Tapered plane three-beam lattice light sheet super-resolution microscopic imaging device and method based on spatial spectrum estimation

Abstract

The invention provides an inclined plane three-beam lattice light sheet super-resolution microscopic imaging device and method based on spatial spectrum estimation. The device comprises a light source, a cylindrical lens group, a reflector, an SLM, a lambda / 4 wave plate, a polarization splitting prism, a mask, a relay lens group, a galvanometer, a dichroscope, an objective lens, an sCMOS camera and other optical components. The SLM is used for enabling the three light beams to interfere to form a lattice light sheet so as to realize periodic illumination of the sample; the same high numerical aperture objective lens is used for exciting and collecting fluorescence, and the limitation of a traditional light sheet fluorescence microscope on sample loading is broken through; a three-stage infinite correction microscopic system is used for imaging and de-magnifying an intermediate image surface, so that distortion caused by high axial magnification and spherical aberration is avoided; a measurement matrix is constructed for a low-resolution image sequence based on a spatial spectrum estimation algorithm, singular value decomposition is performed, a signal threshold is adaptively determined by combining a noise model, fluorescence probe positioning is realized by using orthogonality of a signal and a noise subspace, and finally a high-resolution image is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of super-resolution fluorescence microscopy, in particular to a slanted-plane three-beam lattice light-sheet super-resolution microscopy device and method based on spatial spectrum estimation. BACKGROUND

[0002] Lattice light-sheet microscopy super-resolution fluorescence microscopy technology forms a quasi-non-diffracting lattice light field with two-dimensional or three-dimensional periodic structure through the coherent superposition of multiple plane waves with wave vectors on a conical surface, combines structured illumination technology with light-sheet microscopy imaging technology, and uses a light-sheet microscopy system combined with mutually perpendicular illumination and detection objectives to achieve 150nm super-resolution imaging. However, due to the limitation of the numerical aperture of the objective and the frequency of the lattice pattern, the lateral spatial resolution can only be improved by 1.3-1.5 times, which cannot meet the imaging requirements of resolution better than 150nm for the study of biological processes such as information transmission of neural synapses and regulation of immune response by exosomes. SUMMARY

[0003] The present application aims to solve the problem of low spatial resolution of lattice light-sheet imaging and proposes a slanted-plane three-beam lattice light-sheet super-resolution microscopy device and method based on spatial spectrum estimation.

[0004] The present application is achieved by the following technical solutions, and the present application proposes a slanted-plane three-beam lattice light-sheet super-resolution microscopy device based on spatial spectrum estimation, which comprises a linearly polarized circular input light beam 1, a first cylindrical lens group 2, a second cylindrical lens group 3, a first mirror 4, an SLM 5, λ a / 4 wave plate 6, a polarization beam splitter prism PBS 7, a lens L1 8, a mask 9, a second mirror 10, a first relay lens group 11, z a galvanometer 12, a second relay lens group 13, x a galvanometer 14, a third relay lens group 15, a dichroic mirror 16, an objective O1 7, a fluorescent sample 18, a filter 19, a lens L8 20, an intermediate image plane P2 21, a lens L9 22, an objective O2 23, an objective O3 24, a lens L 10 10 and an sCMOS camera 26. The linearly polarized circular input light beam 1 is first stretched along the x direction by the first cylindrical lens group 2, then compressed along the z direction by the second cylindrical lens group 3, passes through the first mirror 4, and finally forms a thin stripe illumination at the center of the SLM 5; the SLM 5 surface is loaded with a preset xz plane lattice pattern, and the light reflected from the SLM 5 passes through λThe / 4 wave plate 6 is converted into orthogonal linearly polarized light, and then vertically incident on the PBS 7, and then a Fourier diffraction pattern is formed on the plane of the mask 9 through the lens L18; the mask 9 is used to filter out invalid diffraction orders and limit the numerical aperture range of illumination, and the mask 9 is connected in series with the second mirror 10, the first relay lens group 11, z the galvanometer 12, the second relay lens group 13, x the galvanometer 14, the third relay lens group 15, the dichroic mirror 16, and the back pupil surface of the objective O117 are connected in series, so that the lattice light sheet is illuminated in the sample plane; the fluorescence excited by the fluorescent sample 18 is collected through the same high numerical aperture objective O117, separated through the dichroic mirror 16 and the filter 19, and then sequentially passes through three infinite correction microscopic systems, so that the sample focal plane is de-magnified and re-imaged on the focal plane of the sCMOS camera 26; by switching the lattice pattern loaded on the SLM 5, the intensity distribution of the front focal surface of the objective O117 is changed, the radial dynamic scanning of the lattice light sheet is realized, the dynamic scanning covers the whole light sheet illumination plane of the fluorescent sample 18, the scanning of the fluorescent sample 18 changes with time, and the cumulative intensity distribution of all illumination patterns of the SLM 5 is uniform; the sCMOS camera 26 is clock-synchronized with the SLM 5, and the image sequence generated by the fluorescent sample 18 in the illumination plane in the flashing process at each scanning position is collected; further, the volume imaging data is obtained by moving the fluorescent sample 18 to realize layer-by-layer scanning.

