconstructs the right-handed circular polarization phase distribution with the hierarchical features as the input, so as to obtain the right-handed circular polarization phase distribution map of the far-field intensity distribution image. Moreover, the terminal device passes through the second branch of the dual decoder branches of the U-Net architecture, and at the same time reconstructs the left-handed circular polarization phase distribution with the hierarchical features as the input, so as to obtain the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0088] Exemplarily, the U-Net architecture can be composed of three main components: an encoder path dedicated to feature extraction, a bottleneck layer for feature transformation, and a dual decoder path responsible for phase reconstruction. As Figure 4 shown, the shared encoder path (yellow block) in the U-Net architecture first extracts hierarchical features from the far-field intensity distribution image, and the extracted hierarchical features are supplemented by the intermediate pooling layer (gray block). Then, since the U-Ne architecture integrates dual decoder branches (each branch consists of a transposed convolutional layer (brown block) for upsampling and a feature refinement convolutional block (red block)), it can simultaneously reconstruct the phase distributions of right-handed circular polarization (RCP, upper) and left-handed circular polarization (LCP, bottom), that is: the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image. In addition, the U-Ne architecture integrates skip connections (horizontal arrows) by connecting the corresponding encoder and decoder levels to retain spatial details. The feature dimensions of each level are represented by numbers. The outputs (RCP and LCP) demonstrate the ability to recover complementary phase patterns from a single target intensity image (Target, that is: the far-field intensity distribution image), and achieve the polarization selection function through physical information optimization.

[0089] In some embodiments, the U-Net architecture uses a systematic mathematical formulation. Among them, the encoding path passes through continuous convolution and pooling operations, and its mathematical formula can be expressed as:

[0090]

[0091] Among them, el represents the feature map of the l-th layer of the encoding path, * represents the convolution operation, σ is the ReLU activation function, is the 2×2 max pooling operation. d l，RCP and d l，LCP represent the feature maps of the l-th layer of the dual decoder path, [,] represents the feature concatenation along the channel dimension, and U represents upsampling using a 2×2 kernel and a stride of 2 transposed convolution.

[0092] In addition, the U-Net architecture performs feature transformation at the network bottleneck layer, connecting the encoder and the decoder. The mathematical formula is expressed as follows:

[0093] f bottleneck = σ(W bottleneck *e L +b bottleneck ),

[0094] where L is the depth of the encoder path. The double convolutional blocks in the decoder gradually reduce the feature complexity, ensuring a smooth transition from high-level features to detailed phase maps. By reconstructing features in the decoder, each decoding block contains an upsampling layer, two 3×3 convolutional layers, and skip connections. In this way, the U-Net architecture can output the phase distribution maps of RCP and LCP polarized light.

[0095] In this embodiment, by adopting this design of the above U-Net architecture, the hierarchical features of the far-field intensity distribution image can be efficiently captured, and at the same time, the independent and synchronous reconstruction of the right-handed circular polarization (RCP) and left-handed circular polarization (LCP) phase distributions can be achieved through a single forward propagation, obtaining the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image. In this way, the design process is significantly optimized.

[0096] Please refer to Figure 5 , Figure 5 which is Figure 1 the detailed step flow schematic diagram of step S103 in

[0097] In some embodiments, as Figure 5 shown, the above step S103: inputting the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover the metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model may include, but is not limited to, steps S501 to S503 shown below.

[0098] Step S501: Input the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map as model input data into a preset angular spectrum propagation model.

[0099] Step S502: Based on the angular spectrum propagation model, perform far-field intensity prediction calculation with the model input data to obtain a predicted far-field intensity.

[0100] Step S503: Based on a preset intensity-phase gradient relationship, perform reverse optimization on the mean square error between the predicted far-field intensity and the input far-field intensity to obtain the metasurface phase distribution corresponding to the far-field intensity distribution image; the input far-field intensity is the far-field intensity of the far-field intensity distribution image.

[0101] After obtaining the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image, the terminal device further inputs the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map as model input data into a preset angular spectrum propagation model, so as to perform far-field intensity prediction calculation through the angular spectrum propagation model with the model input data to obtain the predicted far-field intensity, that is, through Fourier transform, frequency-domain multiplication and inverse Fourier transform, calculate the far-field intensity distribution corresponding to the model input data, and this far-field intensity distribution is the predicted far-field intensity. Finally, through the pre-established intensity-phase gradient relationship, the mean square error between the predicted far-field intensity and the actual far-field intensity of the far-field intensity distribution image is reversely optimized, so as to obtain the metasurface phase distribution corresponding to the far-field intensity distribution image.

