A millimeter wave hybrid metasurface aperture imaging method and device

Abstract

The application provides a millimeter wave mixed metasurface aperture imaging method and device, and belongs to the technical field of millimeter wave computing imaging and reconfigurable metasurface, and comprises the following steps: performing large field of view scanning in a first frequency band through a frequency diversity metasurface aperture, acquiring global echo data and generating a coarse imaging result; automatically determining a region of interest based on the coarse imaging result; generating a dynamic holographic coding pattern according to coordinate information of the region of interest, wherein the dynamic holographic coding pattern is a phase coding pattern or an amplitude and phase coding pattern, and loading the dynamic holographic coding pattern to a dynamic holographic tuning metasurface aperture; focusing on the region of interest in a second frequency band through the dynamic holographic tuning metasurface aperture, and acquiring local high-resolution echo data; and fusing and reconstructing the coarse imaging result and the local high-resolution echo data to generate a fusion image. The application can reduce the requirements for active channels and mechanical scanning, and improve the millimeter wave near-field imaging speed, resolution and engineering stability.

Description

Technical Field

[0001] This invention belongs to the field of millimeter-wave computational imaging and reconfigurable metasurface technology, and particularly relates to a millimeter-wave hybrid metasurface aperture imaging method and device. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] In the field of millimeter-wave imaging, traditional techniques employ methods such as frequency diversity metasurface aperture, dynamic metasurface antenna, programmable metasurface, and depth prior computation imaging.

[0004] Frequency diversity metasurfaces can generate a series of spatially incoherent or quasi-orthogonal radiation modes through frequency scanning, reducing the need for mechanical scanning and a large number of active channels. However, using frequency diversity metasurfaces alone usually requires a wide operating bandwidth and a relatively accurate measurement matrix, and the ability to resolve local details is limited by the operating frequency band, effective aperture, and mode orthogonality. Dynamic metasurface antennas can change the aperture field distribution through tuning units such as PIN diodes, varactors, MEMS, or liquid crystals to achieve electronic focusing, beam scanning, or coded illumination. However, while dynamic holographic metasurfaces alone can form focusing or complex illumination in local areas, directly covering large scenes requires a large number of state adjustments and scanning steps, resulting in high imaging time and control complexity. Mills-Cross sparse cross-layouts can synthesize a two-dimensional equivalent aperture with fewer linear array elements, but relying solely on sparse arrays is prone to sidelobes, aperture loss, and reconstructive ill-conditioning.

[0005] Therefore, existing technologies cannot simultaneously achieve rapid detection over a large area and high-resolution identification of local areas. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, this invention proposes a millimeter-wave hybrid metasurface aperture imaging method and device. By physically integrating and algorithmically coordinating static frequency diversity aperture and dynamic holographic tuning aperture, it solves the problems of high complexity and insufficient local resolution in large-area imaging.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: In a first aspect, the present invention discloses a millimeter-wave hybrid metasurface aperture imaging method, comprising: By using frequency diversity metasurface apertures to perform large field-of-view scanning in the first frequency band, global echo data is acquired and coarse imaging results are generated. The region of interest is automatically determined based on the coarse imaging results; A dynamic holographic coding pattern is generated based on the coordinate information of the region of interest. The dynamic holographic coding pattern is a phase coding pattern or an amplitude-phase coding pattern. The dynamic holographic coding pattern is then loaded onto the aperture of the dynamic holographic tuned metasurface. By focusing the region of interest in the second frequency band using the dynamic holographic tuned metasurface aperture, local high-resolution echo data can be obtained. The coarse imaging result and the local high-resolution echo data are fused and reconstructed to generate a fused image.