[0005] Further, the objective O117 and the lens L820 jointly constitute a first infinite correction microscopic system; the lens L922 and the objective O223 constitute a second infinite correction microscopic system; and the objective O324 and the lens L 10 25 constitute a third infinite correction microscopic system.

[0006] Further, the first cylindrical lens group 2 includes two cylindrical lenses and is arranged in parallel; and the second cylindrical lens group 3 includes two cylindrical lenses and is arranged in parallel.

[0007] Further, the first relay lens group 11 includes two relay lenses and is arranged in parallel; the second relay lens group 13 includes two relay lenses and is arranged in parallel; and the third relay lens group 15 includes two relay lenses and is arranged in parallel.

[0008] The application further provides a microscopic imaging method based on the spatial spectrum estimation-based oblique plane three-beam lattice light sheet super-resolution microscopic imaging device, and the method comprises the following steps: Step S1: Linearly polarized circular input light beam 1 is modulated by a computer-controlled SLM 5 to make three laser beams interfere with each other to form a lattice illumination with periodic regularity, and scanning of the lattice illumination is achieved by switching the modulation pattern on the SLM 5, so as to generate a uniform cumulative illumination pattern intensity distribution on the surface of the fluorescent sample 18; Step S2: The excitation and collection of fluorescence are achieved by using the same high numerical aperture objective O117, and the sample focal plane is de-magnified and re-imaged by using three infinite-corrected microscopic systems; Step S3: High-precision synchronous acquisition is achieved by using an sCMOS camera 26 in cooperation with the SLM 5 clock, and a low-resolution image sequence under different lattice illumination patterns is obtained; Step S4: A spatial spectrum estimation algorithm is applied to the low-resolution image sequence to reconstruct a high-resolution image, further, low-resolution volumetric imaging data is obtained by moving the fluorescent sample 18 layer by layer for scanning, and a high-resolution three-dimensional volumetric image of the fluorescent sample 18 is reconstructed based on the spatial spectrum estimation algorithm.

[0009] Further, the step S1 specifically comprises: The linearly polarized circular input light beam 1 is first stretched along the direction by the first cylindrical lens group 2, then compressed along the direction by the second cylindrical lens group 3, and finally forms a thin stripe illumination at the center of the SLM 5 through the first mirror 4; the SLM 5 is loaded with a preset x plane lattice pattern to make three laser beams interfere with each other to generate a lattice illumination field with periodic regularity, and spatial decoupling of incident light and reflected light is achieved by polarization light path separation technology, followed by filtering out invalid diffraction orders by the lens L18 and the mask 9. z xz Further, the step S2 specifically comprises: The fluorescent signal excited by the lattice light sheet in the sample plane is collected by the same high numerical aperture objective O117, and then filtered to remove stray light by the dichroic mirror 16 and the optical filter 19, the objective O117 and the lens L820 together constitute a first infinite-corrected microscopic system to generate an intermediate image plane P221, then the lens L922 and the objective O223 constitute a second infinite-corrected microscopic system to de-magnify the intermediate image plane P221, finally, the fluorescent sample 18 is re-imaged by a third infinite-corrected microscopic system composed of the objective O324 and the lens L 10 25 to obtain a low-resolution image thereof.

