A microscopic imaging system and method based on two-stage joint modulation

Abstract

The application discloses a microscopic imaging system and method based on two-stage joint modulation, and relates to the technical field of microscopic imaging. The system comprises a light source assembly, a first-stage light field modulation assembly, a main beam splitter, a microscopic amplification and conjugate adapter optical path assembly, a second-stage dynamic conjugate spatial filter assembly, a camera detection assembly, and a signal control and calculation reconstruction assembly. The signal control and calculation reconstruction assembly controls the cut-in or cut-out state of the first-stage light field modulation assembly and the second-stage dynamic conjugate spatial filter assembly according to a target imaging mode, realizes flexible switching of multiple imaging modes, and reconstructs a super-resolution microscopic image of a sample to be measured based on an acquired digital image. Through the two-stage joint modulation architecture, the application pre-inhibits defocus background, improves the signal-to-noise ratio of an original image, realizes flexible switching of multiple modes, and can obtain a high-contrast super-resolution microscopic image in a thick sample and a strong scattering environment.

Description

Technical Field

[0001] This application relates to the field of microscopic imaging technology, and in particular to a microscopic imaging system and method based on dual-level joint modulation. Background Technology

[0002] Microscopic imaging technology is an important tool for obtaining microscopic structural information in life sciences, medical diagnostics, and other fields. With the increasing demand for imaging the dynamic processes of organelles and the three-dimensional structure of thick tissue samples, microscopic imaging systems not only need higher spatial resolution but also need to maintain a high signal-to-noise ratio and good background suppression capabilities under complex sample conditions. Especially in scenarios with thick samples, strongly scattering samples, or significant defocused fluorescence backgrounds, traditional wide-field microscopy, due to its lack of effective optical tomography, easily introduces a large amount of defocused stray signals, leading to decreased image contrast, details being obscured by the background, and further affecting the stability and imaging quality of subsequent computational reconstruction.

[0003] To improve resolution and imaging quality, various structured illumination microscopy methods have been developed, such as Structured Illumination Microscopy (SIM) and Image Scanning Microscopy (ISM). These techniques typically modulate high-frequency information in the sample to a detectable frequency band by introducing fringes, speckles, or other coded light fields into the excitation optical path, and then combine this with computational reconstruction methods to achieve resolution enhancement or specific optical contrast enhancement. While these methods have achieved good results under thin sample and low background conditions, they still primarily rely on excitation-end encoding and back-end algorithm reconstruction, and their physical suppression of defocused background light on the emission side is relatively limited. When sample thickness increases, background enhancement or scattering intensifies, the original image received by the detector is often subject to strong defocused background interference, leading to a decrease in the signal-to-noise ratio and an increase in artifacts in the reconstruction results.

[0004] On the other hand, techniques such as confocal microscopy and rotating confocal microscopy selectively transmit signals within the focal plane by incorporating pinholes, slits, or other spatial filtering structures in the emission optical path, thereby achieving defocus background suppression at the physical level and exhibiting good optical tomography capabilities. However, traditional confocal or rotating systems are typically based on fixed pinhole arrays or fixed scanning configurations, and their detection-side spatial filtering structures are tightly coupled with the front-end excitation method, limiting system flexibility and making it difficult to simultaneously accommodate multiple structured excitation modes.

[0005] Therefore, traditional microscopic imaging techniques generally suffer from the following shortcomings: First, front-end structured excitation and back-end emission spatial filtering are often designed as independent modules, lacking a unified coordination mechanism; second, although excitation-end encoding can provide super-resolution capabilities, the out-of-focus background will be detected, thus limiting the final reconstruction quality; third, traditional spatial filtering components have limited compatibility with different imaging modalities, which is not conducive to the flexible switching of multiple modes such as lattice, stripe, speckle, and wide field on the same platform; fourth, the lack of a unified physical description between hardware optical path design and computational reconstruction model makes it difficult to further improve the overall system performance.

[0006] Therefore, there is an urgent need to propose a new microscopic imaging system and method that enables the co-design of excitation-end optical field encoding and emission-end spatial filtering. While maintaining compatibility with various structured illuminations, it can physically suppress defocused backgrounds and combine a unified temporal control and computational reconstruction model to achieve high signal-to-noise ratio and super-resolution microscopic imaging under thick sample and strong background conditions. Summary of the Invention

[0007] The purpose of this application is to provide a microscopic imaging system and method based on dual-level joint modulation, which can realize the physical coordination of structured illumination at the excitation end and spatial filtering at the emission end, suppress defocus background signals before photoelectric conversion, improve the signal-to-noise ratio of the original image, support flexible switching of multiple modes, and obtain high-contrast super-resolution microscopic images in thick sample and strong background environments by combining physical model-driven computational reconstruction.

[0008] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a microscopic imaging system based on dual-level joint modulation, comprising: a light source component, a first-level light field modulation component, a master beam splitter, a microscopic magnification and conjugate transition optical path component, a second-level dynamic conjugate spatial filtering component, a camera detection component, and a signal control and computational reconstruction component.

[0009] The master beam splitter is positioned on the outgoing light path of the light source assembly; the first-stage optical field modulation assembly is positioned between the light source assembly and the master beam splitter; the microscopic magnification and conjugate transition optical path assembly is positioned on the reflected light path of the master beam splitter; the second-stage dynamic conjugate spatial filter assembly is positioned between the master beam splitter and the camera detection assembly, and is optically conjugate with the plane containing the sample under test; the camera detection assembly is positioned on the transmitted light path of the master beam splitter; and the signal control and computational reconstruction assembly is connected to the light source assembly, the first-stage optical field modulation assembly, the second-stage dynamic conjugate spatial filter assembly, and the camera detection assembly, respectively.

[0010] The light source assembly is used to output the excitation beam.

[0011] The first-stage optical field modulation component is used to modulate the excitation beam into a structured light illumination field when the received first state command is a cut-in command.

[0012] The main beam splitter is used to reflect the received incident light signal to the micro-magnification and conjugate transition optical path assembly; the incident light signal includes an excitation beam or a structured light illumination field.

[0013] The micro-magnification and conjugate transition optical path assembly is used to focus the incident light signal onto the plane where the sample under test is located, and to collect the emitted light signal excited by the sample under test and return it to the main beam splitter.

[0014] The second-stage dynamic conjugate spatial filter component is used to physically defocus and purify the emitted light signal that will be transmitted through the main beam splitter when the received second state command is a cut-in command.

[0015] The camera detection component converts the received detection light signal into a digital image and sends it to the signal control and calculation reconstruction component; the detection light signal includes the emitted light signal transmitted through the main beam splitter or the emitted light signal purified by physical defocusing.

[0016] The signal control and computational reconstruction component is used to determine a first state command and a second state command based on the target imaging modality; it is also used to obtain a super-resolution microscopic image of the sample to be tested based on the target imaging modality and the digital image; both the first state command and the second state command include an entry command or an exit command.

[0017] Secondly, this application provides a microscopic imaging method based on dual-level joint modulation, comprising: The target imaging mode is acquired, and a first state command and a second state command are determined based on the target imaging mode; the target imaging mode includes a structured light illumination super-resolution mode, a wide-field imaging mode, a rotating disk confocal mode, or a two-level joint modulation enhanced structured light illumination mode.

[0018] Based on the first state command, the incident light signal is controlled to irradiate the sample under test, so as to excite the sample under test to generate an emitted light signal; the incident light signal includes an excitation beam or a structured light illumination field.

[0019] Based on the second state command and the emitted light signal, the detection light signal of the sample to be tested is determined; the detection light signal includes the emitted light signal or the emitted light signal purified by physical defocusing.

[0020] A digital image of the sample is determined based on the probe light signal of the sample.

[0021] If the target imaging mode is a wide-field imaging mode or a rotating disk confocal mode, then the digital image is used as a microscopic image of the sample to be tested.

[0022] If the target imaging mode is a structured light illumination super-resolution mode, then the super-resolution microscopic image of the sample to be tested is determined by the frequency domain decoding matrix equation constructed based on the digital image.

[0023] If the target imaging mode is a dual-level joint modulation enhanced structured light illumination mode, then the super-resolution microscopic image of the sample to be tested is determined based on the digital image and the forward imaging model.

[0024] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a microscopic imaging system and method based on dual-level joint modulation. By constructing a "structured excitation-conjugate filtering" dual-level joint modulation architecture through a first-level optical field modulation component located in the excitation optical path and a second-level dynamic conjugate spatial filter component located in the emission optical path and optically conjugate with the sample plane, a "structured excitation-conjugate filtering" dual-level joint modulation architecture is established. This solves the problem of independent design of excitation-end encoding and emission-end filtering in traditional microscopic imaging techniques, lacking a physical coordination mechanism. It achieves physical pre-suppression of defocused background signals before photoelectric conversion, significantly improving the signal-to-noise ratio of the original image. Through a signal control and computational reconstruction component, a first state command and a second state command are determined according to the target imaging mode, controlling the entry or exit states of the first-level optical field modulation component and the second-level dynamic conjugate spatial filter component, respectively. This solves the problem of poor multi-modal compatibility and difficulty in flexibly switching imaging modes in traditional systems, achieving millisecond-level switching of wide-field imaging mode, turntable confocal mode, structured light illumination super-resolution mode, and dual-level joint modulation enhanced structured light illumination mode within the same system. The super-resolution microscopic image of the sample under test is obtained by signal control and computational reconstruction components based on the target imaging modality and digital image. Combined with the physical model-driven computational reconstruction method, the problem of lack of unified physical description between hardware optical path design and computational reconstruction model is solved. Stable reconstruction from high signal-to-noise ratio original image to super-resolution microscopic image is achieved, and high contrast and super-resolution microscopic imaging effect is obtained in thick sample and strong scattering environment. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic diagram illustrating the structure and interconnections of a microscopic imaging system based on dual-level joint modulation, provided as an embodiment of this application.