[0008] Secondly, the present invention discloses a millimeter-wave hybrid metasurface aperture imaging device, comprising: Hybrid metasurface apertures, including frequency diversity metasurface apertures and dynamic holographic metasurface apertures, wherein the frequency diversity metasurface apertures and the dynamic holographic metasurface apertures are physically integrated on the same dielectric substrate; A sparse transmit-receive array is connected to a hybrid metasurface aperture via electromagnetic coupling. The dynamic control circuit is electrically connected to the aperture of the dynamic holographic metasurface and is used to load the dynamic holographic encoded pattern; The millimeter-wave transceiver link is connected to a sparse transmit-receive array via a signal connection, and is used to generate millimeter-wave excitation signals and receive echo signals. The imaging processing module, connected to the millimeter-wave transceiver link and dynamic control circuit, is used to execute the millimeter-wave hybrid metasurface aperture imaging method described above.

[0009] Thirdly, the present invention discloses an electronic device, including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when run by the processor, complete the steps of the above-described millimeter-wave hybrid metasurface aperture imaging method.

[0010] Fourthly, the present invention discloses a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the above-described millimeter-wave hybrid metasurface aperture imaging method.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention first integrates broadband passive frequency diversity aperture and high-frequency dynamic holographic aperture, reducing the need for full-array phased array channels, phase shifters, and mechanical scanning structures. Then, it utilizes the frequency dimension, spatial dimension, and coded state dimension to create measurement diversity, improving the reconfigurability of low-channel-count systems. By using frequency-selective metasurface coarse imaging to automatically determine the region of interest (ROI), the holographic aperture ROI is encoded and can be supplemented with additional scans, reducing the number of full-field high-resolution scans and increasing speed. Finally, through PSF / transfer function libraries, mutual coupling calibration, and confidence fusion, ill-conditioned inverse problems and cross-band fusion errors are reduced. This allows for rapid detection of anomalies in large scenes first, followed by detailed identification of local areas, making it suitable for security inspection, non-destructive testing, and near-field high-resolution imaging.

[0012] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0013] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0014] Figure 1 This is a schematic diagram of the millimeter-wave hybrid metasurface aperture imaging structure described in Embodiment 1 of the present invention. Detailed Implementation

[0015] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0016] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0017] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0018] Example 1 In one or more embodiments, a millimeter-wave hybrid metasurface aperture imaging method is disclosed, employing, for example... Figure 1 The illustrated millimeter-wave hybrid metasurface aperture imaging structure, consisting of a frequency diversity metasurface aperture and a dynamic holographic metasurface aperture, is physically integrated on the same imaging front end. The frequency diversity metasurface aperture and the dynamic holographic metasurface aperture are integrated on the same dielectric substrate and operate in different frequency bands, with an electromagnetic isolation band between them. The specific steps include: Step S0: Establish the scene coordinate system, aperture coordinate system, and system calibration parameters, and determine the voxel mesh of the space to be imaged.

[0019] In one implementation, the scene coordinate system has its origin at the center of the space under test, with the x-axis and y-axis along the horizontal and vertical directions, respectively, and the z-axis along the depth direction. The aperture coordinate system has its origin at the center of the hybrid metasurface aperture. System calibration parameters include rotation, translation, and scaling parameters between the aperture coordinate system and the scene coordinate system, the mutual coupling matrix between the two apertures, and the amplitude and phase errors of each transceiver channel.

[0020] Furthermore, a system calibration process is established to obtain the transfer functions of frequency diversity metasurface apertures and dynamically holographic tuned metasurface apertures at different frequencies, encoding states, temperatures, or assembly positions. Calibration data includes, but is not limited to, three-dimensional point spread function, measurement matrix, mutual coupling matrix, channel amplitude and phase errors, frequency drift parameters, element state errors, and coordinate mapping parameters.

[0021] As one implementation method, the transfer function is obtained through system calibration. Specifically, a standard point target or standard reflector with known scattering intensity is placed in the imaging area, and its position is sequentially located at preset three-dimensional grid points; echo data is acquired at each frequency point, each holographic encoding state, and each set of transmit and receive channels, and the background field is subtracted.

