A microscope three-dimensional sample image reconstruction method and microscope three-dimensional reconstruction device

Abstract

The present application relates to the technical field of three-dimensional reconstruction, and in particular to a microscope three-dimensional sample image reconstruction method and a microscope three-dimensional reconstruction device. The present application first obtains a two-dimensional image sequence of the sample at different depths through a microscopic imaging module, and simultaneously uses an inclination detection module to measure the X-axis inclination θ of each layer in real time. n The inclination angles θ of the x and y axes n y, then, based on the measured inclination, the image is corrected using a matrix transformation method. The transformation process includes first performing X-axis rotation correction, then Y-axis rotation correction, and finally obtaining the corrected three-dimensional coordinates. Finally, the corrected multi-layer images are rearranged according to depth position to complete accurate three-dimensional reconstruction; the real-time detection and compensation technology of inclination is introduced into microscopic three-dimensional reconstruction, which improves the reconstruction accuracy of three-dimensional images.

Description

Technical Field

[0001] The present application relates to the technical field of three-dimensional reconstruction, and in particular to a microscope three-dimensional sample image reconstruction method and a microscope three-dimensional reconstruction device. Background Art

[0002] In microscopy systems, 3D reconstruction typically uses planar imaging. This involves moving the sample or objective lens axially to acquire multiple 2D images at different depths. These images are then superimposed to achieve 3D reconstruction. This approach has been widely used for 3D imaging of biological samples.

[0003] Prior art, such as a three-dimensional microscope imaging method disclosed in the Chinese invention patent application document with publication number CN114964041A, refers to Figure 1 , Figure 1 The figure below is a flow chart of a conventional 3D reconstruction method. The conventional method uses Z-axis scanning to obtain a series of images and then performs 3D reconstruction. This 3D reconstruction scheme only considers the acquisition and superposition of images in the Z-axis direction, ignoring the influence of factors such as sample gravity, inherent system errors, optical path system errors, and environmental jitter during the actual imaging process. Consequently, each acquired image layer is not strictly parallel. During the 3D reconstruction process, the inclination angle of each layer will randomly vary, resulting in low accuracy of the final reconstructed 3D image. This situation needs further improvement. Summary of the Invention

[0004] To address the problem that existing 3D reconstruction methods ignore the fact that each layer of image acquired during the actual imaging process is not strictly parallel, resulting in low accuracy of the final reconstructed 3D image, the present application provides a microscope 3D sample image reconstruction method and a microscope 3D reconstruction device, which adopt the following technical solutions:

[0005] In a first aspect, the present application provides a method for reconstructing a three-dimensional sample image from a microscope, which is applied to a three-dimensional reconstruction device. The three-dimensional reconstruction device includes a microscopic imaging module and a tilt detection module. The method includes the following steps:

[0006] Control the microscopic imaging module to obtain scanning images of multiple layers of the sample to obtain a series of two-dimensional images M1 to M n， , n is a positive integer;

[0007] The sample inclination angle corresponding to each layer of image is measured by the inclination detection module to obtain the x-axis inclination angle θ of each layer. n The inclination angles θ of the x and y axes n y;

[0008] According to the x-axis inclination angle θ n The inclination angles θ of the x and y axes ny, for the image M at the corresponding level n Perform coordinate transformation to obtain the corrected three-dimensional coordinates (x n ',y n ',z n ');

[0009] Based on the corrected three-dimensional coordinates, the multi-layer images are three-dimensionally reconstructed to obtain a three-dimensional image of the sample.

[0010] By adopting the above technical solution, due to the influence of factors such as sample gravity, system error, and environmental jitter, the sample surface will randomly tilt relative to the objective lens during Z-axis scanning. This tilt will change with the change of scanning depth. For example, when the scanning depth is 10 microns, the sample may tilt 2 degrees toward the X-axis, and when the scanning depth increases to 20 microns, it may tilt 1.5 degrees toward the Y-axis. This tilt that changes with depth will cause the reconstructed three-dimensional image to be misaligned, seriously affecting the reconstruction accuracy. The present application first obtains a two-dimensional image sequence of the sample at different depths through a microscopic imaging module, and uses a tilt detection module to measure the X-axis tilt angle θ of each layer in real time. n The inclination angles θ of the x and y axes n y, then, based on the measured inclination, the image is corrected using a matrix transformation method. The transformation process includes first performing X-axis rotation correction, then Y-axis rotation correction, and finally obtaining the corrected three-dimensional coordinates. Finally, the corrected multi-layer images are rearranged according to depth position to complete accurate three-dimensional reconstruction; the real-time detection and compensation technology of inclination is introduced into microscopic three-dimensional reconstruction, which improves the reconstruction accuracy of three-dimensional images.