[0102] In some embodiments, the terminal device can construct an angular spectrum propagation model by transforming the traditional angular spectrum propagation formula into a differentiable neural network layer. By using the phase distribution predicted by the network (the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image extracted by the hierarchical features based on the U-Net architecture and reconstructed) as the input of the angular spectrum propagation layer, the angular spectrum propagation model calculates the corresponding far-field intensity distribution (predicted far-field intensity) through Fourier transform, frequency-domain multiplication and inverse Fourier transform. In addition, by establishing the gradient relationship between intensity and phase (intensity-phase gradient relationship), the phase distribution prediction can be reversely optimized directly from the intensity error. Finally, by combining the mean square error and physical constraints (such as energy conservation and phase continuity), it can be ensured that the prediction result conforms to physical laws.

[0103] In some embodiments, the design method of the polarization beam splitting metasurface provided by the embodiments of the present application may further include but is not limited to the following steps:

[0104] Transform the preset angular spectrum propagation model into a differentiable neural network layer, and set the angular spectrum propagation model as a partially embedded training loop architecture to construct a deep learning framework driven by physical information.

[0105] It should be noted that the angular spectrum propagation model is a physical information-driven layer in the deep learning framework; the deep learning framework is used to perform neural network training with the far-field intensity distribution image as input data and output the metasurface phase distribution corresponding to the far-field intensity distribution image restored by the angular spectrum propagation model.

[0106] The terminal device can construct a deep learning framework driven by physical information by transforming a preset angular spectrum propagation model into a differentiable neural network layer and setting the angular spectrum propagation model as a partial embedded training loop architecture. Thus, when restoring the metasurface phase distribution corresponding to the far-field intensity distribution image through the angular spectrum propagation model, the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image extracted and reconstructed by the U-Net architecture based on this deep learning architecture can be used as the input of the angular spectrum propagation layer. Then, through the angular spectrum propagation model, Fourier transform, frequency domain multiplication, and inverse Fourier transform are performed to calculate the corresponding far-field intensity distribution, and the gradient relationship between intensity and phase is established, enabling the deep learning architecture to directly optimize the phase distribution prediction in reverse from the intensity error.

[0107] Exemplarily, the deep learning architecture is as Figure 6 shown. The angular spectrum propagation model is embedded as a part of the deep learning architecture into the training loop architecture ( Figure 6 as described by the pink arrow and the orange dashed box in), starting from the intensity input (far-field intensity distribution image), the deep learning architecture processes and predicts phase information through the U-Net architecture, and then inputs the obtained phase (the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image) into the physical model (angular spectrum propagation model) to predict the intensity. The mean square error loss between the predicted intensity and the input intensity is backpropagated through an Adaptive Learning Rate (Adam) optimizer to achieve end-to-end training.

[0108] In addition, Figure 6 the blue dashed box in marks the training loop, the pink arrow represents the backpropagation path, and the orange dashed line highlights the physical information feedback mechanism. The output of the deep learning architecture can be converted into a complex optical field through a formula, and then the predicted far-field intensity is generated through angular spectrum propagation, thereby determining the final phase distribution.

[0109] In this embodiment, after the terminal device reconstructs the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image during RCP and LCP phase distribution reconstruction, an angular spectrum propagation model is introduced as a physical information-driven layer to connect the phase distribution and the far-field intensity pattern. Since the angular spectrum propagation model can accurately describe the relationship between the metasurface phase distribution and the far-field intensity pattern, this rigorous electromagnetic framework provides a solution for paraxial optical systems that is significantly superior in computational efficiency to full-wave simulation while maintaining physical accuracy. In addition, by constructing a deep learning framework and embedding the angular spectrum propagation model directly into the neural network training process to restore the phase distribution from the target far-field intensity distribution, the need for a large number of pre-generated datasets is eliminated.

[0110] In some embodiments, the above step S104: Designing a polarization beam splitting metasurface based on the metasurface phase distribution may include, but is not limited to, the following steps:

[0111] Convert the metasurface phase distribution into nanocolumn structure parameters, and design a nanocolumn superlattice based on the nanocolumn structure parameters to construct a polarization beam splitting metasurface.

[0112] After the terminal device obtains the metasurface phase distribution corresponding to the far-field intensity distribution image, when designing the polarization beam splitting metasurface based on this metasurface phase distribution, by converting the finally determined metasurface phase distribution into nanocolumn structure parameters, and then designing a silicon nanocolumn superlattice based on geometric parameter engineering, a polarization beam splitting metasurface can be constructed.

