A quadrature line scan imaging processing method and system

Abstract

The application discloses a kind of orthogonal line scanning imaging processing method and system, belong to image processing field.The method includes: S1, using two focusing surface same orthogonal line light beams to sample irradiation, control orthogonal line light beams and sample relative motion, obtain two strip images I +45° 、 I ‑45° ;S2, pixel in I +45° 、 I ‑45° Translation is homing, obtains two homing images I +45°′ 、 I ‑45°′ ;S3, two homing images I +45°′ 、 I ‑45°′ Fusion, obtains fusion image I recon . Since the focusing surface of two line light beams is same and focusing direction is orthogonal, so only need to carry out strip image I +45° 、 I ‑45° Pixel translation homing, restore to real position, subsequently through algorithm fusion, can overcome the problem of uneven horizontal spatial resolution under single direction line scanning mode, and promote tomography capacity.The whole process does not need to carry out rotation, scaling and other additional image registration operation, and data processing amount is also greatly reduced, can effectively reduce operation time.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of image processing, and more particularly, relates to a kind of orthogonal line scanning imaging processing method and system. BACKGROUND

[0002] In the field of image processing, background interference makes it difficult to obtain clear focal plane images, which greatly affects the imaging quality of the microscopic system. In order to remove background interference, laser scanning confocal microscopy (LSCM) uses pinholes and other hardware to directly prevent background from entering the detector, and compared with traditional wide-field microscopy, the imaging quality has been significantly improved. Traditional LSCM uses point scanning mode to obtain the complete morphology of the sample, but due to the scanning speed of the scanning device, this technology takes too long when imaging large samples of centimeter size and above, and cannot be applied. In this regard, imaging methods based on line scanning mode have been developed. Compared with point scanning mode, line scanning mode can obtain the detection results of a row of detection arrays at a time, greatly shortening the imaging time. However, the imaging system based on line scanning has the problem of uneven distribution of lateral spatial resolution, i.e. the spatial resolution in the unfocused direction is poorer than that in the focused direction. At present, based on the idea of line illumination modulation, various algorithms have been developed for imaging systems based on line scanning mode, which have improved the imaging quality of the system from different aspects, but have not been able to overcome the inherent defect of uneven distribution of lateral spatial resolution.

[0003] To address the problem of uneven distribution of spatial resolution in optical microscopic systems, the following two approaches can be used to improve it: 1) image processing such as deconvolution, such as the classic Wiener filter, Richard-Lucy deconvolution algorithm, etc. The process of optical system imaging is a convolution process of sample light intensity distribution and point spread function of optical system, therefore, by performing inverse operation such as deconvolution on the image, the spatial resolution of the system can be improved; 2) improvement of optical system, such as the classic 4pi microscope. Optical systems are limited by numerical aperture and cannot detect all the light, which is the fundamental reason for the uneven distribution of spatial resolution. By acquiring multiple detection images at different angles and using algorithms to fuse them, the numerical aperture of the optical system can be effectively expanded, and thus its spatial resolution can be improved.

[0004] Both of the two methods can improve the spatial resolution of the optical system, but each has its own shortcomings: 1) the inverse process of deconvolution requires accurate acquisition of the point spread function of the system, but the point spread function cannot remain constant due to the influence of factors such as environment and light source, so it often needs to be limited, resulting in complex image processing algorithm, and the processing result of large sample imaging cannot be maintained superior, and iteration is often introduced in the deconvolution process, resulting in long operation time; 2) obtaining imaging results of the same sample at different angles often requires sequential imaging at different times, resulting in increased imaging time, and later requires image registration, image fusion and other reconstruction algorithm processing, which has large operation amount and long operation time.

[0005] Therefore, it is meaningful to develop a fast imaging method based on line scanning mode and optimize the uneven problem of transverse spatial resolution distribution through simple design. SUMMARY

[0006] In view of the above defects or improvement needs of the prior art, the present application provides a kind of orthogonal line scanning imaging processing method and system, its purpose is to overcome the uneven problem of transverse spatial resolution of line scanning system by simple design and algorithm, and then quickly obtain the high-quality imaging result of large size sample.