[0011] Optionally, the tilt detection module includes a detection plane, on which a reference position mark point is provided; the tilt detection module is used to measure the sample tilt angle corresponding to each layer of the image to obtain the x-axis tilt angle θ of each layer. n The inclination angles θ of the x and y axes n y, specifically including the following steps:

[0012] Emitting detection light to the sample surface;

[0013] Collect the detection light reflected by the sample to obtain a detection light spot;

[0014] Calculating the offset of the center of mass of the detection light spot relative to the reference position mark point to obtain a lateral offset Δx and a longitudinal offset Δy;

[0015] According to the lateral offset Δx and the longitudinal offset Δy, the x-axis inclination angle θ of the sample is determined. n The inclination angles θ of the x and y axes n y.

[0016] By adopting the above technical solution, since traditional inclination measurement methods, such as mechanical angle sensors or inclinometers, are often large in size and difficult to integrate into microscope systems, and their measurement accuracy and response speed are also difficult to meet the requirements of microscopic imaging; the present application uses the principle of light beam reflection to calculate the inclination angle of the sample by detecting the slight offset of the reflected light spot position; first, a detection plane is set in the microscope system, and a reference point is marked on the detection plane as a reference position. When the emitted detection light is irradiated on the sample surface, the reflected light will form a light spot on the detection plane. If the sample is tilted, the position of the reflected light spot will be offset relative to the reference point. By calculating the lateral offset Δx and longitudinal offset Δy of the center of mass of the light spot relative to the reference point, and then based on the optical geometric relationship, the inclination angle θ of the sample in the X-axis and Y-axis directions can be accurately calculated. n x and θ n y, the spot centroid calculation method is used to significantly improve the accuracy of angle measurement.

[0017] Optionally, before emitting the detection light to the sample surface, the method further includes the following steps:

[0018] Set multiple calibration inclination angles θ1x to θ on the adjustable inclination platform M x, M are positive integers;

[0019] Measure the position offset Δx1 to Δx corresponding to the calibration inclination M ;

[0020] Based on the calibrated inclination angle and the position offset, establishing an x-axis inclination angle response function θx=f(Δx);

[0021] Set multiple calibration inclination angles θ1y to θ on the adjustable inclination platform M y;

[0022] Measure the position offset Δy1 to Δy corresponding to the calibration inclination M ;

[0023] Based on the calibrated inclination angle and the position offset, establishing a y-axis inclination angle response function θy=f(Δy);

[0024] The x-axis tilt angle θ of the sample is determined according to the lateral offset Δx, the longitudinal offset Δy, the x-axis tilt angle response function and the y-axis tilt angle response function. n The inclination angles θ of the x and y axes n y.

[0025] By adopting the above technical solution, due to factors such as inherent aberrations in the optical system, nonlinear response of the detector, and errors in the mechanical structure, the spot displacement and the actual inclination angle are not a simple linear correspondence; this application first uses a high-precision adjustable inclination platform to set a series of known calibration inclination angles in the X-axis and Y-axis directions, and then measures the corresponding spot displacement for each calibration inclination angle to obtain a large number of calibration data point pairs. Based on these discrete data points, the inclination response functions θx=f(Δx) and θy=f(Δy) of the X-axis and Y-axis are established respectively through mathematical fitting methods. In actual measurement, the detected spot displacement is substituted into these response functions to accurately calculate the actual inclination angle of the sample; the measurement accuracy of the system is improved by fitting a large number of calibration points; and the independent calibration of the X-axis and Y-axis directions is taken into account to make the measurement results more accurate.