[0010] Further, the step S3 specifically comprises:

[0011] Further, the step S3 specifically comprises: ​By switching the lattice pattern loaded on the SLM 5, the radial dynamic scanning of the lattice light sheet in the sample plane is realized, which covers the target illumination plane of the entire fluorescent sample 18, and the cumulative intensity distribution of all illumination patterns is uniform, finally, the sCMOS camera 26 is clocked with the SLM 5, and the low-resolution image sequence generated by the fluorescent sample 18 in the illumination plane in the flashing process at each scanning position is collected.

[0012] Further, in step S4, the low-resolution image sequence is described in the form of a mathematical matrix, and the matrix formula of the intensity distribution of each image is combined into a matrix I , and the covariance matrix thereof is calculated R . (1) , wherein is a M-dimensional mapping vector matrix, is a M-dimensional signal covariance matrix, is a M-dimensional noise covariance matrix. Eigenvalue decomposition is performed on the matrix , to obtain (2) , wherein is a N-dimensional imaging data space basis vector matrix, is a N-dimensional diagonal matrix composed of eigenvalues.

[0013] Further, in step S4, according to the imaging system noise model, the number of signal eigenvalues is obtained by maximum likelihood estimation to determine the best signal-to-noise ratio threshold σ0 of any point in the field of view; whether the eigenvalue is greater than the corresponding eigenvector of σ0 is used to distinguish whether it is a signal, to obtain the signal and noise subspaces, and the signal subspace and the noise subspace have orthogonality, the spectral peak search is performed by using the orthogonal characteristics of the two subspaces, to obtain a high-resolution image; further, by moving the fluorescent sample 18, a layer-by-layer scanning is realized to obtain low-resolution volumetric imaging data, and based on the spatial spectrum estimation algorithm, a high-resolution three-dimensional volumetric image of the fluorescent sample 18 is reconstructed.

[0014] The beneficial effects of the present application are: 1. The same numerical aperture objective is used to excite and collect fluorescence, which can break through the fixed limitation of the sample type (glass slide, etc.) and loading space of the light sheet fluorescence microscope, and solve the technical problem that the orthogonal light sheet fluorescence microscope cannot use a high numerical aperture objective.

[0015] ​2. The use of a three-stage infinity-corrected microscope system to image and re-magnify the intermediate image overcomes the defocus problem, so that it will not be distorted due to high axial magnification and related spherical aberration away from the image plane.

[0016] 3. According to the threshold value, the signal and noise subspaces are divided, and the high-resolution image is obtained by using the projection ratio of the mapping vector in the mutually orthogonal signal subspace and noise subspace; 4. According to the noise model of the imaging system, the number of signal eigenvalues is obtained by maximum likelihood estimation to determine the best signal-to-noise ratio threshold σ0 of any point in the field of view, without selecting the threshold to avoid the influence of subjective factors. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present application, and those skilled in the art can obtain other drawings according to the provided drawings without creative labor.

[0018] Figure 1 It is a structure schematic diagram of a slanted plane three-beam lattice light sheet super-resolution microscopic imaging device based on spatial spectrum estimation; Figure 2 It is a three-beam lattice light sheet pattern used for illuminating a fluorescent sample on the focal plane of an objective lens; Figure 3 It is a contrast-frame number result schematic diagram of a slanted plane four-beam lattice light sheet super-resolution microscopic imaging based on spatial spectrum estimation.