[0113] Exemplarily, the terminal device converts the phase distribution diagram (the metasurface phase distribution corresponding to the far-field intensity distribution image) determined finally into specific nanocolumn structure parameters according to the following formula.

[0114]

[0115] Among them, φRCP and φLCP respectively represent the phase distributions of right-handed circularly polarized light (RCP) and left-handed circularly polarized light (LCP), S represents the activation function Sigmod, and Wfinal,RCP and Wfinal,LCP are weight coefficients related to the polarization state. Based on this, the phase modulation of RCP and LCP can be converted into the structure parameters of the nanocolumns (such as diameter, height).

[0116] Next, a specific embodiment of the metasurface beam splitter provided in the embodiments of the present application will be described.

[0117] The metasurface beam splitter provided in the embodiments of the present application is constructed based on the design method of the polarization beam splitting metasurface provided in the above embodiments of the present application. That is, after the terminal device constructs a polarization beam splitting metasurface by executing the specific embodiment of the design method of the polarization beam splitting metasurface provided in the above embodiments of the present application, based on this polarization beam splitting metasurface, a metasurface beam splitter can be further designed.

[0118] In some embodiments, the structure of the designed metasurface beam splitter provided by the embodiments of the present application may include: (1) a substrate layer: composed of an amorphous silicon (α-Si) and silica substrate, with a thickness of 500 nm; (2) a nano-pillar array layer: composed of α-Si nano-pillars with a diameter ranging from 100 to 400 nm, a height of 700 nm, and a period of 750 nm; (3) a control layer: including an input light source control unit, a polarization state detection unit, and an output beam analysis unit.

[0119] In some embodiments, the polarization beam splitting function of the metasurface beam splitter can achieve complete Jones matrix decoupling through geometric parameter engineering and Pancharatnam-Berry (PB) phase modulation, and can independently control the orthogonal polarization states. The mathematical relationship expression of the PB phase and the RCP / LCP phase is as follows:

[0120]

[0121] where Φtotal RCP / LCP is the total phase of RCP or LCP, which is jointly determined by the geometric phase and the PB phase, ΦgeoRCP / LCP(L,W) is the geometric phase, which is related to the length (L) and width (W) of the nano-pillars, θ(x,y) is the rotation angle of the nano-pillars, and is a function of the position (x,y).

[0122] Based on this, the terminal device can obtain the corresponding rotation angle of the nano-pillars to construct the metasurface, and its mathematical expression formula of the rotation angle is as follows:

[0123]

[0124] where ΦtargetRCP / LCP(x,y) is the target phase distribution, that is, the phase values that RCP and LCP are expected to reach at the position (x,y).

[0125] Exemplarily, please refer to Figure 7 , Figure 7 In (a) is a schematic diagram of the unit structure in the metasurface, showing rectangular silicon nano-pillars (α-Si) on a silica substrate (SiO2), and the key geometric parameters are marked in the figure. Figure 7 In (b) is a phase response diagram under right-handed circularly polarized light (RCP) and left-handed circularly polarized light (LCP) irradiation, showing a complete 2π phase coverage in the parameter space when the length and width of the nano-pillars change. In addition, Figure 7 In (c) is a schematic diagram of the supercell configuration. Among them, on the left is a supercell composed of 2×2 nano-pillars, and on the right is a top view of the entire metasurface array including 32×32 supercells. Figure 7In (d), it is the top-view schematic diagram of the 2×2 supercell configuration, showing the spatial distribution of the nanocolumns.

[0126] It should be noted that the connection mechanisms between the realization of the polarization beam splitting function of the metasurface beam splitter and the geometric parameter engineering and PB phase modulation include: phase mapping, parameter optimization process, joint coding of rotation angle and geometric parameters, and complete decoupling of the Jones matrix. Among them:

[0127] 1. Phase mapping: Introduce PB contributions through the rotation angle to achieve complete decoupling of the Jones matrix. Such a correlation ensures that the rotation component of the phase modulation can correctly compensate the geometric phase component, thereby simultaneously achieving the overall phase distributions required for both polarization states.

[0128] 2. Parameter optimization process: The optimization process for each supercell position extracts the target phase from the neural network output, calculates the required rotation angle based on the phase difference to determine the adjusted geometric phase requirement, and finds the optimal nanocolumn size that minimizes the combined phase error of the two polarization states.