[0027] Figure 2This is a schematic diagram of the structure of an optical field coding component and a dynamic spatial filtering mask provided in an embodiment of this application; wherein, Figure 2 (a) is a schematic diagram of the partitioned mask turntable for three-beam interference structured light illumination; Figure 2 (b) is a schematic diagram of the partitioned mask turntable for two-beam interference structured light illumination; Figure 2 (c) is a schematic diagram of an Archimedes spiral pinhole array turntable; Figure 2 (d) is a schematic diagram of a periodic light-transmitting slit turntable.

[0028] Figure 3 Schematic diagrams of various architectures of a first-stage optical field modulation component provided in an embodiment of this application; wherein, Figure 3 (a) is a schematic diagram of a non-coaxial architecture based on off-axis oblique incidence; Figure 3 (b) is a schematic diagram of a vertical coaxial architecture based on polarization state modulation; Figure 3 (c) is a schematic diagram of a cascaded scanning configuration based on a static array and pupil plane deflection; Figure 3 (d) is a schematic diagram of the interference configuration based on diffraction coding and frequency domain filtering.

[0029] Figure 4 This is a schematic diagram of a multifocal lighting pattern generation strategy provided in an embodiment of this application; wherein, Figure 4 (a) is a schematic diagram of the microlens array structure; Figure 4 (b) is an image of the multifocal illumination pattern generated by the microlens array; Figure 4 (c) is a schematic diagram of the Damman grating structure; Figure 4 (d) is a multifocal lighting pattern effect generated by the Daman grating.

[0030] Figure 5 This is a schematic diagram illustrating other lighting pattern generation strategies provided in one embodiment of this application; wherein, Figure 5 (a) is a schematic diagram of the speckle phase plate structure; Figure 5 (b) is a diagram of the dynamic speckle illumination field generated by the speckle phase plate; Figure 5 (c) is a schematic diagram of the phase mask used to generate the Bessel beam; Figure 5 (d) is a diagram of the illumination field effect of the Bessel beam.

[0031] Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application.

[0032] Figure reference numerals: 10-Light source assembly; 11-Excitation beam; 12-Beam shaping assembly; 13-Outgoing parallel beam; 14-Illumination beam; 20-First-stage optical field modulation assembly; 21-Pre-connector assembly; 211-Polarizer; 212-Polarization beam splitter; 213-Waveplate assembly; 22-Optical field encoding assembly; 23-First-stage drive assembly; 24-First relay lens; 25-Second relay lens; 26-Scanning device; 27-Spatial frequency domain mask; 30-Structured illumination field; 31-Plane containing the sample under test; 3 2-Detector plane; 40-Main beam splitter; 50-Second-stage dynamic conjugate spatial filter assembly; 51-Dynamic spatial filter mask; 52-Mask driving assembly; 60-Microscopic magnification and conjugate transition optical path assembly; 61-Eye tube lens; 62-Microscopic objective lens; 70-Sample assembly; 71-Sample scanning assembly; 72-Lateral drive assembly; 73-Axial drive assembly; 80-Conjugate transition assembly; 81-Third relay lens; 82-Fourth relay lens; 90-Camera detection assembly; 100-Signal control and computational reconstruction assembly. Detailed Implementation

[0033] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0034] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0035] In one exemplary embodiment, such as Figure 1 As shown, a microscopic imaging system based on dual-level joint modulation is provided, comprising: The light source assembly 10, the first-stage light field modulation assembly 20, the main beam splitter 40, the microscopic magnification and conjugate transition optical path assembly 60, the second-stage dynamic conjugate spatial filter assembly 50, the camera detection assembly 90, and the signal control and computational reconstruction assembly 100.

[0036] The main beam splitter 40 is disposed on the outgoing light path of the light source assembly 10; the first-stage light field modulation assembly 20 is disposed between the light source assembly 10 and the main beam splitter 40; the microscopic magnification and conjugate transition optical path assembly 60 is disposed on the reflected light path of the main beam splitter 40; the second-stage dynamic conjugate spatial filter assembly 50 is disposed between the main beam splitter 40 and the camera detection assembly 90, and is optically conjugate with the plane where the sample to be measured is located; the camera detection assembly 90 is disposed on the transmitted light path of the main beam splitter 40; the signal control and calculation reconstruction assembly 100 is connected to the light source assembly 10, the first-stage light field modulation assembly 20, the second-stage dynamic conjugate spatial filter assembly 50 and the camera detection assembly 90 respectively.

[0037] The light source assembly 10 is used to output the excitation beam 11.

[0038] The first-stage optical field modulation component 20 is used to modulate the excitation beam 11 into a structured light illumination field when the received first state command is a cut-in command.

[0039] The main beam splitter 40 is used to reflect the received incident light signal to the micro-magnification and conjugate transition optical path assembly 60; the incident light signal includes the excitation beam 11 or the structured light illumination field.

[0040] The microscopic magnification and conjugate transition optical path assembly 60 is used to focus the incident light signal onto the plane where the sample under test is located, and to collect the emitted light signal excited by the sample under test and return it to the main beam splitter 40.

[0041] The second-stage dynamic conjugate spatial filter component 50 is used to physically defocus and purify the emitted light signal transmitted through the main beam splitter 40 when the received second state command is a cut-in command.

[0042] The camera detection component 90 is used to convert the received detection light signal into a digital image and send it to the signal control and calculation reconstruction component 100; the detection light signal includes the emitted light signal transmitted through the main beam splitter 40 or the emitted light signal purified after physical defocusing.

[0043] The signal control and computational reconstruction component 100 is used to determine a first state command and a second state command based on the target imaging modality; it is also used to obtain a super-resolution microscopic image of the sample to be tested based on the target imaging modality and the digital image; both the first state command and the second state command include an entry command or an exit command.

[0044] The signal control and computational reconstruction component 100 controls the first-stage optical field modulation component 20 and the second-stage dynamic conjugate spatial filter component 50 to work together to form a two-stage joint modulation according to the first state command and the second state command. Two-stage joint modulation refers to the signal control and computational reconstruction component 100 uniformly controlling the entry or exit states of the first-stage optical field modulation component 20 and the second-stage dynamic conjugate spatial filter component 50, so that the structured illumination at the excitation end and the spatial filtering at the emission end work together under optical conjugate relations, achieving a combined effect of enhanced transmission of in-focus signals and physical blocking of out-of-focus background signals. The first state command and the second state command independently control the entry or exit of the two stages, which can be flexibly configured according to the target imaging mode to achieve various imaging modes such as wide-field imaging, turntable confocalization, structured illumination super-resolution, and two-stage joint modulation enhanced structured illumination.

[0045] In another exemplary embodiment of this application, the first-stage optical field modulation component 20 includes: a pre-transfer component 21, an optical field encoding component 22, a first-stage driving component 23, and a 4f optical component.

[0046] The pre-transfer assembly 21 is disposed on the outgoing optical path between the light source assembly 10 and the main beam splitter 40, and is used to adjust the incident state of the excitation beam 11 incident on the optical field encoding assembly 22; the incident state includes at least one of incident angle or polarization state.

[0047] The optical field encoding component 22 is disposed on the outgoing optical path of the front-end transition component 21 and is used to perform structured modulation on the incident excitation beam 11 when the first state command is a cut-in command to generate an initial modulated optical field; the structured modulation includes at least one of amplitude modulation, phase modulation, spatial frequency modulation and polarization state modulation.

[0048] The first-stage driving component 23 is connected to the light field encoding component 22 and is used to drive the light field encoding component 22 to cut into or out of the outgoing light path between the light source component 10 and the main beam splitter 40 according to the first state command. When the first state command is the cut-in command, the light field encoding component 22 is driven to perform a spatial evolution operation on the initial modulated light field to obtain a structured illumination field 30. The spatial evolution operation includes at least one of spatial scanning, phase shifting, or pattern replacement.

[0049] The 4f optical component is disposed on the outgoing light path of the light field encoding component 22 and is used to transmit the incident light signal to the main beam splitter 40 in a proportionally conjugate manner.

[0050] As an optional implementation, when the first state command is a cut-in command, the signal control and calculation reconstruction component 100 controls the initial modulation light field pattern refresh of the first-level light field modulation component 20 to maintain nanosecond-level synchronization with the exposure of the camera detection component 90.

[0051] In another exemplary embodiment of this application, the second-level dynamic conjugate spatial filtering component 50 includes a dynamic spatial filtering mask 51 and a mask driving component 52.

[0052] The dynamic spatial filtering mask 51 is located on the transmission light path between the main beam splitter 40 and the camera detection component 90, and is optically conjugate with the plane where the sample to be tested is located. It is used to perform physical defocusing and purification of the emitted light signal transmitted through the main beam splitter 40 through a general mask structure when the second state command is the cut-in command.

[0053] The mask driving component 52 is connected to the dynamic spatial filter mask 51 and is used to drive the dynamic spatial filter mask 51 to cut into or out of the transmission optical path between the main beam splitter 40 and the camera detection component 90 according to the second state command. When the second state command is the cut-in command, the dynamic spatial filter mask 51 is driven to move continuously at a preset rate without needing to maintain phase lock with the pattern refresh of the first-stage light field modulation component 20. This allows the dynamic spatial filter mask 51 to complete at least one full field-of-view scan of the conjugate imaging field of the sample under test within a single exposure time of the camera detection component 90.

[0054] In another exemplary embodiment of this application, the microscopic imaging system based on dual-level joint modulation further includes a beam shaping component 12; the beam shaping component 12 is located between the light source component 10 and the first-level optical field modulation component 20, and is used to perform optical preprocessing operations on the excitation beam 11; the optical preprocessing operations include at least one of beam expansion operation, homogenization operation, wavefront correction operation and polarization adjustment operation.

[0055] In another exemplary embodiment of this application, the front adapter 21 includes a reflector or a polarization control sub-assembly.

[0056] The reflector is disposed in the outgoing light path of the light source assembly 10 and is used to obliquely incident the excitation beam 11 onto the light field encoding assembly 22 at a preset off-axis tilt angle, so as to adjust the incident angle of the excitation beam 11.