[0019] In the drawings, the reference signs are explained as follows: 1, linearly polarized circular input light beam; 2, first cylindrical lens group; 3, second cylindrical lens group; 4, first mirror; 5, SLM; 6, λ / 4 wave plate; 7, PBS; 8, lens L1; 9, mask; 10, second mirror; 11, first relay lens group; 12, z galvanometer; 13, second relay lens group; 14, x galvanometer; 15, third relay lens group; 16, dichroic mirror; 17, objective lens O1; 18, fluorescent sample; 19, filter; 20, lens L8; 21, intermediate image plane P2; 22, lens L9; 23, objective lens O2; 24, objective lens O3; 25, lens L 10 ; 26, sCMOS camera. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application.

[0021] The present application proposes a method for oblique plane three-beam lattice light sheet super-resolution microscopic imaging based on spatial spectrum estimation, which combines the spatial spectrum estimation algorithm with the oblique plane lattice light sheet fluorescence microscopic imaging technology. Under the modulation of a computer-controlled SLM 5, a lattice light sheet illumination field is generated by three-beam laser interference, a measurement data matrix is constructed based on the obtained low-resolution image sequence, and singular value decomposition is performed on the matrix. Further, the sample space is divided into a signal subspace and a noise subspace, and the precise search of the fluorescence probe position is realized based on the orthogonality of the two subspaces, so as to reconstruct a high-resolution image. The same high numerical aperture objective lens O1 17 is used to realize the excitation and collection of the fluorescence of the fluorescent sample 18, the SLM 5 modulation is used to realize the movement scanning of the lattice light sheet with time, the low-resolution volumetric imaging data is obtained by layer-by-layer scanning, and the spatial spectrum estimation algorithm is used to process to obtain a high-resolution three-dimensional volumetric image of the fluorescent sample 18.

[0022] Specifically, in combination with Figures 1-3 , the present application proposes an oblique plane three-beam lattice light sheet super-resolution microscopic imaging device based on spatial spectrum estimation, Figure 1 which is a structural schematic diagram of the microscopic imaging device of the present embodiment. The microscopic imaging device comprises a linearly polarized circular input light beam 1, a first cylindrical lens group 2, a second cylindrical lens group 3, a first mirror 4, an SLM 5, λ a / 4 wave plate 6, a polarization beam splitter prism PBS 7, a lens L1 8, a mask 9, a second mirror 10, a first relay lens group 11, z a galvanometer 12, a second relay lens group 13, x a galvanometer 14, a third relay lens group 15, a dichroic mirror 16, an objective lens O1 17, a fluorescent sample 18, a filter 19, a lens L8 20, an intermediate image plane P2 21, a lens L9 22, an objective lens O2 23, an objective lens O3 24, a lens L 10 25 and an sCMOS camera 26. The linearly polarized circular input light beam 1 is first stretched along the x direction by the first cylindrical lens group 2, and then compressed along the z direction by the second cylindrical lens group 3, passes through the first mirror 4, and finally forms a thin stripe illumination at the center of the SLM 5; the SLM 5 surface is loaded with a preset xz plane lattice pattern, and the light reflected from the SLM 5 passes through λThe / 4 wave plate 6 converts the orthogonal linearly polarized light into vertical incident PBS 7, and then forms a Fourier diffraction pattern on the plane of the mask 9 through the lens L18; the mask 9 is used to filter out invalid diffraction orders and limit the numerical aperture range of illumination, and the mask 9 is connected in series with the second mirror 10, the first relay lens group 11, z the galvanometer 12, the second relay lens group 13, x the galvanometer 14, the third relay lens group 15, the dichroic mirror 16, and the back pupil plane of the objective O117 are connected in series, so that the lattice light sheet is illuminated on the sample plane; the fluorescence excited by the fluorescent sample 18 is collected by the same high numerical aperture objective O117, separated through the dichroic mirror 16 and the filter 19, and then sequentially passes through three infinite correction microscopic systems to demagnify and reimage the sample focal plane on the focal plane of the sCMOS camera 26; by switching the lattice pattern loaded on the SLM 5, the intensity distribution of the front focal plane of the objective O117 is changed, the radial dynamic scanning of the lattice light sheet is realized, the dynamic scanning covers the whole light sheet illumination plane of the fluorescent sample 18, the fluorescent sample 18 is scanned over time, and the cumulative intensity distribution of all illumination patterns of the SLM 5 is uniform; the sCMOS camera 26 is clocked synchronously with the SLM 5, and the image sequence generated by the fluorescent sample 18 in the illumination plane in the flashing process at each scanning position is collected; further, the volume imaging data is obtained by moving the fluorescent sample 18 to realize layer-by-layer scanning.