[0026] Optionally, the coordinate transformation is expressed in matrix form, where:

[0027] The x-axis rotation matrix is: The y-axis rotation matrix is:

[0028] By adopting the above technical solution, the present application first constructs an X-axis rotation matrix, the matrix elements of which are composed of trigonometric functions of the inclination angle θnx, in the form of a 3×3 orthogonal matrix; similarly, a Y-axis rotation matrix is ​​constructed, the elements of which are composed of trigonometric functions of the inclination angle θny. For any point in space, the X-axis rotation transformation is performed first, and then the Y-axis rotation transformation is performed. The corrected spatial coordinates can be obtained through matrix multiplication.

[0029] Optionally, according to the x-axis inclination angle θ n The inclination angles θ of the x and y axes n y, for the image M at the corresponding level n Perform coordinate transformation to obtain the corrected three-dimensional coordinates (x n ',y n ',z n '), specifically comprising the following steps:

[0030] Based on the x-axis inclination angle θ n x and x-axis rotation matrix, perform x-axis rotation transformation on the image to obtain the first transformation coordinate; based on the y-axis inclination angle θ n y and y-axis rotation matrix, perform y-axis rotation transformation on the first transformed coordinates to obtain the corrected three-dimensional coordinates (x n ',y n ',z n ').

[0031] Optionally, based on the x-axis inclination angle θ nx and x-axis rotation matrix, perform x-axis rotation transformation on the image to obtain the first transformation coordinates, specifically including the following steps:

[0032] Get the initial coordinates (x, y, z);

[0033] Based on the x-axis inclination angle θ n x and x-axis rotation matrices transform the initial coordinates (x, y, z) to obtain the first transformed coordinates (x1, y1, z1), where x1 = x; y1 = ycosθ n x-zsinθ n x; z1 = ysinθ n x+zcosθ n x.

[0034] Optionally, based on the y-axis inclination angle θ n y and y-axis rotation matrix, perform y-axis rotation transformation on the first transformed coordinates to obtain the corrected three-dimensional coordinates (x n ',y n ',z n '), specifically comprising the following steps:

[0035] Get the first transformation coordinates (x1, y1, z1);

[0036] Based on the y-axis inclination angle θ n y and y-axis rotation matrix, perform y-axis rotation transformation on the first transformed coordinate (x1, y1, z1) to obtain the corrected three-dimensional coordinate (x n ',y n ',z n '),

[0037] Among them, x n '=x1cosθ n y+z1sinθ n y;y n '=y1;z n '=-x1sinθ n y+z1cosθ n y.

[0038] Optionally, each layer image M n Corresponding to a depth position z n Based on the corrected three-dimensional coordinates, the multi-layer image is three-dimensionally reconstructed to obtain a three-dimensional image of the sample, which specifically includes the following steps:

[0039] The corrected three-dimensional coordinates (x n ',y n ',z n ')According to different z nThe values ​​are arranged to obtain a calibrated three-dimensional image.

[0040] By adopting the above technical solution, the present application first assigns each layer image Mn its corresponding depth position identifier z n , this depth value is the actual position data recorded by the displacement stage during the image acquisition process, and then the image coordinates (x n ',y n ',z n ') According to its depth value z n The arrangement ensures the accurate positioning of each layer of image in the Z-axis direction, while maintaining the precise position after correction in the XY plane, realizing the three-dimensional reconstruction of multi-layer images.

[0041] In a second aspect, the present application provides a microscope three-dimensional reconstruction device, the device comprising:

[0042] Microscopic imaging module, used to obtain scanning images of multiple layers of the sample;

[0043] Inclination detection module, including:

[0044] A detection light source, used for emitting detection light;

[0045] A detection plane, wherein a reference position marking point is provided on the detection plane;

[0046] a detector, disposed on the detection plane, for receiving the detection light reflected by the sample;

[0047] a dichroic mirror, configured to reflect the probe light and transmit visible light;

[0048] A processor, configured to execute the microscope three-dimensional sample image reconstruction method according to any one of claims 1 to 8;

[0049] Memory, used to store program instructions and data.

[0050] Optionally, the device further includes:

[0051] A sample stage, used for moving in the z-axis direction to acquire sample images at different depths;

[0052] Adjustable inclination platform for setting multiple calibration inclination angles during the calibration process;

[0053] The display unit is used to display the reconstructed three-dimensional image.