[0022] For the m-th measurement state and the j-th spatial voxel, we can obtain:

[0023] in, Given the scattering coefficient of a standard target, For the measured echo, It is additive complex white Gaussian noise. Since the transfer function is used, and the white noise is negligible after subtracting the empty background, therefore:

[0024] All voxels, frequencies, encoding states, and transmit / receive channels The combination yields a three-dimensional transfer function library or measurement matrix. In practice, an initial transfer function can be obtained first using electromagnetic simulation, and then corrected using a small amount of standard target measurement data.

[0025] The mutual coupling matrix can be obtained through system self-calibration. Specifically, under targetless or standard load conditions, the scattering parameters Sij between each transmitting antenna, receiving antenna, frequency diversity aperture region, and dynamic holographic aperture region are measured to form the mutual coupling matrix C(f,c), where f represents the frequency and c represents the coding state.

[0026] In subsequent applications, the mutual coupling matrix will be introduced into the forward imaging model:

[0027] Where y is the measured echo, H is the ideal transmission matrix, and x is the scattering distribution of the target to be reconstructed. The reconstruction can be performed using:

[0028] in, These are the weighting coefficients; The penalty function applied to the target x to be reconstructed.

[0029] Pre-compensation can also be performed first:

[0030] Then, the compensated y' is used for imaging. This reduces the impact of the coupling between the two metasurface apertures and the crosstalk between antennas on the imaging results.

[0031] Element state error refers to the deviation between the actual amplitude and phase response of a dynamic holographic metasurface element under different control states and the design values. It can be obtained through near-field scanning, far-field pattern testing, or element / subarray S-parameter measurement.

[0032] For the i-th unit and the s-th state, it can be represented as:

[0033] in, This is the corrected actual response; ; Operating frequency; For ideal unit response; Let be the complex error factor, and let e(i,s,f) = a(i,s,f) • exp(jΔφ(i,s,f)), where a is the amplitude error and Δφ is the phase error. In subsequent generation of the holographic encoded pattern and calculation of the point spread function, the ideal unit response is no longer used; instead, the modified actual response is used. This allows the focus position, sidelobe level, and reconstruction model to more closely approximate the real hardware.

[0034] Step S1: Perform a large field-of-view scan in the first frequency band using a frequency diversity metasurface aperture. The frequency diversity metasurface aperture generates diverse illumination or reception modes that vary with frequency, acquires global echo data of the scene under test, and generates coarse imaging results.

[0035] Preferably, the frequency diversity metasurface apertures are arranged in a dynamic holographic metasurface aperture configuration that surrounds or partially surrounds the surface.

[0036] The frequency diversity metasurface aperture can be constructed using rectangular slots, complementary open rings, patch slots, H-shaped / cross-shaped / Jerusalem cross-shaped elements, or combinations thereof, with different sizes, orientations, coupling strengths, or aperiodic distributions. Each element has different resonant frequencies and phase responses within its operating bandwidth, resulting in a low-correlation spatial radiation mode with frequency.

[0037] Preferably, the frequency diversity aperture operates in the 24 GHz to 40 GHz, 26.5 GHz to 35.5 GHz, 57 GHz to 71 GHz, or other millimeter-wave broadband bands, with 32 to 512 frequency sampling points. The sampling points can be equally spaced or adaptively selected based on the channel matrix condition number, mode correlation, and signal-to-noise ratio. The cell layout of the frequency diversity aperture can be optimized by maximizing the mutual information of the measurement matrix, minimizing column correlation, controlling the aperture Q value, or constraining radiation efficiency.

[0038] To enhance engineering feasibility, the frequency diversity aperture can remain passive and be driven by a millimeter-wave source for frequency sweeping only; alternatively, a small number of switching states can be set to switch to different sub-aperture combinations before scanning at different frequencies, thereby increasing the number of modes without significantly increasing hardware complexity.

[0039] Preferably, the first frequency band is 26.5 GHz to 35.5 GHz.