[0129] 3. Joint coding of rotation angle and geometric parameters: Integrate the rotation angle θ with the length L and width W of the nanocolumn into the parameter vector [L, W, θ], and optimize each unit position to simultaneously meet the phase requirements of RCP and LCP.

[0130] 4. Complete decoupling of the Jones matrix: Achieve complete control of the transmission / reflection coefficients by precisely controlling the geometric parameters and rotation angle of the nanocolumns, thereby independently regulating the two orthogonal polarization states and completing the polarization beam splitting function.

[0131] In this embodiment, the specific parameters (such as diameter, height, period, etc.) of the nanocolumns mapped in the metasurface beam splitter are determined by optimization, so as to achieve the best polarization beam splitting effect. By adopting this structural design, the device can achieve high-performance indicators that are difficult to reach by traditional optical elements while maintaining micro-nano dimensions.

[0132] Next, a finite-difference time-domain (FDTD) simulation verification example of the metasurface beam splitter provided in the embodiment of the present application is proposed.

[0133] When verifying the performance of the metasurface beam splitter provided by the embodiments of the present application through the FDTD simulation software, parameter scans are performed on nanocolumns of different scales to obtain the corresponding diagrams of RCP and LCP. According to the phase distribution obtained from the above U-net architecture, FDTD is used for mapping to construct the metasurface, and the physical model information of the metasurface is obtained. This model file is imported into FDTD to set the simulation area, and then conditions such as the light source and polarization state are set for simulation. The results show that the metasurface beam splitter provided by the embodiments of the present application has the characteristics of a high polarization extinction ratio (PER) (34.11 dB) and a high transmission efficiency (63.91%).

[0134] It should be noted that during the implementation of PER, the logarithm of the ratio of the intensity of the transmitted polarized light (desired polarization state) to the cross-polarized light (undesired polarization state) is calculated, that is: PER = 10log 10 (I_desired / I_undesired). And PER values of 36.63 dB, 28.11 dB, and 37.6 dB are measured at three focal positions respectively, with an average of 34.11 dB, far exceeding the 20 dB threshold required for polarization-sensitive applications. Thus, it is confirmed that the metasurface beam splitter provided by the embodiments of the present application achieves an extinction ratio of 34.11 dB, significantly superior to the related art.

[0135] In some embodiments, the metasurface beam splitter provided by the embodiments of the present application can be a three-beam splitter, corresponding to left-handed circularly polarized light (LCP), right-handed circularly polarized light (RCP), and 45° linearly polarized light respectively. This three-beam splitter successfully restores the required phase distribution through an improved U-Net architecture and an angular spectrum propagation model. And the performance of this three-beam splitter is verified through FDTD simulation, and its simulation results also show that this three-beam splitter has a high extinction ratio and a high transmission efficiency.

[0136] In some embodiments, the metasurface beam splitter provided by the embodiments of the present application can also be a polarization beam splitter with arbitrary beam directivity.

[0137] In some embodiments, based on the fact that the metasurface beam splitter provided by the embodiments of the present application can achieve arbitrary beam directivity and polarization control, the metasurface beam splitter provided by the embodiments of the present application can be applied to a variety of optical applications.

[0138] Exemplarily, the specific applications of the metasurface beam splitter provided by the embodiments of the present application can include:

[0139] a. Polarization imaging system: Utilize the characteristic of the high extinction ratio of the metasurface beam splitter to realize the acquisition of polarization information of biological tissues and enhance the contrast of medical images.

[0140] b. Optical communication multiplexing: High-bandwidth information transmission can be achieved through the property of the metasurface beam splitter that enables polarization-state-independent control, thereby increasing the communication capacity.

[0141] c. Quantum information processing: The metasurface beam splitter can provide a high-fidelity manipulation platform for qubits based on polarization encoding.

[0142] d. Augmented Reality (AR) display: Based on the metasurface beam splitter's ability to simultaneously control the beam directions of different polarization states, multi-level information presentation can be achieved.

[0143] e. Optical security encryption: Utilize the polarization-dependent diffraction pattern of the metasurface beam splitter as a security verification mechanism.

[0144] f. Advanced spectroscopic analyzer: Combine the polarization and spatial control capabilities of the metasurface beam splitter to achieve the separation and analysis of complex spectral components.