[0023] The objective O117 and the lens L820 jointly constitute a first infinite correction microscopic system; the lens L922 and the objective O223 constitute a second infinite correction microscopic system; and the objective O324 and the lens L 10 25 constitute a third infinite correction microscopic system.

[0024] The first cylindrical lens group 2 includes two cylindrical lenses and is arranged in parallel; and the second cylindrical lens group 3 includes two cylindrical lenses and is arranged in parallel.

[0025] The first relay lens group 11 includes two relay lenses and is arranged in parallel; the second relay lens group 13 includes two relay lenses and is arranged in parallel; and the third relay lens group 15 includes two relay lenses and is arranged in parallel.

[0026] The application further provides a microscopic imaging method based on the spatial spectrum estimation-based oblique plane three-beam lattice light sheet super-resolution microscopic imaging device, and the method comprises the following steps: Step S1: The linearly polarized circular input light beam 1 is modulated through the computer-controlled SLM 5, so that three laser beams interfere with each other to form a lattice illumination with periodicity; by switching the modulation pattern on the SLM 5, the scanning of the lattice illumination is realized, so that a uniform cumulative illumination pattern intensity distribution is generated on the surface of the fluorescent sample 18. Step S2: The same high numerical aperture objective lens O117 is used to excite and collect fluorescence, and three infinity-corrected microscope systems are used to demagnetize and re-image the focal plane of the sample. Step S3: Use the sCMOS camera 26 in conjunction with the SLM5 clock to achieve high-precision synchronous acquisition and obtain low-resolution image sequences under different lattice illumination patterns; Step S4: Apply a spatial spectrum estimation algorithm to the low-resolution image sequence to reconstruct a high-resolution image. Further, move the fluorescent sample 18 to achieve layer-by-layer scanning to obtain low-resolution volume imaging data, and reconstruct a high-resolution three-dimensional volume image of the fluorescent sample 18 based on the spatial spectrum estimation algorithm.

[0027] Step S1 specifically includes: Linearly polarized circular input beam 1 first passes through the first cylindrical lens group 2 along... x directional stretching, then passing through the second cylindrical lens group 3 along z The directional compression, after passing through the first reflector 4, ultimately forms a thin stripe of illumination at the center of the SLM5; the surface of the SLM5 is loaded with a preset... xz A planar lattice pattern is used to generate a periodic lattice illumination field by interfering with the three laser beams. Spatial decoupling of incident and reflected light is achieved through polarization path separation technology. Invalid diffraction levels are then filtered out by lens L18 and mask 9.

[0028] Step S2 specifically includes: The fluorescence signal generated by the lattice light plate in the sample plane is collected by the same high numerical aperture objective lens O117, and stray light is filtered out sequentially by a dichroic mirror 16 and a filter 19. The objective lens O117 and lens L820 together constitute the first infinity correction microscopy system, generating the intermediate image plane P221. Subsequently, lens L922 and objective lens O223 constitute the second infinity correction microscopy system to demagnetize the intermediate image plane P221. Finally, the image is filtered through objective lens O324 and lens L923. 10 The third infinity-corrected microscopy system, consisting of 25, re-images the fluorescent sample 18 to obtain a low-resolution image.