[0040] In this embodiment, the hybrid aperture size is 12 cm × 12 cm. The outer region is a frequency diversity metasurface aperture, employing an aperiodic rectangular slot array, operating in the frequency band of 26.5 GHz to 35.5 GHz, sampling 101 frequency points. The central region is a dynamic holographic tuned metasurface aperture, operating at a frequency of 77 GHz, with a unit spacing no greater than half a wavelength, employing a 1-bit or 2-bit phase state. A 1 mm to 3 mm grounding via is provided between the two regions.

[0041] Coarse image reconstruction can employ matched filtering, least squares, sparse regularization, compressed sensing, wavenumber domain reconstruction, back projection, or depth unfolding networks, without specific limitations here.

[0042] Step S2: Based on the coarse imaging results, perform background suppression, threshold segmentation, clustering, connected component analysis, or learned target scoring to obtain suspected defects, metallic foreign objects, or abnormal reflection areas, thereby automatically determining the region of interest.

[0043] In this embodiment, the learning target score can be based on an existing machine learning classifier or can adopt the feature-weighted score set in this scheme.

[0044] Example calculation method:

[0045] in, Here, C(r) represents the coarse-image normalized reflectance intensity, and C(r) represents the local contrast. Let S be the anomaly probability output by the trained model, B(r) be the background similarity, and α, β, γ, and η be the weights. When Score(r) is greater than a preset threshold, or belongs to the top K connected regions with the highest scores, the region is identified as a Region of Interest (ROI).

[0046] If a learning model is not used, ROI can also be determined using only threshold segmentation, clustering, and connected component analysis. Therefore, this scheme includes both a learning-based scoring method and a non-learning-based implementation method.

[0047] Step S3: Generate a dynamic holographic coding pattern based on the coordinate information of the region of interest (ROI). The dynamic holographic coding pattern is a phase coding pattern or an amplitude-phase coding pattern. Then, load the dynamic holographic coding pattern onto the aperture of the dynamic holographic tuned metasurface.

[0048] First, a dynamic holographic coded pattern matching the ROI is selected or optimized from a pre-set coded pattern library and a 3D point spread function (PSF) / transfer function library. The pre-set coded pattern library selects / corrects the code based on the focal position, sidelobes, and signal-to-noise ratio (SNR), thereby reducing real-time computation and improving focusing quality.

[0049] Then, based on the formation of one or more focal points on the ROI by the dynamic holographic coding pattern, the coding pattern is loaded onto the aperture of the dynamic holographic tuned metasurface.

[0050] Step S4: Focus the region of interest in the second frequency band using the dynamic holographic tuned metasurface aperture through single-point, multi-point, linear focal, area focal, or randomly coded illumination to obtain local high-resolution echo data.

[0051] The dynamic holographic tunable metasurface aperture comprises multiple tunable units, each possessing at least two electromagnetic states. These electromagnetic states can be 1-bit phase states, 2-bit or multi-bit phase states, amplitude switching states, amplitude-phase combined states, or polarization switching states. Tuning devices may include PIN diodes, varactor diodes, RF MEMS, liquid crystals, phase change materials, graphene, or other controllable electromagnetic materials.

[0052] The holographic coding pattern can be solved based on the backpropagation field of the target focus as a virtual point source, Rayleigh-Sommerfeld diffraction integral, angular spectrum propagation, equivalent wavenumber domain transmission, or a full-wave simulation correction model. Based on the propagation phase from the target point to the aperture element, the feed reference wave phase, and the element discrete state constraints, the coding pattern that maximizes the field strength, minimizes the sidelobes, or achieves multi-focus energy balance at the region of interest (ROI) can be obtained.

[0053] Dynamic holographic apertures are preferably operated near 60 GHz, 77 GHz, 94 GHz, 110 GHz, or the W band to achieve higher angular and lateral resolution with a relatively small physical aperture. Their focusing modes may include on-axis focusing, off-axis focusing, multi-point focusing, line scanning, area scanning, and random speckle coding modes.