[0145] Next, please refer to Figure 8 , the embodiment of the present application also provides a design device for a polarization beam splitting metasurface. The design device for a polarization beam splitting metasurface can implement the above design method of the polarization beam splitting metasurface. The design device for a polarization beam splitting metasurface includes:

[0146] An acquisition module, configured to acquire a far-field intensity distribution image;

[0147] An image processing module, configured to perform feature extraction and reconstruction on the far-field intensity distribution image to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image;

[0148] An intensity-phase mapping module, configured to input the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover the metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model;

[0149] A metasurface design module, configured to perform the design of a polarization beam splitting metasurface based on the metasurface phase distribution.

[0150] In some embodiments, the intensity-phase mapping module is further configured to input the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map as model input data into a preset angular spectrum propagation model; perform far-field intensity prediction calculation on the basis of the angular spectrum propagation model using the model input data to obtain a predicted far-field intensity; and perform backpropagation optimization on the mean square error between the predicted far-field intensity and the input far-field intensity based on a preset intensity-phase gradient relationship to obtain a metasurface phase distribution corresponding to the far-field intensity distribution image; the input far-field intensity being the far-field intensity of the far-field intensity distribution image.

[0151] In some embodiments, the design device of the polarization beam splitting metasurface further includes:

[0152] a neural network training model, configured to convert a preset angular spectrum propagation model into a differentiable neural network layer and set the angular spectrum propagation model as a partial embedded training loop architecture to construct a deep learning framework driven by physical information;

[0153] wherein, the angular spectrum propagation model is a physical information driven layer in the deep learning framework; the deep learning framework is configured to perform neural network training using the far-field intensity distribution image as input data and output a metasurface phase distribution corresponding to the far-field intensity distribution image recovered based on the angular spectrum propagation model.

[0154] In some embodiments, the image processing module is further configured to input the far-field intensity distribution image into a preset convolutional neural network architecture; the convolutional neural network architecture includes a shared encoder and dual decoder branches; and perform feature extraction and reconstruction on the far-field intensity distribution image based on the shared encoder and the dual decoder branches to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0155] In some embodiments, the image processing module is further configured to perform feature extraction on the far-field intensity distribution image based on the shared encoder to obtain hierarchical features; and reconstruct the right-handed circular polarization phase distribution using the hierarchical features as input based on the first branch in the dual decoder branches to obtain the right-handed circular polarization phase distribution map of the far-field intensity distribution image, and reconstruct the left-handed circular polarization phase distribution using the hierarchical features as input based on the second branch in the dual decoder branches to obtain the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

[0156] In some embodiments, the metasurface design module is further configured to convert the metasurface phase distribution into nano-pillar structure parameters and perform nano-pillar superlattice design based on the nano-pillar structure parameters to construct a polarization beam splitting metasurface.

[0157] It should be noted that the specific implementation of the design device of the polarization beam splitting metasurface provided in the embodiments of the present application is basically the same as the specific embodiments of the above-mentioned design method of the polarization beam splitting metasurface, and will not be elaborated here.

[0158] The embodiments of the present application further provide an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above-mentioned design method of the polarization beam splitting metasurface is implemented. The electronic device can be any intelligent terminal device including a tablet computer, a personal computer, etc.

[0159] Please refer to Figure 9 , Figure 9 which schematically shows the hardware structure of the electronic device in some embodiments. The electronic device may include:

[0160] A processor 901, which can be implemented in the form of a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0161] A memory 902, which can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 902 can store an operating system and other application programs. When implementing the technical solutions provided in the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 902, and the processor 901 is used to call and execute the design method of the polarization beam splitting metasurface in the embodiments of the present application;

[0162] An input / output interface 903, which is used to implement information input and output;

[0163] A communication interface 904, which is used to implement communication interaction between this device and other devices, and can implement communication through a wired method (such as USB, network cable, etc.) or through a wireless method (such as mobile network, WIFI, Bluetooth, etc.);

[0164] A bus 905, which transmits information between various components of the device (such as the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);

[0165] Among them, the processor 901, the memory 902, the input / output interface 903, and the communication interface 904 are communicatively connected to each other inside the device through the bus 905.

[0166] An embodiment of the present application also provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the above-mentioned design method of the polarization beam splitting metasurface.

[0167] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above networks include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0168] An embodiment of the present application also provides a computer program product storing a computer program, which when executed by a processor implements the above-mentioned design method of the polarization beam splitting metasurface.

[0169] The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art will know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0170] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or combine certain steps, or different steps.

[0171] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0172] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices can be implemented as software, firmware, hardware, and appropriate combinations thereof.