[0057] The polarization control sub-assembly includes a polarizer 211, a polarization beam splitter 212, and a waveplate assembly 213. The polarizer 211 is disposed in the output optical path of the light source assembly 10 and is used to purify the polarization of the excitation beam 11. The polarization beam splitter 212 is disposed in the output optical path of the polarizer 211 and is used to reflect the polarization-purified excitation beam 11 to the waveplate assembly 213 and guide the excitation beam 11 after passing through the waveplate assembly 213 to be perpendicularly incident on the optical field encoding assembly 22, so as to adjust the incident angle of the excitation beam 11. The waveplate assembly 213 is disposed between the polarization beam splitter 212 and the optical field encoding assembly 22 and is used to rotate the polarization direction of the excitation beam 11 after being reflected by the polarization beam splitter 212, so as to adjust the polarization state of the excitation beam 11.

[0058] In another exemplary embodiment of this application, the microscopic imaging system based on dual-level joint modulation further includes a sample scanning component 71.

[0059] The sample scanning component 71 is used to carry and drive the sample to be tested to move; wherein, the sample scanning component 71 includes a lateral driving component 72 and an axial driving component 73; the lateral driving component 72 is used to drive the sample to be tested to perform lateral planar movement, and the axial driving component 73 is used to drive the sample to be tested to perform axial depth movement.

[0060] Based on the same inventive concept, this application also provides a microscopic imaging method based on dual-level joint modulation applied to the aforementioned microscopic imaging system based on dual-level joint modulation. The solution provided by this method is similar to the implementation scheme described in the above system. Therefore, the specific limitations in one or more embodiments of the microscopic imaging method based on dual-level joint modulation provided below can be found in the limitations of the microscopic imaging system based on dual-level joint modulation described above, and will not be repeated here.

[0061] In one exemplary embodiment, a microscopic imaging method based on dual-level joint modulation is provided, comprising: The target imaging mode is acquired, and a first state command and a second state command are determined based on the target imaging mode; the target imaging mode includes a structured light illumination super-resolution mode, a wide-field imaging mode, a rotating disk confocal mode, or a two-level joint modulation enhanced structured light illumination mode.

[0062] Based on the first state command, the incident light signal is controlled to irradiate the sample under test, so as to excite the sample under test to generate an emitted light signal; the incident light signal includes an excitation beam 11 or a structured light illumination field.

[0063] Based on the second state command and the emitted light signal, the detection light signal of the sample to be tested is determined; the detection light signal includes the emitted light signal or the emitted light signal purified by physical defocusing.

[0064] A digital image of the sample is determined based on the probe light signal of the sample.

[0065] If the target imaging mode is a wide-field imaging mode or a rotating disk confocal mode, then the digital image is used as a microscopic image of the sample to be tested.

[0066] If the target imaging mode is a structured light illumination super-resolution mode, then the super-resolution microscopic image of the sample to be tested is determined by the frequency domain decoding matrix equation constructed based on the digital image.

[0067] If the target imaging mode is a dual-level joint modulation enhanced structured light illumination mode, then the super-resolution microscopic image of the sample to be tested is determined based on the digital image and the forward imaging model.

[0068] In another exemplary embodiment of this application, determining a first state command and a second state command based on the target imaging modality specifically includes: When the target imaging mode is the structured light illumination super-resolution mode, the first state command is the cut-in command and the second state command is the cut-out command.

[0069] When the target imaging mode is a wide-field imaging mode, the first state command is a cut-out command, and the second state command is a cut-out command.

[0070] When the target imaging mode is the turntable confocal mode, the first state command is the cut-out command, and the second state command is the cut-in command.

[0071] When the target imaging mode is a dual-level joint modulation enhanced structured light illumination mode, the first state command is a cut-in command, and the second state command is a cut-in command.

[0072] In another exemplary embodiment of this application, the super-resolution microscopic image of the sample under test is determined by a frequency domain decoding matrix equation constructed based on digital images, specifically including: Acquire digital images of multiple test samples under multiple illumination directions and multiple preset phase shifts.

[0073] The digital images in each frame are transformed from the spatial domain to the frequency domain using the Fourier transform method to obtain multiple frames of frequency domain images.

[0074] Based on the frequency domain image under each illumination direction, the frequency domain decoding matrix equation for each illumination direction is constructed respectively.

[0075] The frequency domain decoding matrix equations for each illumination direction are inverted to obtain the spectral components of each order for each illumination direction; each spectral component includes a DC component and at least one AC component.

[0076] The AC components of each order of the spectrum under each illumination direction are subjected to spectral shifting to obtain the spectral shifted AC components under each illumination direction.

[0077] The AC component of the spectrum shifted under each illumination direction is superimposed with the DC component of each order of the spectrum under the corresponding illumination direction to obtain the recombined spectrum under each illumination direction.

[0078] The recombined spectra under all lighting directions are merged to obtain the complete spectrum.

[0079] The complete spectrum is transformed from the frequency domain to the spatial domain using the inverse Fourier transform method to obtain a super-resolution microscopic image of the sample under test.

[0080] In another exemplary embodiment of this application, a super-resolution microscopic image of the sample to be tested is determined based on a digital image and a forward imaging model, specifically including: The forward imaging model of two-level joint modulation is constructed using the following formula: .

[0081] in, This represents the forward imaging model; Super-resolution micrographs of the sample to be analyzed. The sample under test is located in a planar position. Pixel value at; Represents the two-dimensional spatial coordinates of the plane containing the sample to be tested; This indicates the output of the forward imaging model and... The corresponding predicted digital image of the j-th frame is located at the plane position of the test sample. The corresponding pixel value; This represents the camera exposure time corresponding to the acquisition of the j-th frame of the digital image of the sample under test; This indicates the position of the digital image of the j-th frame of the sample being acquired at time t, projected onto the plane containing the sample. Structured excitation light field distribution at the location; This represents the convolution operator; Indicates the position in the plane where the sample to be tested is located. The diffusion function value of the optical point at the transmitting end; This indicates the position of the dynamic spatial filter mask 51 in the plane where the sample is located at time t when acquiring the j-th frame of the digital image of the sample to be tested. The transmission function value at that location.

[0082] The objective function for reconstructing the inverse problem is constructed using the following formula: .

[0083] in, The objective function value represents the inverse problem reconstruction; M represents the total number of frames in the digital image. This indicates the position of the j-th frame of the digital image of the test sample in the plane containing the test sample. Pixel value at; Represents the square of the L2 norm; Represents the regularization coefficient; This represents the regularization constraint.

[0084] An optimization algorithm is used to iteratively solve the forward imaging model until the iteration termination condition is met, at which point the iteration ends and a super-resolution microscopic image of the sample to be tested is obtained. The optimization algorithm includes the alternating direction multiplier method or the fast iterative shrinkage threshold algorithm. The iteration termination condition includes the inverse problem reconstruction objective function value being less than a preset objective function threshold, or the change in the inverse problem reconstruction objective function value within a preset number of consecutive iterations being less than a preset change, or the maximum number of iterations being reached.

[0085] This application no longer designs the excitation and detection ends as isolated units, but instead uses a two-stage optical path coupling of "structured excitation-conjugate matched filtering" to jointly determine the physical quality of the original image. The first-stage driving component 23 is responsible for generating a structured excitation field with frequency coding or energy concentration capabilities; the second-stage dynamic conjugate spatial filtering component 50 is responsible for selective transmission based on the in-focus emission distribution corresponding to this excitation field. By utilizing the spatial distribution difference between the in-focus signal and the out-of-focus background on the conjugate surface, physical background suppression is achieved before photoelectric conversion, thereby providing high signal-to-noise ratio basic data for subsequent computational reconstruction. The following uses a specific microscopic imaging process based on two-stage joint modulation as an example to illustrate this application.

[0086] In one exemplary embodiment, the general imaging process of dual-level joint modulation specifically includes the following steps: Step 1: Parameter initialization and component configuration.

[0087] The signal control and computational reconstruction component 100 (hereinafter referred to as "control component 100"), in response to the target imaging mode in the imaging mode command input by the user, first determines the physical switching state between the first-level optical field modulation component 20 and the second-level dynamic conjugate spatial filter component 50: If it is a structured light illumination super-resolution mode: the control component 100 drives the first-level light field modulation component 20 (such as driving the light field encoding component 22) to enter the illumination light path; at the same time, it drives the dynamic spatial filter mask 51 to cut out the detection light path.

[0088] If the instruction is wide-field imaging mode: control component 100 drives the first-stage optical field modulation component 20 to cut out the illumination optical path; at the same time, it drives the dynamic spatial filter mask 51 to cut out the detection optical path.

[0089] If the instruction is a turntable confocal mode: the control component 100 drives the first-stage light field modulation component 20 to cut out the illumination light path; at the same time, it drives the dynamic spatial filter mask 51 to cut into the detection light path.

[0090] If the instruction is a bipolar joint modulation enhanced structured light illumination mode: the control component 100 drives the first-stage optical field modulation component 20 to enter the illumination optical path; at the same time, it drives the dynamic spatial filter mask 51 to enter the detection optical path.

[0091] Step 2, two-stage asymmetric timing synchronization control.

[0092] Based on the cut-in / cut-out states of the first-stage optical field modulation component 20 and the second-stage dynamic conjugate spatial filter component 50 determined in step 1, the control component 100 initializes its parameters and generates a strictly aligned hardware trigger signal stream: Excitation end hard synchronization: For the first-stage optical field modulation component 20 in the cut-in state, the control component 100 sends motion or addressing parameters (such as turntable step angle, mask phase shift or DMD (Digital Micromirror Device) refresh sequence) to the first-stage driving component 23, and sends nanosecond-level trigger signals to the light source component 10 and the camera detection component 90 to ensure that the refresh of the structured excitation pattern is strictly aligned with the camera exposure shutter.

[0093] Filter-end integration synchronization: For the second-stage dynamic conjugate spatial filter component 50 in the cut-in state, the control component 100 sends a continuous drive signal to the mask drive component 52 to set the rotation / translation rate of the mask to match the camera exposure cycle, so as to realize physical spatial integration within a single camera exposure.

[0094] If the first-stage optical field modulation component 20 cuts out the illumination optical path or the dynamic spatial filter mask 51 is in the cut-out state, then its corresponding dynamic synchronization configuration is ignored.

[0095] Step 3: Sample excitation, physical filtering (depending on the mode) and signal acquisition.