[0054] Preferably, the second frequency band is 70GHz to 85GHz.

[0055] The central holographic aperture loads a corresponding coded pattern based on the ROI's three-dimensional coordinates, forming one or more focal points on the ROI to acquire local echoes and perform high-resolution reconstruction. In this embodiment, the high-resolution reconstruction of local echoes can employ focused scanning imaging, multi-coded illumination inversion, sparse prior, total variational constraints, generative priors, or lightweight convolutional networks; no specific limitations are specified here.

[0056] In the coarse imaging reconstruction and high-resolution reconstruction of local echoes described above, the learning module is used to reduce computational load and improve noise resistance. Its input and output are constrained by the physical transmission model to avoid relying solely on the black-box network.

[0057] Step S5: The coarse imaging result and the local high-resolution echo data are fused and reconstructed. Combined with mutual coupling calibration, channel error compensation and confidence fusion, a fused image with both large scene coverage and local high-resolution details is generated.

[0058] The system performs coordinate registration, frequency band normalization, confidence estimation, and fusion reconstruction on coarse imaging results, local high-resolution echo data, and calibration compensation data, outputting a three-dimensional or two-dimensional image with both wide field of view coverage and local high-resolution details.

[0059] First, the coarse imaging results and local high-resolution echo data are unified into the same coordinate system, and pixel-level or voxel-level weights are generated based on resolution, signal-to-noise ratio, sidelobe level, and calibration reliability, thereby suppressing artifacts and boundary abrupt changes. Specifically: Confidence levels are calculated for both the coarse imaging result and the high-resolution holographic result. Taking a specific pixel or voxel r as an example, the holographic image... The weights can be defined as:

[0060] coarse imaging image The weights can be defined as:

[0061] Where SNR is the signal-to-noise ratio, Res is the resolution evaluation value, Cal is the calibration confidence level, PSL is the sidelobe level, and w1, w2, w3, and w4 are weighting coefficients, which can be set to the same value of 0.25 to average the effect of each part. Mutual coupling calibration and channel error compensation are reflected in Cal: if the mutual coupling compensation error or channel amplitude and phase error corresponding to a certain region is large, the confidence level of that region is reduced.

[0062] The fusion weights are obtained after normalization:

[0063]

[0064] in, Weights for holographic images; Weights for coarse imaging images; Weights for holographic image fusion; The weights for coarse imaging image fusion.

[0065] Then, based on the weighted fusion of coarse imaging results and local high-resolution echo data, holographic high-resolution results are prioritized for the interior of the ROI; for the ROI boundary or low-confidence region, weighted fusion or multi-scale fusion is used to avoid edge discontinuity caused by simple pixel replacement.

[0066] In this embodiment, the weighted fusion can be written as:

[0067] in, It is a holographic high-resolution image. It is a coarse imaging image. These are the fusion weights of the holographic images; This is the merged image.

[0068] Multi-scale fusion can be achieved using pyramid decomposition. First, the coarse imaging image and the holographic image are decomposed into low-frequency and high-frequency layers, respectively:

[0069]

[0070] in, The high-frequency component of a holographic image; The low-frequency component of the holographic image; This refers to the low-frequency portion of the coarse imaging image. This refers to the high-frequency components of the coarse-grained image. Prioritize using the high-frequency components within the ROI. Preserve high-resolution details; at ROI boundaries and Weighted transition, preferably with both weights being 0.5, avoids abrupt changes in the image.

[0071] Furthermore, an adaptive coding and scanning mechanism is adopted: when the ROI confidence is insufficient, the sidelobes are too high, or the local reconstruction residual exceeds the threshold, the system automatically adjusts the holographic coding pattern or adds a small amount of frequency / state sampling to form a closed-loop adaptive scanning. Specifically: When the peak-to-sidelobe ratio of the local reconstruction result is lower than the preset threshold, the ROI boundary is discontinuous, the reconstruction residual is higher than the threshold, or the target category confidence is insufficient, the system does not re-execute the full large-scene scan, but instead selects a supplementary coding pattern related to the current error direction from the coding pattern library.