[0173] In the description of this application and the above-mentioned accompanying drawings, terms such as "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that comprises a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0174] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects and indicates that three relationships can exist. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expression refers to any combination of these items, including any combination of single items (ones) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0175] In several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the above-mentioned division of units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be in electrical, mechanical, or other forms.

[0176] The units described above as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0177] In addition, in each embodiment of the present application, each functional unit can be integrated into one processing unit, or each unit can physically exist independently, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0178] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present application. The foregoing storage medium includes: various media that can store programs, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0179] The preferred embodiments of the embodiments of the present application have been described above with reference to the accompanying drawings. However, this does not limit the scope of the rights of the embodiments of the present application. Any modification, equivalent replacement, and improvement made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall fall within the scope of the rights of the embodiments of the present application.

Claims

1. A design method of a polarization beam splitting metasurface, characterized in that The method includes: Obtaining a far-field intensity distribution image; Performing feature extraction and reconstruction on the far-field intensity distribution image to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image; Inputting the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover a metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model; Designing a polarization beam splitting metasurface based on the metasurface phase distribution.

2. The method according to claim 1, characterized in that, The step of inputting the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to recover a metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model includes: Inputting the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map as model input data into a preset angular spectrum propagation model; Performing far-field intensity prediction calculation on the model input data based on the angular spectrum propagation model to obtain a predicted far-field intensity; Performing inverse optimization on the mean square error between the predicted far-field intensity and the input far-field intensity based on a preset intensity-phase gradient relationship to obtain a metasurface phase distribution corresponding to the far-field intensity distribution image; the input far-field intensity is the far-field intensity of the far-field intensity distribution image.

3. The method according to claim 1, wherein The method further includes: Converting the preset angular spectrum propagation model into a differentiable neural network layer and setting the angular spectrum propagation model as a partially embedded training loop architecture to construct a physics-informed deep learning framework; Wherein, the angular spectrum propagation model is a physics-informed layer in the deep learning framework; the deep learning framework is used to perform neural network training with the far-field intensity distribution image as input data and output a metasurface phase distribution corresponding to the far-field intensity distribution image recovered based on the angular spectrum propagation model.

4. The method according to claim 1, wherein The step of performing feature extraction and reconstruction on the far-field intensity distribution image to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image includes: Inputting the far-field intensity distribution image into a preset convolutional neural network architecture; the convolutional neural network architecture includes a shared encoder and dual decoder branches; Performing feature extraction and reconstruction on the far-field intensity distribution image based on the shared encoder and the dual decoder branches to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image.

5. The method according to claim 4, wherein The step of performing feature extraction and reconstruction on the far-field intensity distribution image based on the shared encoder and the dual decoder branches to obtain a right-handed circular polarization phase distribution map and a left-handed circular polarization phase distribution map of the far-field intensity distribution image includes: Performing feature extraction on the far-field intensity distribution image based on the shared encoder to obtain hierarchical features; Based on the first branch in the dual decoder branches, reconstruct the right-handed circular polarization phase distribution with the hierarchical features as the input to obtain the right-handed circular polarization phase distribution map of the far-field intensity distribution image, and based on the second branch in the dual decoder branches, reconstruct the left-handed circular polarization phase distribution with the hierarchical features as the input to obtain the left-handed circular polarization phase distribution map of the far-field intensity distribution image.

6. The method according to any one of claims 1 to 5, characterized in that, The polarization beam splitting metasurface design based on the metasurface phase distribution includes: Convert the metasurface phase distribution into nano-pillar structure parameters, and perform nano-pillar superlattice design based on the nano-pillar structure parameters to construct a polarization beam splitting metasurface.

7. A design device for a polarization beam splitting metasurface, characterized in that, The device includes: An acquisition module for acquiring a far-field intensity distribution image; An image processing module for performing feature extraction and reconstruction on the far-field intensity distribution image to obtain the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map of the far-field intensity distribution image; An intensity-phase mapping module for inputting the right-handed circular polarization phase distribution map and the left-handed circular polarization phase distribution map into a preset angular spectrum propagation model to restore the metasurface phase distribution corresponding to the far-field intensity distribution image based on the angular spectrum propagation model; A metasurface design module for performing polarization beam splitting metasurface design based on the metasurface phase distribution.

8. A metasurface beam splitter, characterized in that, The metasurface of the metasurface beam splitter is constructed based on the design method of the polarization beam splitting metasurface according to any one of claims 1 to 6.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the design method of the polarization beam splitting metasurface according to any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the design method of the polarization beam splitting metasurface according to any one of claims 1 to 6 is implemented.

Images