[0096] In the timing sequence of step 2 above, the excitation beam 11 output by the light source component 10 is modulated (or not modulated) by the pre-amplifier and the conjugate transition optical path component 60 and focused onto the sample component 70, and the excited emission signal is traced back along the original path.

[0097] Physical purification process: If the dynamic spatial filter mask 51 is in the cut-in state, the returned transmitted signal passes through the mask. The in-focus transmitted signal is efficiently transmitted due to conjugate mapping, while the out-of-focus background signal is physically blocked due to spatial distribution mismatch; if it is in the cut-out state, the transmitted signal is transmitted in full throughput.

[0098] Data storage: The camera detection component 90 converts the received light signal into a digital image, which is then cached by the control component 100 to obtain the digital image or digital image sequence in the corresponding mode.

[0099] Step 4: Establish the bi-level conjugate mapping relationship with the compensation.

[0100] By adjusting the bias of the pattern refresh matrix or mask actuator, color difference shift or mechanical error is compensated to ensure the continued validity of the two-level joint modulation relationship.

[0101] Step 5: Image calculation and reconstruction, and result output.

[0102] The control component 100 retrieves digital images or sequences of digital images from the cache and performs reconstruction based on the currently executing target imaging mode: For structured light illumination super-resolution mode or bipolar joint modulation enhanced structured light illumination mode, it combines temporal control prior parameters (such as phase shift step size) and executes inverse solving algorithms (such as defocusing calculation, frequency domain decoding super-resolution, or physical heuristic deep learning reconstruction) to reconstruct a high-frequency spatial information image that breaks through the diffraction limit, thus obtaining a super-resolution microscopic image of the sample under test. For wide-field imaging mode or rotating disk confocal mode, it directly outputs a single frame digital image as the microscopic image of the sample under test, or executes a basic deconvolution algorithm to improve contrast, thus obtaining a microscopic image of the sample under test.

[0103] As an optional implementation method, the first-level optical field modulation and mask generation method specifically includes: A preferred physical architecture employs a mask turntable coupled with a stepper motor. By etching specific interference patterns onto the partitioned turntable, high-speed switching and stable projection of different spatial modes are achieved. The optical field encoding component 22 is configured as a mask turntable with multiple independent partitions. The first-stage driving component 23 (such as a stepper motor) drives the mask turntable to perform a step-by-step rotation of "one move, one stop, one shot". When a specific partition of the turntable is driven by the motor to enter the illumination optical path and stop, a uniformly emitted parallel beam 13 is transmitted through the partition, projecting the specific transmission pattern etched on the mask partition onto the downstream plane 31 where the sample to be tested is located in a proportionally conjugate manner. This architecture transforms complex phase modulation into purely physical transmission projection, greatly improving the system's stability and phase shift accuracy.

[0104] For traditional two-dimensional structured light illumination (SIM) super-resolution imaging, the mask pattern on the rotating disk is generated based on the two-beam interference principle. One-dimensional periodic fringe gratings with different directions and phases are etched on different sector sections of the mask rotating disk, such as... Figure 2 As shown in (b), a schematic diagram of a partitioned mask turntable for two-beam interference structured light illumination is provided. When the corresponding partition is cut into the optical path, the intensity distribution of the excitation field generated by the transmitted light field at plane 31 where the sample to be tested is located is as follows: .

[0105] in, Indicates the position of the sample under test in the plane. The excitation field intensity distribution at the location; The average intensity of illumination; The modulation contrast of the interference fringes; This is the spatial angular frequency vector in the k-th rotation direction; For the first The phase shift amount per step. In a typical configuration, the mask turntable can be divided into 9 partitions (corresponding to 3 directions, with 3 phase shifts in each direction). The stepper motors switch between the partitions sequentially, so that the corresponding structured light mode can be accurately projected onto the sample to complete image acquisition.

[0106] Besides achieving a similar illumination pattern and imaging process to dual-beam SIM through projection, three-beam SIM can also be implemented, with the mask pattern on the turntable generated based on the principle of three-beam interference. Figure 2 As shown in (a), a schematic diagram of a partitioned mask turntable for three-beam interference structured light illumination is provided. The three beams are distributed at a specific angle (e.g., 120 degrees) in the spatial frequency domain, and their interference intensity exhibits a two-dimensional honeycomb lattice distribution in space. Since the computational reconstruction of three-beam interference SIM requires at least 7 phase shifts to ensure the full rank of the solution matrix, the optical field encoding turntable is designed with a structure of 7 (or multiples thereof) independent partitions. During the digital generation and physical etching stages of the mask, considering that the mask itself will undergo a physical rotation angle α when the stepper motor rotates each partition to the effective excitation field of view, the interference lattice direction projected onto the sample plane is kept absolutely constant. The interference pattern in each partition of the turntable is pre-compensated with a reverse rotation coordinate of -α during generation. During data acquisition, the stepper motor drives the turntable to sequentially switch these 7 partitions, and the transmitted beam can accurately project a three-beam interference structured light excitation mode with a strict 7-step phase shift sequence onto the sample plane.

[0107] It should be specifically stated that the structured light field generation method based on mask turntable projection protected in this application is not limited to two-beam or three-beam interference SIM. Those skilled in the art should understand that by changing the transmission function matrix of the turntable mask, the system of this application is also applicable to structured light illumination or higher-dimensional spatial lattice illumination modes generated by four-beam, six-beam, or even more-beam interference. As long as the physical architecture and working mechanism of "etching a preset multi-beam interference pattern distribution onto the turntable partition and achieving mode and phase shift switching through step rotation" are adopted, they should fall within the protection scope of this application.

[0108] Furthermore, to expand the applicable scenarios of this application system, such as Figures 3-5 As shown, the first-stage optical field encoding component 22 and the driving mechanism can also adopt the following equivalent configuration to generate other structured excitation fields.

[0109] Multifocal array excitation: The outgoing parallel beam 13 is incident on the microlens array (MLA) which serves as the optical field encoding component 22. Figure 3 c and 4a, 4b) or Daman grating ( Figure 3 d and Figure 4 Static transmission arrays (c, 4d) have wavefronts segmented into discrete high-contrast focal arrays. To achieve spatial scanning of this array on the sample surface, a scanning device 26 (such as a scanning galvanometer) is positioned at the Fourier plane (pupil plane) of the first relay lens 24, and its deflection angle is controlled by a galvanometer driver, which serves as the first-stage driving component 23. After the scanning array is projected onto the sample surface via the second relay lens 25, and with the physical purification of the second-stage filtering component, high signal-to-noise ratio image scanning microscopy (ISM) imaging can be achieved. Figure 3 (c) Figure 4 (a) and Figure 4 As shown in (b), a schematic diagram of a cascaded scanning configuration based on a static array and pupil plane deflection, a schematic diagram of a microlens array structure, and a schematic diagram of a multifocal illumination pattern generated by the microlens array are provided.

[0110] Dynamic speckle field excitation: The outgoing parallel beam 13 passes through the speckle phase plate or frosted glass diffuser, which serves as the optical field encoding component 22, such as... Figure 5 (a) and Figure 5 As shown in (b), a schematic diagram of the speckle phase plate structure and an image of the dynamic speckle illumination field generated by the speckle phase plate are provided. The control component 100 sends a high-frequency drive signal through the actuator motor, which serves as the first-stage drive component 23, causing the speckle plate to generate continuous spatial displacement in the optical path. The transmitted beam is transmitted through the first relay lens 24 and the second relay lens 25, generating a dynamic coherent speckle field with statistical fluctuations on the plane 31 where the sample is located, for blind structured light microscopy imaging.

[0111] Special beam excitation (such as a Bessel beam): The outgoing parallel beam 13 first passes through a pre-transfer assembly 21 containing a polarizer 211, a polarization beam splitter 212, and a waveplate assembly 213, such as... Figure 3 As shown in (b), a schematic diagram of a vertical coaxial architecture based on polarization state modulation is provided. Polarization state purification and alignment are performed, followed by perpendicular incident light onto the target surface of a pure phase-type liquid crystal spatial light modulator (SLM), which serves as the light field encoding component 22. A high-frequency backplane driving circuit, serving as the first-stage driving component 23, dynamically loads a conical phase mask matrix onto the SLM. The modulated light field, after passing through the first relay lens 24 and the second relay lens 25, generates a Bessel beam array with concentrated energy in the central main lobe and possessing an extremely long axial non-diffraction depth on the plane 31 where the sample is located, to resist the scattering effects of deep biological tissues, such as... Figure 5 (c) and Figure 5 As shown in (d), a schematic diagram of the phase mask used to generate the Bessel beam and a diagram of the illumination field effect of the Bessel beam are provided.

[0112] As an optional implementation method, the second-stage dynamic conjugate spatial filtering method specifically includes: The control component 100 regulates the second-stage dynamic conjugate spatial filter component 50, and performs physical defocusing and purification of the emitted light field through a high-speed integration mechanism that is spatiotemporally decoupled from the excitation end.

[0113] Unlike traditional filtering components that require strict alignment with the excitation pattern in spatial coordinates, the dynamic spatial filtering mask 51 in this application is configured as a general mask structure with full field-of-view spatial translation invariance.

[0114] Control component 100 can invoke masks with the following typical structures to adapt to all first-level excitation modes (such as lattice, stripe, speckle, or uniform wide field): Archimedes spiral pinhole array turntable: Utilizing a high-density arrangement of pinholes along a spiral, this ensures that the scanning trajectory of the pinholes uniformly covers the entire conjugate imaging field of view (within the case of the spiral) as the turntable rotates. Taking =12 as an example, (For symmetrical sectors of the mask pattern). For example... Figure 2 As shown in (c), a schematic diagram of an Archimedes spiral pinhole array turntable is provided; Slit turntable: Employs periodic light-transmitting slits, providing higher light throughput than pinhole arrays, such as... Figure 2 As shown in (d), it provides a method for... A schematic diagram of a periodic light-transmitting slit turntable with a value of 4.

[0115] The purpose of this configuration is to completely eliminate the spatial coordinate binding requirement between the second-level filter mask and the first-level structured excitation pattern in terms of physical structure, thus eliminating the extremely demanding two-level micron-level alignment and calibration process.