[0072] In this embodiment, a supplementary coding pattern related to the current error direction is determined by error sensitivity or correlation coefficient. Specifically, this involves the predicted response to the current reconstructed residual e and the candidate supplementary coding pattern. Calculate the relevance:

[0073] Select Relevance The largest or exceeding a preset threshold encoding pattern is used as a supplementary encoding pattern. For example, if the residual mainly varies along the depth direction, multifocal encoding that is sensitive to depth is selected; if the residual is concentrated at the lateral edge, linear focal encoding is selected; if there are suspected multiple scattering centers, random speckle encoding is selected.

[0074] For example, one could choose multifocal coding which is sensitive to the target depth direction, linear coding which is sensitive to the lateral edges, or random speckle coding which is sensitive to the number of scattering centers.

[0075] In this embodiment, the rescanning process forms a closed loop of "coarse imaging - ROI judgment - holographic enhancement - quality evaluation - a small number of rescanning", which can reduce the total number of samplings while ensuring imaging quality.

[0076] This invention physically integrates a static frequency diversity metasurface aperture and a dynamic holographic tuned metasurface aperture into the same imaging front end. The system first uses the frequency diversity aperture to perform a large field-of-view rapid scan in a broadband frequency band to obtain a coarse imaging result of the scene. Then, based on the coarse imaging result, it automatically determines the region of interest (ROI) and drives the dynamic holographic aperture to focus on the ROI in the high frequency band to obtain local high-resolution echo data. Finally, it generates a fused image with both large scene coverage and local high-resolution details through confidence fusion, realizing the collaborative work of wide-field rapid detection and local high-resolution focusing.

[0077] Example 2 In one or more embodiments, a millimeter-wave hybrid metasurface aperture imaging device is disclosed, comprising: The hybrid metasurface aperture includes a frequency diversity metasurface aperture and a dynamic holographic metasurface aperture. The frequency diversity metasurface aperture and the dynamic holographic metasurface aperture are physically integrated on the same dielectric substrate, with an electromagnetic isolation band between them. The frequency diversity metasurface aperture is used to generate diverse illumination or reception modes that vary with frequency in a first frequency band. The dynamic holographic metasurface aperture is used to load phase-encoded patterns or amplitude-phase-encoded patterns according to ROI coordinates in a second frequency band to form single-point, multi-point, or extended region focusing.

[0078] The sparse transmit-receive array is connected to the hybrid metasurface aperture via electromagnetic coupling.

[0079] The dynamic control circuit is electrically connected to the aperture of the dynamic holographic metasurface and is used to load dynamic holographic encoded patterns.

[0080] The millimeter-wave transceiver link is connected to a sparse transmit-receive array via a signal connection to generate millimeter-wave excitation signals and receive echo signals.

[0081] The imaging processing module, connected to the millimeter-wave transceiver link and the dynamic control circuit, is used to execute the imaging method described in Embodiment 1.

[0082] In this embodiment, the frequency diversity metasurface aperture and the dynamic holographic metasurface aperture can be located on the same dielectric substrate, the same metal ground layer, and the same mounting reference surface.

[0083] Preferably, the first aperture region surrounds or semi-surrounds the second aperture region, forming a layout of wide-field detection in the outer ring and local focusing in the center; or the two are arranged side by side, one above the other, in a checkerboard pattern, or in alternating blocks. An electromagnetic isolation strip, a wave-absorbing / grounding via fence, or a non-resonant buffer unit is set between the two apertures to reduce cross-aperture mutual coupling.