[0029] Step S3 specifically includes: By switching the lattice pattern loaded on the SLM5, the lattice light sheet is dynamically scanned radially in the sample plane. The dynamic scan covers the entire target illumination plane of the fluorescent sample 18, and the cumulative intensity distribution of all illumination patterns is uniform. Finally, the sCMOS camera 26 is synchronized with the clock of the SLM5 to acquire a low-resolution image sequence generated by the fluorescent sample 18 in the illuminated plane during the flashing process at each scanning position.

[0030] In step S4, the low-resolution image sequence is described in the form of a mathematical matrix, and the matrix formulas of the intensity distribution of each frame image are combined into a matrix. I Then calculate its covariance matrix. R : (1) In the formula for dimensional mapping vector matrix, for 3D signal covariance matrix, for 3D noise covariance matrix; For matrix Perform eigenvalue decomposition to obtain (2) In the formula for 3D imaging data spatial basis vector matrix, Composed of eigenvalues A diagonal matrix.

[0031] In step S4, based on the imaging system noise model, the number of signal eigenvalues ​​is obtained through maximum likelihood estimation to determine the optimal signal-to-noise ratio threshold σ0 for any point in the field of view; whether a eigenvalue is greater than the eigenvector corresponding to σ0 is used to distinguish whether it is a signal, thus obtaining signal and noise subspaces. The signal subspace and noise subspace are orthogonal, and the orthogonality of the two subspaces is used to perform spectral peak search to obtain a high-resolution image; furthermore, by moving the fluorescent sample 18, layer-by-layer scanning is achieved to obtain low-resolution volumetric imaging data, and a high-resolution three-dimensional volumetric image of the fluorescent sample 18 is reconstructed based on the spatial spectrum estimation algorithm.

[0032] Figure 2 Showing Figure 1 The microscopic imaging apparatus shown illuminates the lattice pattern of the fluorescent sample 18 within the focal plane of objective lens O117. xz The cross-section visually reflects the light field modulation effect achieved by the device on the excitation side. Figure 3is the contrast-frame number result of three-beam lattice light sheet super-resolution microscopic imaging based on spatial spectrum estimation. The results show that when the number of reconstructed image frames is greater than 100, the resolution is better than 100 nm, which breaks through the limit of the traditional diffraction limit. Unlike the conventional multi-angle imaging method which depends on additional rotating optical path, this method can realize the periodic distribution of complex light field at the excitation end through the fine regulation of the lattice light sheet, and realize super-resolution imaging combined with spatial spectrum estimation reconstruction, which not only effectively improves the lateral resolution, but also has advantages in maintaining low light toxicity and large volume imaging capability. This optical path structure breaks through the limitation of traditional light sheet fluorescence microscope in sample type (such as glass slide loading) and loading space due to its efficient defocusing correction and dynamic scanning capability, and solves the technical bottleneck that orthogonal light sheet microscope cannot use high numerical aperture objective. In summary, this method, through the fine light field regulation of lattice light sheet and the super-resolution reconstruction algorithm based on spatial spectrum estimation, has important significance for high-resolution three-dimensional long-term observation of living samples.

[0033] The above describes the device and method of oblique plane three-beam lattice light sheet super-resolution microscopic imaging based on spatial spectrum estimation in detail, and the principle and implementation mode of the present application are described by applying specific examples; the above examples are only used to help understand the method and core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed, and the above description should not be understood as limiting the present application.