[0116] As an optional implementation, during the image acquisition cycle, the control component 100 performs an independent timing control method for the second-stage dynamic conjugate spatial filter component that is completely asynchronous and decoupled from the first-stage modulation: Decoupled high-speed drive: Control component 100 sends a constant speed drive command to mask drive component 52, causing it to drive the dynamic spatial filter mask 51 (such as a turntable) to perform high-speed continuous rotation (speed set to...). This rotation timing does not need to be phase-locked with the pattern refresh and phase shifting cycle of the first-order light field.

[0117] Uniform scanning of exposure cycle: Control component 100 sets the single exposure time of camera detection component 90. At this time Inside, the high-speed rotating dynamic spatial filter mask 51 completes at least one full field-of-view scan on the conjugate surface of the emission optical path (i.e., ,in (This refers to the number of symmetrical sectors in the mask pattern). The purpose of this step is to utilize the detector in time... The integral effect within the mask transforms the instantaneous discrete transmission distribution of the mask into an "equivalent semi-transparent and semi-reflective" filtering plane that is absolutely uniform in the time dimension, eliminating the shadows or raster artifacts left by the mask's own structure on the final image.

[0118] As an optional implementation, the control component 100 constructs a two-level joint modulation forward imaging model based on the system's underlying physical mechanism, and reconstructs a high-fidelity, super-resolution image of the target from the acquired digital image sequence by constructing and solving the inverse problem objective function. Let... To uniformly map to the two-dimensional spatial coordinates of the plane 31 where the sample to be tested is located.

[0119] In this implementation, the control component 100 establishes a positive mathematical mapping relationship from the "true distribution of samples" to the "image acquired by the camera" based on the system's optical transmission link; that is, a positive physical model of two-level joint modulation. This model not only encompasses the first-level structured illumination but also innovatively embeds the spatiotemporal integration process of the second-level dynamic conjugate spatial filtering. For the acquired j-th frame of digital image ( The forward physics model constructed by control component 100 Satisfies the following integral equation: .

[0120] in, To and The corresponding theoretical digital image of the j-th frame is located at the plane position of the sample under test. The corresponding pixel value; Super-resolution micrographs of the sample to be analyzed. The sample under test is located in a planar position. The pixel value at that location (the actual phosphor density or reflectance spatial distribution) is the final output of the inverse problem solution. This indicates the position of the digital image of the j-th frame of the sample being acquired at time t, projected onto the plane containing the sample. The structured excitation light field distribution at the location is the structured excitation light field distribution projected by the first-level light field modulation component 20 onto the surface of the sample under test. According to the aforementioned modes, it can be a discrete lattice, coherent fringes, or speckle that varies with time t. Indicates the position in the plane where the sample to be tested is located. The optical point spread function value at the transmitter is the optical point spread function of the micro-magnification and conjugate transition optical path component 60 at the transmitter, which characterizes the intrinsic optical blur of the system. This indicates the position of the dynamic spatial filter mask 51 in the plane where the sample is located at time t when acquiring the j-th frame of the digital image of the sample to be tested. The transmission function value at that location is the transmission function of the dynamic spatial filter mask 51 in the second-stage dynamic conjugate spatial filter assembly 50. During camera exposure, this mask can move with the turntable or translation stage. This represents the single exposure time of the camera detection component 90. This integration operation rigorously characterizes the physical space averaging effect under temporal decoupling. The plane position of the sample to be tested The system detects noise, including Poisson shot noise and Gaussian readout noise.

[0121] In this implementation, after acquiring the M-frame digital image sequence actually output by the camera detection component 90, the control component 100 transforms the image reconstruction task into an optimization problem. The control component 100 constructs and solves the following inverse problem reconstruction objective function to achieve inverse problem reconstruction: .

[0122] in, The optimal value of S minimizes the objective function J(S). Specifically, considering the mixed Poisson-Gaussian noise characteristics in microscopic imaging, the specific objective function constructed by the control component 100 includes a fidelity term. With regularization constraints : .

[0123] Among them, the data fidelity item is It is used to measure the difference between the observed results and the ideal forward propagation results. The objective function for the inverse problem reconstruction is represented by M; M represents the total number of frames in the digital image. This indicates the position of the j-th frame of the digital image of the sample being tested, actually acquired by the camera detection component 90, within the plane containing the sample. Pixel value at; It represents the square of the L2 norm, which characterizes the residual energy between the model prediction and the actual observation. This represents the regularization coefficient, used to balance the weights of data fit and prior constraints; This represents the regularization constraint.

[0124] After the objective function is constructed, the control component 100 calls the internal computational reconstruction engine and uses the Alternating Direction Method of Multipliers (ADMM) and the Fast Iterative Shrinkage-Thresholding Algorithm (FISTA) to efficiently solve the above inverse problem.

[0125] Since the second-stage dynamic conjugate spatial filter component 50 is already in the hardware level (i.e., in the forward model) The parameters cut off most of the out-of-focus background light, greatly reducing the positive operator. The condition number. Therefore, when performing gradient descent iterations of this objective function, the control component 100 can converge to the global optimum with fewer iterations. This effectively avoids the structural artifacts that are easily generated when removing backgrounds using pure algorithms, and ultimately outputs high-contrast, super-resolution microscopic images.

[0126] As an optional implementation, when only the first-level cut is performed, the high-frequency spatial information of the sample is modulated and aliased into the passband of the system's optical transfer function by the structured illumination field 30. The control component 100 calls the corresponding frequency domain decoding model (here, a traditional reconstruction algorithm, which is independently selectable from the steps described above—building a two-level joint modulation forward imaging model based on the system's underlying physical mechanism and reconstructing a high-fidelity, super-resolution image of the target from the acquired digital image sequence by constructing and solving the inverse problem objective function) according to the specific physical configuration of the first-level optical field encoding component 22. If the first stage uses a two-beam interference mask (e.g., configured with 3 rotation directions, each direction divided into 3 phase-shift partitions), the excitation field contains one DC component C0 and two AC components C +1 C -1 In a specific direction, the camera captures 3 frames with a preset phase shift (e.g., ...). The image of ) has the following frequency domain equation: .

[0127] in, Representing spectral coordinates, C0, C +1 C -1 These represent the 0-frequency component and the ±1-level components in the frequency domain, respectively. , , These respectively represent the preset phase shifts as follows: The spectrum of the image observed at that time.

[0128] Control component 100 by controlling the above The mixing matrix is ​​inverted to separate the components of each order, and this operation is repeated in three directions. Finally, the spectrum is shifted and recombined by generalized Wiener filtering.

[0129] If a three-beam interference mask is used in the first stage, a single interference pattern generates one DC component and six AC components in the frequency domain (corresponding to three pairs of symmetrical high-frequency extrema). The control component drives the mask turntable to switch in steps, performing at least seven phase shifts (if a 1-2-4 phase shift step rule is used, the resulting system phase shift matrix is ​​set to...). Its frequency domain decoding equation is: .

[0130] in, , , , These correspond to ±2-level spectral components and ±3-level spectral components, respectively.

[0131] Because the preset physical phase shift of the turntable partition ensures As a full-rank matrix, control component 100 calculates... (or least squares pseudo-inverse) can decouple all high-frequency AC components, and then reassemble and stitch them together in the spatial frequency domain. Finally, a high-fidelity lattice super-resolution image is reconstructed by inverse Fourier transform.

[0132] In one exemplary embodiment, such as Figure 1 As shown, a schematic diagram of the structure of a microscopic imaging system based on dual-level joint modulation and the connection relationships between the structures is provided.

[0133] The light source assembly 10 is used to output the excitation beam 11. The light source assembly 10 can be a single-mode laser, a multimode laser, a high-power light-emitting diode (LED), or a combination thereof. To expand the system's application scope, the light source assembly 10 can also be a broadband supercontinuum laser, a tunable laser, or an ultrafast femtosecond laser (to support nonlinear multiphoton excitation modes). During multi-wavelength / multicolor imaging, multicolor or wide-field illumination beams 14 can be formed through fiber combining, free-space combining, dichroic mirror combining, acousto-optic tunable filter (AOTF) timing modulation, or other wavelength multiplexing methods.

[0134] A beam shaping component 12 is disposed downstream of the light source component 10 and is used to expand, homogenize, correct the wavefront, and adjust the polarization of the excitation beam 11 to generate a uniform outgoing parallel beam 13. The beam shaping component 12 includes, but is not limited to, multimode fiber, single-mode polarization-maintaining fiber, liquid waveguide, filter pinhole, beam expander lens group, aspherical lens, achromatic lens, mirror group, microlens homogenizing array, or a combination thereof.

[0135] The first-stage optical field modulation component 20 receives the outgoing parallel beam 13 and performs structured modulation on its spatial amplitude, phase, spatial frequency, or polarization state to generate a preset structured illumination field 30 (e.g., a multifocal array lattice, phase-shifting fringes, or a specific coded speckle). This structured illumination field 30 is ultimately strictly conjugated to a preset target region on the plane 31 where the sample under test is located, achieving selective excitation of a specific spatial mode.

[0136] As an optional implementation, the first-stage optical field modulation component 20 includes: a pre-transfer component 21, an optical field encoding component 22, a first-stage driving component 23, a first relay lens 24, and a second relay lens 25.

[0137] The front adapter component 21 is a reflector used to adjust the incident angle of the collimated outgoing parallel beam 13, so as to realize the flexible folding of the beam and ensure that it is incident on the light field encoding component 22 at a specific angle.

[0138] The optical field encoding component 22 here is a partitioned turntable. Different partitions of the turntable have different transmission masks distributed on them, which are used to modulate the amplitude of the outgoing parallel beam 13 and give it specific spatial pattern attributes. For example Figure 2 (a) and Figure 2As shown in (b), schematic diagrams of partitioned mask turntables for three-beam interference structured light illumination and two-beam interference structured light illumination are provided. The structured illumination field 30 modulated by this mask is ultimately projected onto the plane 31 where the sample to be tested is located. Equivalently, any optical and / or electrical method that can be used to impart specific spatial pattern properties to the incident beam can be replaced, including but not limited to active digitally programmable arrays (such as digital micromirror devices (DMDs) utilizing pixelated electro-optic / magneto-optic effects, liquid crystal spatial light modulators (SLMs), or acousto-optic deflection arrays), passive microstructure arrays (such as microlens arrays, static speckle generators, or diffractive optical elements (DOEs)), or spatial frequency domain micromask structures (such as multi-segmented stripe masks or amplitude-type transmission gratings).