[0084] In this embodiment, the antenna of the sparse transmit-receive array preferably adopts a Mills-Cross or equivalent cross layout. The transmit units are arranged along a first direction, and the receive units are arranged along a second direction, with the first and second directions perpendicular to each other. A two-dimensional equivalent aperture is synthesized from a one-dimensional transmit aperture and a one-dimensional receive aperture. This layout, combined with the frequency diversity mode and dynamic focusing mode of the metasurface, forms multi-dimensional sampling, including frequency dimension, spatial dimension, coded state dimension, and polarization dimension.

[0085] The frequency diversity metasurface aperture region and the dynamic holographic metasurface aperture region are integrated side-by-side on the same substrate. The transmitting antenna is arranged along the upper and lower edges of the aperture, and the receiving antenna is arranged along the left and right edges of the aperture, forming a Mills-Cross sparse cross layout. The transmitting and receiving antennas are arranged parallel to the metasurface aperture, with a spacing of 5 mm to 15 mm. This structure can synthesize a two-dimensional equivalent sampling aperture without increasing the number of channels in a two-dimensional full array.

[0086] When constructing the measurement matrix, the imaging processing module uses the transmission position, reception position, frequency diversity mode, and holographic coding state as a joint index. For coarse imaging, a small number of Tx / Rx combinations are used to quickly complete scene localization. In this embodiment, the actual number of transmit / receive combinations activated does not exceed 10% to 50% of the full array combination count, preferably 20% to 30%. For local fine imaging, only Tx / Rx combinations that contribute significantly to the ROI direction and a small number of high-information-content coding states are activated. That is, the sensitivity of this combination to the ROI is greater than 50% of the maximum sensitivity, or it is in the top 30% of the sensitivity ranking. In addition, the number of supplementary coding states is 5% to 30% of the initial number of coding states, or 1 to 20 coding states are added with each supplementary scan.

[0087] This device achieves compact integration at the hardware level and coordinated operation at the functional level by physically integrating the frequency diversity metasurface aperture region and the dynamic holographic metasurface aperture region on the same dielectric substrate, in conjunction with a sparse transmit / receive array and an imaging processing module.

[0088] This invention utilizes a static frequency diversity metasurface to achieve rapid detection of a large field of view, and then uses a dynamic holographic metasurface to perform high-resolution focusing on the automatically determined ROI. Finally, a global-local collaborative image is obtained through calibration model and confidence fusion.

[0089] This invention can be used in scenarios such as security inspection, non-destructive testing, packaging defect detection, material electromagnetic property measurement, near-field three-dimensional radar imaging, and medical auxiliary testing.

[0090] Example 3 This embodiment provides an electronic device, including a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, they complete the steps of the above-described millimeter-wave hybrid metasurface aperture imaging method.

[0091] Example 4 This embodiment provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the above-described millimeter-wave hybrid metasurface aperture imaging method.

[0092] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0093] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0094] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment, whereby a series of operational steps are performed to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0095] The descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0096] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A millimeter-wave hybrid metasurface aperture imaging method, characterized in that, include: By using frequency diversity metasurface apertures to perform large field-of-view scanning in the first frequency band, global echo data is acquired and coarse imaging results are generated. The region of interest is automatically determined based on the coarse imaging results; A dynamic holographic coding pattern is generated based on the coordinate information of the region of interest. The dynamic holographic coding pattern is a phase coding pattern or an amplitude-phase coding pattern. The dynamic holographic coding pattern is then loaded onto the aperture of the dynamic holographic tuned metasurface. By focusing the region of interest in the second frequency band using the dynamic holographic tuned metasurface aperture, local high-resolution echo data can be obtained. The coarse imaging result and the local high-resolution echo data are fused and reconstructed to generate a fused image.

2. The millimeter-wave hybrid metasurface aperture imaging method as described in claim 1, characterized in that, A dynamic holographic encoded pattern is generated based on the coordinate information of the region of interest, specifically as follows: The system selects or optimizes dynamic holographic coded patterns that match the region of interest from a preset coded pattern library and a 3D point spread function / transfer function library; wherein the preset coded pattern library selects or modifies the coding based on the focal position, side lobes and SNR.