[0054] In summary, this application includes at least one of the following beneficial technical effects:

[0055] 1. This application first uses a microscopic imaging module to obtain a two-dimensional image sequence of the sample at different depths, and uses a tilt detection module to measure the X-axis tilt angle θ of each layer in real time. n The inclination angles θ of the x and y axes n y, then, based on the measured inclination angle, the image is corrected using a matrix transformation method. The transformation process includes first performing X-axis rotation correction, then Y-axis rotation correction, and finally obtaining the corrected three-dimensional coordinates. Finally, the corrected multi-layer images are rearranged according to depth position to complete accurate three-dimensional reconstruction. The real-time detection and compensation technology of inclination angle is introduced into microscopic three-dimensional reconstruction, which improves the reconstruction accuracy of three-dimensional images.

[0056] 2. Traditional inclination angle measurement methods, such as mechanical angle sensors or inclinometers, are often large and difficult to integrate into microscope systems. At the same time, their measurement accuracy and response speed cannot meet the requirements of microscopic imaging. The present application uses the principle of light beam reflection to calculate the inclination angle of the sample by detecting the slight offset of the reflected light spot position. First, a detection plane is set in the microscope system, and a reference point is marked on the detection plane as a reference position. When the emitted detection light is irradiated on the sample surface, the reflected light forms a light spot on the detection plane. If the sample is tilted, the position of the reflected light spot will be offset relative to the reference point. By calculating the lateral offset Δx and longitudinal offset Δy of the center of mass of the light spot relative to the reference point, and then based on the optical geometric relationship, the inclination angle θ of the sample in the X-axis and Y-axis directions can be accurately calculated. n x and θ n y, the spot centroid calculation method is used to significantly improve the accuracy of angle measurement;

[0057] 3. Due to factors such as inherent aberrations in the optical system, nonlinear response of the detector, and errors in the mechanical structure, the relationship between the light spot displacement and the actual inclination angle is not a simple linear correspondence. This application first uses a high-precision adjustable inclination platform to set a series of known calibration inclination angles in the X-axis and Y-axis directions, and then measures the corresponding light spot displacement for each calibration inclination angle to obtain a large number of calibration data point pairs. Based on these discrete data points, the inclination response functions θx=f(Δx) and θy=f(Δy) of the X-axis and Y-axis are established respectively through mathematical fitting methods. In actual measurement, the detected light spot displacement is substituted into these response functions to accurately calculate the actual inclination angle of the sample. The measurement accuracy of the system is improved by fitting a large number of calibration points. Independent calibration of the X-axis and Y-axis directions is taken into account to make the measurement results more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a flowchart of the prior art three-dimensional reconstruction method;

[0059] Figure 2This is a schematic structural diagram of a microscope three-dimensional reconstruction device according to an embodiment of the present application;

[0060] Figure 3 This is a flow chart of a method for reconstructing a three-dimensional microscope sample image according to an embodiment of the present application;

[0061] Figure 4 This is a flow chart of step S320 in a microscope three-dimensional sample image reconstruction method according to an embodiment of the present application;

[0062] Figure 5 is a schematic diagram of measuring position distance in an embodiment of the present application;

[0063] Figure 6 This is a flow chart of step S321 in a microscope three-dimensional sample image reconstruction method according to an embodiment of the present application;

[0064] Figure 7 This is a flow chart of step S330 in a microscope three-dimensional sample image reconstruction method according to an embodiment of the present application.

[0065] Explanation of the accompanying symbols: 1. Sample; 2. Objective lens; 3. Inclination detection module; 31. Dichroic mirror; 32. Detection light source; 33. Detector; 331. Plane reference position mark point; 332. Centroid of reflected light; 4. Microscopic imaging module. DETAILED DESCRIPTION

[0066] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and appended claims of this application, the singular expressions "a," "an," "said," "above," "the," and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to any or all possible combinations comprising one or more of the listed items.