[0139] The first-stage driving component 23 is used to drive the partitioned turntable to rotate in steps, and can also drive the turntable to translate radially, controlling its different radial positions across the optical axis and its exit from the optical path. Equivalently, the first-stage driving component 23 can adopt different hardware implementations for different light field encoding components 22, as long as it can perform spatial scanning, phase shifting, or pattern replacement on the structured illumination field 30. If the light field encoding component 22 is a passive microstructure or mask, the first-stage driving component 23 uses mechanical / acoustic-optical deflection actuators such as optical scanning mirrors, acousto-optic deflectors, precision rotary motors, or polyhedral scanning mirrors; if the light field encoding component 22 is an active digital array, the driving unit directly manifests as high-frequency electronically controlled refresh addressing logic and backplane driving circuitry.

[0140] The first relay lens 24 and the second relay lens 25 constitute a 4f optical assembly, which is used to establish the optical conjugate mapping of the system and ensure that the structured illumination field 30, after being encoded and driven, is transmitted to the downstream optical path without loss according to the predetermined spatial geometry and magnification.

[0141] The illumination beam 14, after passing through the first-stage optical field modulation component 20, reaches the main beam splitter 40 (such as a dichroic mirror, polarizing beam splitter, or broadband neutral beam splitter), undergoes reflection (or transmission), and enters the second-stage dynamic conjugate spatial filter component 50. The second-stage dynamic conjugate spatial filter component 50 is positioned precisely optically conjugate to the plane 31 where the final sample is located, and is used to perform physical defocusing and spatial matching filtering on the emitted light field within a single exposure integration period of the camera. To accommodate multimodal imaging and cover different physical occlusion mechanisms, the second-stage dynamic conjugate spatial filter component 50 includes a dynamic spatial filter mask 51 and a dynamic mask driving component 52.

[0142] The dynamic spatial filtering mask assembly 51 here is a rotating disk containing a pinhole array or slit array with an Archimedean spiral arrangement, used to block out-of-focus background signals and highly transmit focal plane signals, such as... Figure 2 (c) and Figure 2 As shown in (d), schematic diagrams of an Archimedean spiral pinhole array turntable and a periodic light-transmitting slit turntable are provided. Equivalently, any physical or digital interface with a preset spatial transmission function can be used as a substitute, such as a linearly translated physical mask (e.g., a one-dimensional translational slit), a digitally programmable dynamic mask (e.g., a DMD or transmissive LCD (Liquid Crystal Display) array), or a dynamic filter film based on phase change materials.

[0143] The mask driving component 52 is used to drive the dynamic spatial filter mask component 51 to move continuously along a preset trajectory. It can drive the dynamic spatial filter mask component 51 to perform rapid rotation and can also drive the turntable to perform radial translation, controlling its different radial positions across the optical axis and its exit from the optical path. Equivalently, if the dynamic spatial filter mask component 51 uses a linear translation physical mask (such as a one-dimensional translation slit), then the mask driving component 52 is a linear / two-dimensional translation of a piezoelectric displacement stage; if the dynamic spatial filter mask component 51 uses a digitally programmable dynamic mask (such as a DMD or transmissive LCD array), then the mask driving component 52 is used to control the high-frequency refresh of the transmissivity matrix of the digital mask. In short, any scanning method that achieves spatial physical integration of the emitted signal within the camera's exposure time, and any motion device that controls the module's entry and exit from the optical path, can be equivalently replaced.

[0144] In an optional embodiment, the second-stage dynamic conjugate spatial filter component 50 may also be disposed only in the emission optical path behind the main beam splitter 40, and only perform unidirectional filtering and control on the emission signal.

[0145] The excitation beam, after being transmitted or reflected through the second-stage dynamic conjugate spatial filter component 50, is divided into several structured wavefronts, and then enters the micro-magnification and conjugate transition optical path component 60, and finally irradiates a specific xy plane of the sample component 70 or a sample volume with a preset focal depth range (i.e., the plane 31 where the sample to be tested is located), to achieve sparse or coherent excitation of a specific structural pattern.

[0146] The emitted beam carrying sample spatial information (such as fluorescence, scattered light, or second harmonic) returns along its original path. After being collected by the microscopic amplification and conjugate transition optical path component 60, it is physically defocused and purified again by the second-stage dynamic conjugate spatial filter component 50. The purified emitted beam passes through (or is reflected from) the main beam splitter 40 and is finally projected proportionally by the conjugate transition component 80 onto the detection plane 32 of the camera detection component 90.

[0147] The microscopic magnification and conjugate transition optical path assembly 60 is configured to ensure that the first-stage modulated structured illumination field 30 is accurately projected onto the plane 31 where the sample is located, while simultaneously collecting the emitted light. It consists of a microscope objective 62, a tube lens 61, and necessary conjugate relay mirrors, ensuring that a strict conjugate relationship is maintained between the second-stage dynamic conjugate spatial filter assembly 50, the camera detection plane 32, and the plane 31 where the sample is located.

[0148] The sample scanning component 71 and its associated lateral drive component 72 and axial drive component 73 are used to perform three-dimensional spatial movement of the sample to be tested.

[0149] The camera detection component 90 employs a pixel array with photoelectric conversion capabilities, including a scientific-grade CMOS camera (sCMOS), an electron multiplier CCD (EMCCD), a single-photon avalanche diode (SPAD) array, or an event-triggered neuromorphic camera, to output image data within a preset time sequence.

[0150] The signal control and computation reconstruction component 100 (composed of an FPGA (Field-Programmable Gate Array) acquisition card, a microcontroller, and host computer software) is configured as the central brain of the system, and is divided into three main logic pipelines: System hardware synchronization: The signal control and computational reconstruction component 100 performs rigorous synchronization control on the light source component 10, the first-stage light field modulation component 20, the second-stage dynamic conjugate spatial filtering component 50, the microscopic magnification and conjugate transition optical path component 60, the sample scanning component 71, and the camera detection component 90 by outputting high-precision analog signals, digital communication commands, or nanosecond-level hardware trigger signals. Specifically, this central control is responsible for dynamically coordinating laser output power, pattern refresh rate, scanning cycle, turntable / mask spatial positioning, sample stage three-dimensional motion trajectory, and camera exposure sequence to ensure absolute accuracy of the light field modulation state during the acquisition of each frame of image.

[0151] Status and data management: responsible for the logical scheduling of imaging tasks (such as mode switching and disk area selection) and the high-speed caching, conjugate calibration and error compensation of raw data.

[0152] Computational reconstruction: Extract the original image sequence from the data pipeline, combine it with the temporal control prior parameters (such as phase shift and scanning trajectory) to construct an imaging physical model, and reconstruct high-frequency spatial information images that exceed the diffraction limit by executing inverse solving algorithms (such as defocusing calculation, pixel reassignment, frequency domain decoding super-resolution or physical heuristic deep learning reconstruction, etc.).

[0153] As an alternative implementation, depending on the modulation method and the physical properties of the device, the first-stage optical field modulation component can also have the following equivalent configurations: (1) Folded modulation architecture based on reflective active array, such as Figure 3 (a) and Figure 3 As shown in (b), two typical optical path topologies are illustrated in which the front-end adapter 21 and the optical field coding component 22 work together.

[0154] ① Based on off-axis oblique incidence, non-coaxial architecture, such as Figure 3 As shown in (a), the pre-transition component 21 is configured as a reflection-guided optical path (e.g., including a total reflection mirror) to control the outgoing parallel beam 13 to be obliquely incident on the optical field encoding component 22 at a preset off-axis tilt angle (taking the effective target surface of a digital micromirror device, DMD, as an example here). This off-axis oblique incident architecture allows the outgoing beam after being reflected and encoded by the target surface to be naturally separated from the incident beam in terms of spatial propagation angle, enabling high-throughput entry into the next stage optical path without the need for additional beam splitting elements.

[0155] ② Vertical coaxial architecture based on polarization state modulation, such as Figure 3 As shown in (b), for the polarization-dependent optical field encoding component 22 (taking the effective target surface of a pure phase-type liquid crystal spatial light modulator SLM as an example), the pre-transition component 21 is configured as a vertical coaxial optical path including a polarization state preprocessing network. Specifically, this configuration includes: Polarizer 211: Located on the incident beam path, it is used to perform polarization purification on the outgoing parallel beam 13, thereby increasing the polarization extinction ratio of the incident light to meet the stringent physical requirements for polarization purity of the SLM target surface.

[0156] Polarization beam splitter 212: disposed downstream of the polarizer 211, used to reflect a specific linearly polarized component (such as S-polarized light) and guide it to be incident perpendicularly onto the SLM target surface.

[0157] Waveplate assembly 213 (preferably a half-wave plate): interposed between the polarization beam splitter 212 and the SLM target surface 22. The waveplate assembly 213 is used to precisely rotate the polarization vector direction of the linearly polarized light, ensuring that the polarization direction of the beam is strictly parallel and aligned with the effective phase modulation fast axis of the SLM target surface, so as to maximize diffraction efficiency and suppress unmodulated zero-order background light.

[0158] For the active digital array (such as DMD or SLM) in the above paradigm, the first-stage driving component 23 corresponds to a high-frequency electronically controlled refresh addressing logic and backplane driving circuit, used to update the modulation pattern of the encoding engine in real time. The modulated outgoing light then enters a 4f system composed of a first relay lens 24 and a second relay lens 25 to proportionally and faithfully conjugate the modulation pattern onto the target plane of the subsequent system.