Claims

1. A device for oblique plane three-beam-lattice light sheet super-resolution microscopic imaging based on spatial spectrum estimation, characterized in that, The microscopic imaging device comprises a linearly polarized circular input light beam (1), a first cylindrical lens group (2), a second cylindrical lens group (3), a first mirror (4), an SLM (5), λ a 1 / 4 wave plate (6), a polarization beam splitter prism PBS (7), a lens L1 (8), a mask (9), a second mirror (10), a first relay lens group (11), z a galvanometer (12), a second relay lens group (13), x a galvanometer (14), a third relay lens group (15), a dichroic mirror (16), an objective O1 (17), a fluorescent sample (18), a filter (19), a lens L8 (20), an intermediate image plane P2 (21), a lens L9 (22), an objective O2 (23), an objective O3 (24), a lens L 10 (25) and an sCMOS camera (26). The linearly polarized circular input beam (1) first passes through the first cylindrical lens group (2) along... x directional stretching, then passing through the second cylindrical lens group (3) along z Directional compression, passing through the first reflector (4), finally forms a thin stripe of illumination at the center of the SLM (5); the surface of the SLM (5) is loaded with a preset... xz Planar lattice pattern, light reflected from SLM(5) passes through λ After the 4-wave plate (6) converts the light into orthogonally linearly polarized light, it is incident perpendicularly on the PBS (7), and then forms a Fourier diffraction pattern on the mask (9) plane through the lens L1 (8); the mask (9) is used to filter out invalid diffraction orders and limit the numerical aperture range of the illumination. The mask (9) is connected to the second mirror (10) and the first relay lens group (11). z Galvanometer (12), second relay lens group (13) x The galvanometer (14), the third relay lens group (15), the dichroic mirror (16), and the objective lens O1 (17) are connected in series to illuminate the lattice light sheet in the sample plane. The fluorescence emitted by the fluorescent sample (18) is collected by the same high numerical aperture objective lens O1 (17), separated by the dichroic mirror (16) and the filter (19), and then passed through three infinity-corrected microscopy systems to demagnetize the sample focal plane and re-image it onto the focal plane of the sCMOS camera (26). By switching the lattice pattern loaded on the SLM (5), the objective lens is changed. The intensity distribution of the front focal plane of O1 (17) enables radial dynamic scanning of the lattice light sheet. The dynamic scanning covers the entire light sheet illumination plane of the fluorescent sample (18), and the fluorescent sample (18) is scanned over time. The cumulative intensity distribution of all illumination patterns of SLM (5) is uniform. The sCMOS camera (26) is synchronized with the clock of SLM (5) to acquire the image sequence generated by the fluorescent sample (18) in the illuminated plane at each scanning position during the flashing process. Furthermore, by moving the fluorescent sample (18), layer-by-layer scanning is achieved to obtain volume imaging data.

2. The microscopic imaging device of claim 1, wherein, The objective O1 (17) and the lens L8 (20) jointly constitute a first infinity- corrected microscopic system; the lens L9 (22) and the objective O2 (23) constitute a second infinity- corrected microscopic system; the objective O3 (24) and the lens L 10 (25) constitute a third infinity-corrected microscopic system.

3. The microscopic imaging device of claim 1, wherein, The first cylindrical lens group (2) comprises two cylindrical lenses and is arranged in parallel; the second cylindrical lens group (3) comprises two cylindrical lenses and is arranged in parallel.

4. The microscopic imaging device of claim 1, wherein, The first relay lens group (11) comprises two relay lenses and is arranged in parallel; the second relay lens group (13) comprises two relay lenses and is arranged in parallel; and the third relay lens group (15) comprises two relay lenses and is arranged in parallel.

5. A microscopic imaging method based on the spatial-spectrum-estimation-based oblique-plane three-beam-lattice-light-sheet super-resolution microscopic imaging apparatus according to claim 1, characterized in that, The method comprises the following steps: Step S1: a linearly polarized circular input light beam (1) is modulated via a computer-controlled SLM (5) to make three laser beams interfere with each other to form lattice illumination with periodicity, scanning of the lattice illumination is realized by switching the modulation pattern on the SLM (5), and a uniform cumulative illumination pattern intensity distribution is generated on the surface of a fluorescent sample (18); Step S2: excitation and collection of fluorescence are realized by using the same high numerical aperture objective O1 (17), and a sample focal plane is de-magnified and re-imaged by using three infinite correction microscopic systems; Step S3: high-precision synchronous acquisition is realized by using an sCMOS camera (26) in cooperation with a clock of the SLM (5), and a low-resolution image sequence under different lattice illumination patterns is obtained; Step S4: a spatial spectrum estimation algorithm is applied to the low-resolution image sequence to reconstruct a high-resolution image, further, low-resolution volumetric imaging data are acquired by layer-by-layer scanning of the fluorescent sample (18), and a high-resolution three-dimensional volumetric image of the fluorescent sample (18) is reconstructed based on the spatial spectrum estimation algorithm.