[0067] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0068] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0069] In a first aspect, the present application provides a microscope three-dimensional reconstruction device, such as Figure 2As shown, the device includes a microscopic imaging module 4 and a tilt detection module 3. The microscopic imaging module 4 is used to obtain scanning images of multiple layers of the sample; the tilt detection module 3 is used to measure the sample tilt corresponding to each layer of the image; Figure 3 , the method comprises the following steps:

[0070] S310, control the microscopic imaging module to obtain scanning images of multiple layers of the sample, and obtain a series of two-dimensional images M1 to M n， , n is a positive integer.

[0071] S320, measure the sample inclination corresponding to each layer of image through the inclination detection module, and obtain the x-axis inclination θ of each layer n The inclination angles θ of the x and y axes n y.

[0072] S330, according to the x-axis inclination angle θ n The inclination angles θ of the x and y axes n y, for the image M at the corresponding level n Perform coordinate transformation to obtain the corrected three-dimensional coordinates (x n ',y n ',z n ').

[0073] S340 , based on the corrected three-dimensional coordinates, reconstruct the multi-layer image in three dimensions to obtain a three-dimensional image of the sample.

[0074] In one embodiment, referring to Figure 2 The tilt detection module 3 includes a detection light source 32, a detection plane, a detector 33, and a dichroic mirror 31. The detection light source 32 is used to emit detection light; a reference position mark is provided on the detection plane; the detector 33 is arranged on the detection plane and is used to receive the detection light reflected by the sample 1; and the dichroic mirror 31 is used to reflect the detection light and transmit visible light. The detection light emitted by the detection light source 32 enters the objective lens 2 through the dichroic mirror 31, is reflected by the sample, passes through the objective lens 2 and the dichroic mirror 31, and finally enters the detector 33. Furthermore, the module also includes a processor and a memory. The processor is used to execute the above-mentioned microscope three-dimensional sample image reconstruction method; the memory is used to store program instructions and data.

[0075] Reference Figure 4 In step S320, the sample inclination corresponding to each layer of image is measured by the inclination detection module to obtain the x-axis inclination θ of each layer. n The inclination angles θ of the x and y axes n y, specifically including the following steps:

[0076] S321. Emit detection light to the sample surface.

[0077] S322. Collect the detection light reflected by the sample to obtain a detection light spot.

[0078] S323. Calculate the offset of the center of mass of the detection light spot relative to the reference position mark point to obtain a lateral offset Δx and a longitudinal offset Δy.

[0079] S324, determine the x-axis inclination angle θ of the sample based on the lateral offset Δx and the longitudinal offset Δy n The inclination angles θ of the x and y axes n y.

[0080] like Figure 5 As shown, 33 is a detector, 331 is a plane reference position mark point, the centroid of the reflected light emitted by the detection light source 32 and reflected by the sample 1 is 332, and the lateral position distance between the centroid of the reflected light of the detection light spot and the plane reference position mark point 331 is Δx, and the longitudinal position distance is Δy.

[0081] In one embodiment, the device also includes a sample stage, an adjustable inclination platform and a display unit. The sample stage is used to move in the z-axis direction to obtain sample images at different depth positions; the adjustable inclination platform is used to set multiple calibration inclinations during the calibration process; and the display unit is used to display the reconstructed three-dimensional image.

[0082] Reference Figure 6 In step S321, before emitting the detection light to the sample surface, the method further includes the following steps:

[0083] S610: Set multiple calibration inclination angles θ1x to θ1x on the adjustable inclination platform. M x, M are positive integers.

[0084] S620, measuring the position offset Δx1 to Δx corresponding to the calibration inclination M .

[0085] S630 : Based on the calibrated tilt angle and the position offset, establish an x-axis tilt angle response function θx=f(Δx).

[0086] In the same way, the y-axis tilt angle response function is established through steps S640-S660. It can be understood that there is no fixed order for establishing the y-axis tilt angle response function and the x-axis tilt angle response function.

[0087] S640, setting multiple calibration inclination angles θ1y to θ1y on the adjustable inclination platform M y.

[0088] S650, measuring the position offset Δy1 to Δy corresponding to the calibration inclination M .

[0089] S660 : Based on the calibrated inclination angle and the position offset, establish a y-axis inclination response function θy=f(Δy).