[0159] (2) Cascaded deflection architecture based on a transmissive passive array. For example... Figure 3 (c) and Figure 3 As shown in (d), two typical architectural configurations are illustrated for the optical field encoding component 22, the first-stage driving component 23, and the subsequent optical path. The uniform outgoing parallel beam 13 is first folded and guided by the front adapter component 21 (taking the total reflection mirror as an example) and then incident on the subsequent optical path.

[0160] ① Cascaded scanning configuration based on static array and pupil plane deflection, such as Figure 3 As shown in (c), in this configuration, the encoding engine and the driving mechanism are functionally decoupled in physical space. The optical field encoding component 22 employs a static transmission array (such as a microlens array (MLA)) to convert the outgoing parallel beam 13 into a static discrete multifocal array. To achieve spatial scanning of this array, an additional scanning device 26, such as a scanning galvanometer, is added and placed at the Fourier plane (pupil plane) of the first relay lens 24. The first-stage driving component (galvanometer driver) 23 controls the deflection of the scanning device 26 (scanning galvanometer), and the static optical field undergoes controlled angular deflection when passing through this conjugate node. The first-stage driving component 23 and the scanning device 26 include, but are not limited to, mechanical / acousto-optic deflection actuators such as optical scanning galvanometers, acousto-optic deflectors, precision rotary motors, or polyhedral scanning mirrors.

[0161] ② Interference configuration based on diffraction coding and frequency domain filtering, such as Figure 3As shown in (d), in this configuration, the system utilizes a physical masking mechanism in the spatial frequency domain to purify specific spatial modes. The optical field encoding component 22 employs a transmission diffraction device (such as a one-dimensional phase grating, a diffractive optical element DOE, or a partitioned fringe disk) to split the outgoing parallel beam 13 into multiple diffraction orders (such as 0th order and ±1st order) carrying different spatial frequencies. The first-stage driving component 23 drives the optical field encoding component 22 to perform translation or rotation to achieve phase shift and direction switching, and can simultaneously control the insertion and withdrawal of the optical field encoding component 22 and the spatial frequency domain mask 27 into the optical path. The spatial frequency domain mask 27 is placed at the spatial spectral plane of the first relay lens 24, and its transmission function is configured to block unwanted diffraction orders (such as centrally blocking 0th order light), allowing only the target order to pass through the second relay lens 25, thereby causing interference recombination at the target conjugate plane 30 to generate high-contrast structured fringes.

[0162] The specific components, optical path configurations, and algorithm flows described in the embodiments of this application are preferred embodiments listed to illustrate the core idea of ​​the application (i.e., two-stage joint modulation and computational reconstruction of "structured excitation-conjugate matched filtering"). Those skilled in the art can employ various equivalent alternatives to achieve the same technical objective (physical defocusing and computational super-resolution) without departing from the core concept of this invention. Specific alternatives include, but are not limited to: 1. Equivalent replacement of the first-stage optical field modulation component.

[0163] Hardware device alternatives: The optical field encoding component 22 for generating the structured excitation light field is not limited to digital micromirror devices (DMDs), liquid crystal spatial light modulators, microlens arrays, or Dammann gratings. Acousto-optic deflection arrays, axial conical lenses, dynamic scattering devices, or other devices capable of spatially shaping the beam amplitude, phase, or polarization state can also be used. The differences between these devices lie only in refresh rate and diffraction efficiency, but all can achieve the purpose of active spatial encoding at the excitation end.

[0164] Alternative Scanning and Evolution Methods: The first-stage driving component 23, which drives the structured light field to move across the sample surface, can be replaced by an acousto-optic deflector (AOD), an electro-optic deflector (EOD), a polyhedral scanning mirror, or a piezoelectric sample stage to directly drive the sample to move relative to the surface, in addition to digital refresh of an electronically controlled array or deflection of a mechanical galvanometer. That is, this application does not limit it to beam scanning; any method that can achieve the relative spatial evolution of the light field and the sample is an equivalent alternative to this application.

[0165] 2. Equivalent replacement of the second-stage dynamic conjugate spatial filter component.

[0166] Physical Form and Switching Mechanism Alternatives: The physical form of the second-stage dynamic conjugate spatial filter component 50 is by no means limited to a rotating turntable. Without altering the core concept of this application, it can be completely replaced by: a linear sliding multi-segment filter (switching filter sub-regions via one-dimensional translation); a digitally programmable mask array (such as placing a transmissive LCD array or another DMD at the emitter, and using electronic control to change the pixel transmittance to equivalently move the physical mask); a pluggable disk or MEMS (Micro-Electro-Mechanical System) array. All of these alternatives can achieve multimodal spatially selective filtering on the conjugate surface of the emitted optical path.

[0167] Microstructure substitution for filter function: The implementation of filter sub-regions is not limited to hard openings such as pinholes or slits (pure amplitude type). They can also be replaced by grayscale transmission structures, phase-type microstructures, diffraction-type microstructures, or three-dimensional microstructures formed by local refractive index modulation.

[0168] Alternative operating states: This filter component can be used to perform spatial integration and averaging in the time dimension by high-speed motion (such as rotation or high-frequency translation), or it can be used as a static filter element in certain specific modes (such as fixed single-point scanning). Both methods can achieve the purpose of physically blocking out-of-focus backgrounds.

[0169] 3. Equivalent substitutions for matching strategies.

[0170] The matching relationship between the first-level excitation mode and the second-level filter sub-region is not limited to an absolute one-to-one correspondence. A many-to-one grouping correspondence can also be adopted; even the optimal matching strategy can be dynamically and arbitrarily combined across modes by the system control center based on real-time feedback of sample scattering degree and signal-to-noise ratio through software algorithms.

[0171] 4. Equivalent substitution of reverse solution and reconstruction algorithms.

[0172] After physical pre-suppression is completed on the detection side, the control center extracts high-quality raw data from the cache for processing using a reverse computation algorithm. This algorithm is not limited to pixel redistribution or frequency domain coherent decoding (SIM algorithm). Wiener deconvolution, Richardson-Lucy deconvolution, compressed sensing reconstruction, or even physics-inspired deep neural network denoising and super-resolution algorithms can also be used. The replacement of the above algorithms is only a change in the subsequent numerical processing tools; the fundamental premise for their efficient convergence still relies on the high signal-to-noise ratio underlying data provided by the hardware dual-level modulation of this application.

[0173] In summary, this application has the following beneficial effects: 1. Thanks to the dual-level joint modulation configuration, this application achieves physical background suppression before photoelectric conversion, significantly improving image contrast and signal-to-noise ratio, and can suppress defocused background light at the source. The contrast of digital images is several times higher than that of traditional solutions, especially in environments with thick samples and strong scattering, where its signal-to-noise ratio advantage is extremely high.

[0174] 2. Thanks to the timing control method, the second-stage dynamic conjugate spatial filter component 50 (such as a turntable mask) is decoupled from the first-stage modulation mode in time and space. The second-stage mask only needs to perform high-speed continuous rotation asynchronously with the excitation end. By utilizing the time integration effect of the camera within a single exposure, uniform filtering across the entire field of view can be achieved in a statistically significant sense. This avoids the problem that traditional two-stage modulation systems must perform microsecond-level or nanosecond-level phase locking, enabling the system to maintain extremely high stability even under high-speed imaging.

[0175] 3. By pre-setting different spatial pattern matrices and mask driving parameters in the signal control and computational reconstruction component 100, this application software-izes the hardware functionality. The first-stage optical field modulation component 20 (such as DMD / SLM) can instantly switch the excitation field morphology, while the second-stage dynamic conjugate spatial filtering component 50, which has multiple partitions (such as pinhole region, slit region, and through region), can adapt synchronously. This modular design allows the system to adaptively adjust the physical imaging mechanism according to sample characteristics (such as live cell dynamic scanning or deep imaging of thick tissue). A single device can be compatible with multiple modes such as ISM, SIM, speckle illumination, and wide field. Imaging mode switching takes only milliseconds and does not require replacement of any core optical path, greatly expanding the scientific research and industrial application scenarios of the device.

[0176] 4. In computational imaging, the quality of inverse problem reconstruction is highly dependent on the condition number of the forward model. Because this application pre-truncates most of the low-frequency defocus background at the physical level through the second-stage dynamic conjugate spatial filter component 50, the useful information ratio in the raw data acquired by the system is extremely high, significantly reducing the ill-conditioned nature of the forward operator. Therefore, during objective function optimization, the algorithm can more stably separate high-frequency components, significantly reducing parameter estimation bias and image artifacts induced by high background noise.

[0177] 5. The control component 100 can monitor and compensate for deviations caused by wavelength chromatic aberration or mechanical repetitive positioning in real time. Through dynamic deviation compensation at the software level, the physical limitations of the optical hardware are overcome, ensuring that the system maintains strict structured excitation-conjugate matched filter physical constraints throughout its entire lifecycle.

[0178] 6. This application achieves on-demand illumination through the first-stage light field modulation component 20, exciting only the region of interest. Combined with the efficient filtering of the second-stage dynamic conjugate spatial filter component 50, each photon arriving at the camera carries more effective in-focus information. This dual purification through both physical and computational methods allows users to obtain high-quality images with lower optical power, protecting the physiological activity of biological samples.

[0179] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 6 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores microscopic imaging processing data based on dual-level joint modulation. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a microscopic imaging method based on dual-level joint modulation.

[0180] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer equipment to which the present application is applied. Specific computer equipment may include, for example, [the following is a list of possible additional structures]. Figure 6 The embodiments show more or fewer components, combinations of certain components, or different component arrangements. In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, which the processor executes to implement the steps in the above-described method embodiments.