6. The method of microscopic imaging according to claim 5, wherein, The step S1 specifically comprises: The linearly polarized circular input beam (1) first passes through the first cylindrical lens group (2) along... x directional stretching, then passing through the second cylindrical lens group (3) along z Directional compression, after passing through the first reflector (4), finally forms a thin stripe of illumination at the center of the SLM (5); the surface of the SLM (5) is loaded with a preset... xz Planar lattice pattern is used to generate a periodic lattice illumination field by interfering with the three laser beams. Spatial decoupling of incident and reflected light is achieved by polarization path separation technology. Then, invalid diffraction levels are filtered out by lens L1 (8) and mask (9).

7. The method of microscopic imaging according to claim 5, wherein, The step S2 specifically comprises: The lattice light sheet excites fluorescence signal generated in the sample plane, which is collected by the same high numerical aperture objective O1 (17), and in turn filtered by a dichroic mirror (16) and a filter (19) to remove stray light. The objective O1 (17) and a lens L8 (20) together constitute a first infinity-corrected microscope system to generate an intermediate image plane P2 (21). Subsequently, a lens L9 (22) and an objective O2 (23) constitute a second infinity-corrected microscope system to de-magnify the intermediate image plane P2 (21). Finally, a third infinity-corrected microscope system constituted by an objective O3 (24) and a lens L 10 (25) reimages the fluorescent sample (18) to obtain a low-resolution image thereof.

8. The method of microscopic imaging according to claim 5, wherein, The step S3 specifically comprises: By switching the lattice pattern loaded on the SLM (5), radial dynamic scanning of the lattice light sheet in the sample plane is realized, the dynamic scanning covers the entire target illumination plane of the fluorescent sample (18), and the cumulative intensity distribution of all illumination patterns is uniform, finally, the sCMOS camera (26) is synchronized with the clock of the SLM (5), and a low-resolution image sequence generated by the fluorescent sample (18) in the illuminated plane in the flashing process at each scanning position is collected.

9. The method of microscopic imaging according to claim 5, wherein, In step S4, the sequence of low resolution images is described in the form of a mathematical matrix, combining the matrix formula of the intensity distribution of each image frame into a matrix I , whose covariance matrix is then calculated R : （1） wherein is a M x N mapping vector matrix, is a M x M signal covariance matrix, is a M x M noise covariance matrix; on the matrix performing eigenvalue decomposition to obtain （2） wherein is is is a matrix of dimension is a diagonal matrix of dimension 10. The method of microscopic imaging according to claim 9, wherein, In step S4, according to an imaging system noise model, the number of signal eigenvalues is obtained by maximum likelihood estimation to determine the best signal-to-noise ratio threshold σ0 of any point in the field of view; whether it is a signal is distinguished according to whether the eigenvalue is greater than the eigenvector corresponding to σ0, a signal subspace and a noise subspace are obtained, the signal subspace and the noise subspace have orthogonality, spectral peak search is performed by using the orthogonal characteristics of the two subspaces, and a high-resolution image is obtained; further, low-resolution volumetric imaging data are acquired by layer-by-layer scanning of the fluorescent sample (18), and a high-resolution three-dimensional volumetric image of the fluorescent sample (18) is reconstructed based on the spatial spectrum estimation algorithm.

Images