[0090] According to the lateral offset Δx, longitudinal offset Δy, x-axis tilt response function and y-axis tilt response function, the x-axis tilt angle θ of the sample is determined. n The inclination angles θ of the x and y axes n y.

[0091] In one embodiment, the coordinate transformation is expressed in matrix form, where:

[0092] The x-axis rotation matrix is: The y-axis rotation matrix is:

[0093] Reference Figure 7 In step S330, according to the x-axis inclination angle θ n The inclination angles θ of the x and y axes n y, for the image M at the corresponding level n Perform coordinate transformation to obtain the corrected three-dimensional coordinates (x n ',y n ',z n '), specifically comprising the following steps:

[0094] S710, based on the x-axis inclination angle θ n x and x-axis rotation matrix, perform x-axis rotation transformation on the image to obtain the first transformation coordinates.

[0095] In this embodiment, the initial coordinates (x, y, z) are first obtained; then based on the x-axis inclination angle θ n x and x-axis rotation matrices transform the initial coordinates (x, y, z) to obtain the first transformed coordinates (x1, y1, z1), where x1 = x; y1 = ycosθ n x-zsinθ n x; z1 = ysinθ n x+zcosθ n x.

[0096] S720, based on the y-axis inclination angle θ n y and y-axis rotation matrix, perform y-axis rotation transformation on the first transformed coordinate to obtain the corrected three-dimensional coordinate (x n ',y n ',z n ').

[0097] In this embodiment, the first transformation coordinate (x1, y1, z1) is obtained first; then based on the y-axis inclination angle θ n y and y-axis rotation matrix, perform y-axis rotation transformation on the first transformation coordinate (x1, y1, z1) to obtain the corrected three-dimensional coordinate (x n ',y n ',zn '), where x n '=x1cosθ n y+z1sinθ n y;y n '=y1;z n '=-x1sinθ n y+z1cosθ n y.

[0098] In one embodiment, each layer image M n Corresponding to a depth position z n In step S340, based on the corrected three-dimensional coordinates, the multi-layer image is three-dimensionally reconstructed to obtain a three-dimensional image of the sample. The corrected three-dimensional coordinates (x n ',y n ',z n ')According to different z n The values ​​are arranged to obtain a calibrated three-dimensional image.

[0099] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0100] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for reconstructing a three-dimensional sample image from a microscope, characterized in that: Applied to a three-dimensional reconstruction device, the three-dimensional reconstruction device includes a microscopic imaging module and a tilt detection module, and the method includes the following steps: Control the microscopic imaging module to obtain scanning images of multiple layers of the sample to obtain a series of two-dimensional images M1 to M n , n is a positive integer; The sample inclination angle corresponding to each layer of image is measured by the inclination detection module to obtain the x-axis inclination angle θ of each layer. n The inclination angles θ of the x and y axes n y; According to the x-axis inclination angle θ n The inclination angles θ of the x and y axes n y, using matrix transformation method to transform the image M of the corresponding layer n Perform coordinate transformation to obtain the corrected three-dimensional coordinates (x n ',y n ',z n '); Based on the corrected three-dimensional coordinates, the multi-layer image is three-dimensionally reconstructed to obtain a three-dimensional image of the sample; The tilt detection module includes a detection plane, on which a reference position mark point is provided; the tilt detection module measures the sample tilt angle corresponding to each layer of the image to obtain the x-axis tilt angle θ of each layer. n The inclination angles θ of the x and y axes n y, specifically including the following steps: Emitting detection light to the sample surface; Collect the detection light reflected by the sample to obtain a detection light spot; Calculating the offset of the center of mass of the detection light spot relative to the reference position mark point to obtain a lateral offset Δx and a longitudinal offset Δy; According to the lateral offset Δx and the longitudinal offset Δy, the x-axis inclination angle θ of the sample is determined. n The inclination angles θ of the x and y axes n y.