[0181] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0182] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0183] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0184] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0185] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0186] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0187] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A microscopic imaging system based on dual-level joint modulation, comprising: The light source assembly, the first-stage optical field modulation assembly, the main beam splitter, the micro-magnification and conjugate transition optical path assembly, the second-stage dynamic conjugate spatial filter assembly, the camera detection assembly, and the signal control and computational reconstruction assembly. The beam emitter is located on the output optical path of the light source assembly, which outputs an excitation beam. The modulation assembly, positioned between the light source assembly and the beam emitter, modulates the excitation beam into a structured light illumination field when the first state command is a cut-in command. The beam emitter reflects the incident light signal to the optical path assembly. The optical path assembly, on the reflected optical path of the beam emitter, focuses the incident light signal onto the plane of the sample under test, collecting the emitted light signal of the sample and returning it to the beam emitter. The incident light signal includes either the excitation beam or the structured light illumination field. The filtering assembly, positioned between the beam emitter and the detection assembly, is optically conjugate with the sample plane and, in the second state command... When the cut-in command is given, the emitted light signal transmitted by the beam is physically defocused and purified; the detection component converts the detection light signal into a digital image on the transmitted light path of the beam and sends it to the reconstruction component; the detection light signal includes the emitted light signal after transmission or purification; the reconstruction component is connected to the light source component, the modulation component, the filtering component and the detection component respectively, determines the first and second state commands according to the target imaging mode, and obtains the super-resolution microscopic image of the sample based on the mode and the digital image, including: if it is a wide-field imaging or rotating disk confocal mode, the digital image is used as the microscopic image; If it is a structured light illumination super-resolution mode, the microscopic image is determined based on the frequency domain decoding matrix equation constructed from the digital image, including: acquiring multiple frames of digital images of the sample in multiple illumination directions and multiple preset phase shifts; converting each frame of digital images to the frequency domain through Fourier transform to obtain multiple frames of frequency domain images; constructing a frequency domain decoding matrix equation based on the frequency domain image in each direction; performing matrix inversion on the matrix equation to obtain the spectral components of each order in each direction; each spectral component includes a DC component and at least one AC component; performing spectral shift on the AC component to obtain a spectral shifted AC component; superimposing the shifted AC component with the corresponding DC component to obtain a reconstructed spectrum; merging the reconstructed spectra in all directions to obtain a complete spectrum; and converting the complete spectrum to the spatial domain through inverse Fourier transform to obtain the microscopic image. If it is a two-level joint modulation enhanced structured light illumination mode, the microscopic image is determined based on the digital image and the forward imaging model; both the first and second state commands include an in-cut command or an out-cut command.

2. The microscopic imaging system based on dual-level joint modulation according to claim 1, characterized in that, The first-stage optical field modulation component includes: a pre-connection component, an optical field encoding component, a first-stage driving component, and a 4f optical component; The pre-transition component is disposed on the outgoing optical path between the light source component and the main beam splitter, and is used to adjust the incident state of the excitation beam incident on the optical field encoding component; the incident state includes at least one of incident angle or polarization state; The optical field encoding component is disposed on the outgoing optical path of the front-end adapter component and is used to perform structured modulation on the incident excitation beam when the first state command is a cut-in command to generate an initial modulated optical field; the structured modulation includes at least one of amplitude modulation, phase modulation, spatial frequency modulation and polarization state modulation; The first-stage driving component is connected to the light field encoding component and is used to drive the light field encoding component to cut into or out of the outgoing light path between the light source component and the main beam splitter according to a first state command. When the first state command is a cut-in command, the light field encoding component is driven to perform a spatial evolution operation on the initial modulated light field to obtain a structured illumination field. The spatial evolution operation includes at least one of spatial scanning, phase shifting, or pattern replacement. The 4f optical component is disposed on the outgoing light path of the optical field encoding component and is used to transmit the incident light signal to the main beam splitter in a proportionally conjugate manner.

3. The microscopic imaging system based on dual-level joint modulation according to claim 1, characterized in that, The second-level dynamic conjugate spatial filtering component includes: a dynamic spatial filtering mask and a mask driving component; The dynamic spatial filtering mask is located on the transmission light path between the main beam splitter and the camera detection component, and is optically conjugate with the plane where the sample to be tested is located. It is used to perform physical defocusing and purification of the emitted light signal transmitted through the main beam splitter through a general mask structure when the second state command is the cut-in command. The mask driving component is connected to the dynamic spatial filter mask and is used to drive the dynamic spatial filter mask to cut into or out of the transmission optical path between the main beam splitter and the camera detection component according to the second state command. When the second state command is the cut-in command, the dynamic spatial filter mask is driven to move continuously at a preset rate so that the dynamic spatial filter mask completes at least one full field-of-view scan of the conjugate imaging field of the sample under test within a single exposure time of the camera detection component.

4. The microscopic imaging system based on dual-level joint modulation according to claim 1, characterized in that, The microscopic imaging system based on dual-level joint modulation also includes a beam shaping component; The beam shaping component is located between the light source component and the first-stage optical field modulation component, and is used to perform optical preprocessing operations on the excitation beam; the optical preprocessing operations include at least one of beam expansion operation, homogenization operation, wavefront correction operation and polarization adjustment operation.

5. The microscopic imaging system based on dual-level joint modulation according to claim 2, characterized in that, The front-end adapter assembly includes a reflector or a polarization control sub-assembly. The reflector is disposed in the outgoing light path of the light source assembly and is used to obliquely incident the excitation beam onto the light field encoding assembly at a preset off-axis tilt angle, so as to adjust the incident angle of the excitation beam. The polarization control sub-assembly includes a polarizer, a polarization beam splitter, and a waveplate assembly. The polarizer is disposed in the output optical path of the light source assembly and is used to purify the polarization of the excitation beam. The polarization beam splitter is disposed in the output optical path of the polarizer and is used to reflect the polarization-purified excitation beam to the waveplate assembly and guide the excitation beam after passing through the waveplate assembly to be perpendicularly incident on the optical field encoding assembly, so as to adjust the incident angle of the excitation beam. The waveplate assembly is disposed between the polarization beam splitter and the optical field encoding assembly and is used to rotate the polarization direction of the excitation beam reflected by the polarization beam splitter, so as to adjust the polarization state of the excitation beam.

6. The microscopic imaging system based on dual-level joint modulation according to claim 1, characterized in that, The microscopic imaging system based on dual-level joint modulation also includes a sample scanning component; The sample scanning component is used to carry and drive the sample to be tested to move; wherein, the sample scanning component includes a lateral driving component and an axial driving component; the lateral driving component is used to drive the sample to be tested to perform lateral planar movement, and the axial driving component is used to drive the sample to be tested to perform axial depth movement.

7. A microscopic imaging method based on dual-level joint modulation, characterized in that, The microscopic imaging method based on dual-level joint modulation is applied to the signal control and computational reconstruction component of the microscopic imaging system based on dual-level joint modulation as described in any one of claims 1-6, wherein the microscopic imaging method based on dual-level joint modulation includes: The target imaging mode is acquired, and a first state command and a second state command are determined based on the target imaging mode; the target imaging mode includes a structured light illumination super-resolution mode, a wide-field imaging mode, a rotating disk confocal mode, or a two-level joint modulation enhanced structured light illumination mode; Based on the first state command, the incident light signal is controlled to irradiate the sample under test, so as to excite the sample under test to generate an emitted light signal; the incident light signal includes an excitation beam or a structured light illumination field; Based on the second state command and the emitted light signal, the detection light signal of the sample to be tested is determined; the detection light signal includes the emitted light signal or the emitted light signal purified by physical defocusing. A digital image of the sample is determined based on the probe light signal of the sample. If the target imaging mode is a wide-field imaging mode or a rotating disk confocal mode, then the digital image is used as a microscopic image of the sample to be tested; If the target imaging mode is a structured light illumination super-resolution mode, then the super-resolution microscopic image of the sample to be tested is determined by the frequency domain decoding matrix equation constructed based on the digital image. If the target imaging mode is a dual-level joint modulation enhanced structured light illumination mode, then the super-resolution microscopic image of the sample to be tested is determined based on the digital image and the forward imaging model.

8. The microscopic imaging method based on dual-level joint modulation according to claim 7, characterized in that, Determining the first state command and the second state command based on the target imaging modality specifically includes: When the target imaging mode is the structured light illumination super-resolution mode, the first state command is the cut-in command and the second state command is the cut-out command; When the target imaging mode is a wide-field imaging mode, the first state command is a cut-out command, and the second state command is a cut-out command; When the target imaging mode is the turntable confocal mode, the first state command is the cut-out command, and the second state command is the cut-in command; When the target imaging mode is a dual-level joint modulation enhanced structured light illumination mode, the first state command is a cut-in command, and the second state command is a cut-in command.

9. The microscopic imaging method based on dual-level joint modulation according to claim 7, characterized in that, Based on digital images and forward imaging models, super-resolution microscopic images of the sample under test are determined, specifically including: The forward imaging model of two-level joint modulation is constructed using the following formula: ； in, This represents the forward imaging model; Super-resolution micrographs of the sample to be analyzed. The sample under test is located in a planar position. Pixel value at; Represents the two-dimensional spatial coordinates of the plane containing the sample to be tested; This indicates the output of the forward imaging model and... The corresponding predicted digital image of the j-th frame is located at the plane position of the test sample. The corresponding pixel value; This represents the camera exposure time corresponding to the acquisition of the j-th frame of the digital image of the sample under test; This indicates the position of the digital image of the j-th frame of the sample being acquired at time t, projected onto the plane containing the sample. Structured excitation light field distribution at the location; This represents the convolution operator; Indicates the position in the plane where the sample to be tested is located. The diffusion function value of the optical point at the transmitting end; This indicates the position of the dynamic spatial filter mask at time t in the plane containing the sample when acquiring the j-th frame of the digital image of the sample. The transmission function value at that location; The objective function for reconstructing the inverse problem is constructed using the following formula: ； in, The objective function for the inverse problem reconstruction is represented by M; M represents the total number of frames in the digital image. This indicates the position of the j-th frame of the digital image of the test sample in the plane containing the test sample. Pixel value at; Represents the square of the L2 norm; Represents the regularization coefficient; Represents the regularization constraint term; An optimization algorithm is used to iteratively solve the forward imaging model until the iteration termination condition is met, at which point the iteration ends and a super-resolution microscopic image of the sample to be tested is obtained. The optimization algorithm includes the alternating direction multiplier method or the fast iterative shrinkage threshold algorithm. The iteration termination condition includes the inverse problem reconstruction objective function value being less than a preset objective function threshold, or the change in the inverse problem reconstruction objective function value within a preset number of consecutive iterations being less than a preset change, or the maximum number of iterations being reached.

Images