3. The millimeter-wave hybrid metasurface aperture imaging method as described in claim 1, characterized in that, The coarse imaging result and the local high-resolution echo data are fused and reconstructed to generate a fused image, specifically as follows: The coarse imaging results and local high-resolution echo data are unified into the same coordinate system, and pixel-level or voxel-level weights are generated based on resolution, signal-to-noise ratio, sidelobe level and calibration reliability. Based on the weighted fusion of coarse imaging results and local high-resolution echo data, holographic high-resolution results are used for the interior of the region of interest; weighted fusion or multi-scale fusion is used for the boundary of the region of interest.

4. The millimeter-wave hybrid metasurface aperture imaging method as described in claim 3, characterized in that, The process of generating pixel-level or voxel-level weights based on resolution, signal-to-noise ratio, sidelobe level, and calibration confidence is as follows: Holographic images The weights are: coarse imaging image The weights are: Wherein, SNR is the signal-to-noise ratio, Res is the resolution evaluation value, Cal is the calibration confidence level, PSL is the sidelobe level, and w1, w2, w3, and w4 are weighting coefficients.

5. The millimeter-wave hybrid metasurface aperture imaging method as described in claim 3, characterized in that, The multi-scale fusion specifically refers to: The coarse imaging image and the holographic image are decomposed into low-frequency layer and high-frequency layer, respectively: in, The high-frequency component of a holographic image; The low-frequency component of the holographic image; This refers to the low-frequency portion of the coarse imaging image. This refers to the high-frequency portion of the coarse imaging image; High-resolution details are preserved in the high-frequency portion of the holographic image within the region of interest. The low-frequency components of the holographic image and the low-frequency components of the coarse imaging image are weighted and fused at the boundary of the region of interest.

6. The millimeter-wave hybrid metasurface aperture imaging method as described in claim 1, characterized in that, An adaptive coding and scanning mechanism is adopted: when the confidence level of the region of interest is insufficient, the sidelobes are too high, or the local reconstruction residual exceeds the threshold, the holographic coding pattern is adjusted or a small amount of frequency / state sampling is added to form a closed-loop adaptive scanning. Specifically: When the peak-to-sidelobe ratio of the local reconstruction result is lower than the preset threshold, the ROI boundary is discontinuous, the reconstruction residual is higher than the threshold, or the confidence of the target class is insufficient, a supplementary coding pattern related to the current error direction is selected from the coding pattern library.

7. The millimeter-wave hybrid metasurface aperture imaging method as described in claim 6, characterized in that, Determine the supplementary coding pattern related to the current error direction using error sensitivity or correlation coefficient: Predicted response to the current reconstructed residual e and candidate supplementary coding pattern Calculate the relevance: Select the coding pattern with the highest relevance as the supplementary coding pattern.

8. A millimeter-wave hybrid metasurface aperture imaging device, characterized in that, include: Hybrid metasurface apertures, including frequency diversity metasurface apertures and dynamic holographic metasurface apertures, wherein the frequency diversity metasurface apertures and the dynamic holographic metasurface apertures are physically integrated on the same dielectric substrate; A sparse transmit-receive array is connected to a hybrid metasurface aperture via electromagnetic coupling. The dynamic control circuit is electrically connected to the aperture of the dynamic holographic metasurface and is used to load the dynamic holographic encoded pattern; The millimeter-wave transceiver link is connected to a sparse transmit-receive array via a signal connection, and is used to generate millimeter-wave excitation signals and receive echo signals. An imaging processing module, connected to a millimeter-wave transceiver link and a dynamic control circuit, is used to execute the millimeter-wave hybrid metasurface aperture imaging method according to any one of claims 1-7.

9. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the millimeter-wave hybrid metasurface aperture imaging method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, complete the millimeter-wave hybrid metasurface aperture imaging method according to any one of claims 1-7.

Images