2. The microscope three-dimensional sample image reconstruction method according to claim 1, characterized in that: Before emitting the detection light to the sample surface, the method further includes the following steps: Set multiple calibration inclination angles θ1x to θ on the adjustable inclination platform M x, M are positive integers; Measure the position offset Δx1 to Δx corresponding to the calibration inclination M ; Based on the calibrated inclination angle and the position offset, establishing an x-axis inclination angle response function θx=f(Δx); Set multiple calibration inclination angles θ1y to θ on the adjustable inclination platform M y; Measure the position offset Δy1 to Δy corresponding to the calibration inclination M ; Based on the calibrated inclination angle and the position offset, establishing a y-axis inclination angle response function θy=f(Δy); The x-axis tilt angle θ of the sample is determined according to the lateral offset Δx, the longitudinal offset Δy, the x-axis tilt angle response function and the y-axis tilt angle response function. n The inclination angles θ of the x and y axes n y.

3. The microscope three-dimensional sample image reconstruction method according to claim 1, characterized in that: The coordinate transformation is expressed in matrix form, where The x-axis rotation matrix is: The y-axis rotation matrix is:

4. The microscope three-dimensional sample image reconstruction method according to claim 3, characterized in that: According to the x-axis inclination angle θ n The inclination angles θ of the x and y axes n y, for the image M at the corresponding level n Perform coordinate transformation to obtain the corrected three-dimensional coordinates (x n ',y n ',z n '), specifically comprising the following steps: Based on the x-axis inclination angle θ n x and x-axis rotation matrix, perform x-axis rotation transformation on the image to obtain the first transformation coordinates; Based on the y-axis inclination angle θ n y and y-axis rotation matrix, perform y-axis rotation transformation on the first transformed coordinates to obtain the corrected three-dimensional coordinates (x n ',y n ',z n ').

5. The microscope three-dimensional sample image reconstruction method according to claim 4, characterized in that: Based on the x-axis inclination angle θ n x and x-axis rotation matrix, perform x-axis rotation transformation on the image to obtain the first transformation coordinates, specifically including the following steps: Get the initial coordinates (x, y, z); Based on the x-axis inclination angle θ n x and x-axis rotation matrices transform the initial coordinates (x, y, z) to obtain the first transformed coordinates (x1, y1, z1), where x1 = x; y1 = ycosθ n x-zsinθ n x; z1 = ysinθ n x+zcosθ n x.

6. The microscope three-dimensional sample image reconstruction method according to claim 5, characterized in that: Based on the y-axis inclination angle θ n y and y-axis rotation matrix, perform y-axis rotation transformation on the first transformed coordinates to obtain the corrected three-dimensional coordinates (x n ',y n ',z n '), specifically comprising the following steps: Get the first transformation coordinates (x1, y1, z1); Based on the y-axis inclination angle θ n y and y-axis rotation matrix, perform y-axis rotation transformation on the first transformed coordinate (x1, y1, z1) to obtain the corrected three-dimensional coordinate (x n ',y n ',z n '), where, x n ' = x1 cosθ n y + z1 sinθ n y; y n ' = y1; z n ' = -x1 sinθ n y + z1 cosθ n y.

7. The microscope three-dimensional sample image reconstruction method according to claim 1, characterized in that: Each layer image M n Corresponding to a depth position z n Based on the corrected three-dimensional coordinates, the multi-layer image is three-dimensionally reconstructed to obtain a three-dimensional image of the sample, which specifically includes the following steps: The corrected three-dimensional coordinates (x n ',y n ',z n ')According to different z n The values ​​are arranged to obtain a calibrated three-dimensional image.

8. A microscope three-dimensional reconstruction device, characterized in that: The device comprises: A microscopic imaging module (4) for obtaining scanning images of multiple layers of the sample (1); The tilt detection module (3) comprises: A detection light source (32) for emitting detection light; A detection plane, wherein a reference position marking point is provided on the detection plane; a detector (33), arranged on the detection plane, for receiving detection light reflected by the sample (1); a dichroic mirror (31) for reflecting the probe light and transmitting visible light; A processor, configured to execute the microscope three-dimensional sample image reconstruction method according to any one of claims 1 to 7; Memory, used to store program instructions and data.

9. The microscope 3D reconstruction device according to claim 8, characterized in that: The device further comprises: A sample stage, used for moving in the z-axis direction to acquire sample images at different depths; Adjustable inclination platform for setting multiple calibration inclination angles during the calibration process; The display unit is used to display the reconstructed three-dimensional image.

Images