[0007] To achieve the above-mentioned purpose, according to one aspect of the present application, an orthogonal line scanning imaging processing method is provided, comprising the following steps:

[0008] S1, two orthogonal line beams with the same focusing surface are used to irradiate the sample, the relative motion of the orthogonal line beam and the sample in the diagonal direction of the orthogonal line beam is controlled, and two strip images of the orthogonal line beam are obtained 、 ;

[0009] S2, the pixels in the two strip images 、 are translated and reset to obtain two reset images 、 ;

[0010] S3, the two reset images 、 are fused to obtain a fused image .

[0011] Through the above technical scheme, since the focusing surfaces of the two orthogonal line beams are the same and the focusing directions are orthogonal, the two orthogonal line beams have nearly consistent distribution in three-dimensional space, so the strip images of the two orthogonal line beams 、 Then, simply shift the pixels back to their true positions, and then combine the two repositioned images. , Fusion can overcome the problem of uneven spatial resolution in the lateral direction of imaging results in unidirectional line scanning mode, and improve tomographic capabilities, resulting in fused images. This refers to the focal plane image of the sample. The entire processing does not require additional image registration operations such as image rotation and scaling, and the amount of data processed is greatly reduced, which can effectively reduce the computation time.

[0012] Preferably, obtaining the strip image in S1 specifically includes:

[0013] The detector scans N frames of images of the sample in a strip scanning manner;

[0014] Then, the center row pixels of each frame are taken as the image result of one frame, or the single frame processing result obtained after processing a single frame / adjacent multiple frames of images with the line illumination modulation algorithm is taken as the image result of one frame. The image results of N frames under two orthogonal line beams are stitched together in chronological order to form a strip image. , .

[0015] Preferably, the velocity of the relative motion between the orthogonal line beam and the sample in S1 ,in, This is the equivalent pixel size during tilted scanning. This is the exposure time for a single frame of the detector.

[0016] Preferably, the pixel repositioning process in S2 is as follows:

[0017] , ;

[0018] , Where i represents the strip image The i-th pixel in the n-th row, where j represents the strip image. The j-th pixel in the n-th row of the image, where m is the total number of pixels in a single row of the image.

[0019] Preferably, the strip images obtained by stitching together in time sequence in S1 , The size is m×N, and the repositioned image in S2 , The size is m×(m+N), where m represents the total number of pixels in a row in a single frame image, and N represents the number of image frames detected by the detector.

[0020] Preferably, the two repositioned images in S3 , Multiplication and fusion are performed using a point spread function.

[0021] Preferably, the point spread function of the unidirectional line beam imaging in the orthogonal line beam is: , ;in Let be the diffusion function of the illumination point. The diffusion function of the probe point;

[0022] ,in Let be the point spread function for linear array illumination. Let be the point spread function of the objective lens;

[0023] ,in Indicates the pixel unit size of the detector;

[0024] The point spread function of the image of the other line beam corresponding to the unidirectional line beam in the orthogonal line beam is: , The superscript T indicates matrix transpose;

[0025] The point spread function after fusion is , For the impulse function under ideal conditions, .

[0026] Preferably, the two repositioned images in S3 , Fusion is performed using an intensity-based minimum projection method: ;

[0027] Alternatively, fusion can be performed using a frequency-domain-based maximum projection method: , and These represent the two-dimensional discrete Fourier transform and the inverse two-dimensional discrete Fourier transform, respectively.

[0028] Alternatively, fusion can be performed using a deconvolution-based method: , ]},in, and These represent the two-dimensional discrete Fourier transform and the inverse two-dimensional discrete Fourier transform, respectively. Represents the image of repositioning The estimated value at each iteration during the deconvolution process. Represents the image of repositioning In the deconvolution process, the estimated value for each iteration, in the first iteration... , In subsequent iterations, , , The point spread function represents the line beam imaging at a -45° direction. The point spread function represents the line beam imaging at +45° direction, and the superscript T indicates matrix transpose. This invention also proposes an orthogonal line scanning system for performing the above-described orthogonal line scanning image processing method, comprising:

[0029] The imaging module illuminates the sample with two orthogonal line beams having the same focal plane, and controls the relative movement of the orthogonal line beams and the sample along the diagonal direction of the orthogonal line beams to acquire two strip images of the orthogonal line beams. , ;

[0030] The repositioning module is used to reposition two strip images. , The pixels in the image are shifted and realigned to their original positions to obtain two realigned images. , ;

[0031] The fusion module is used to merge two repositioned images. , Perform fusion to obtain a fused image. .

[0032] Preferably, the imaging module specifically includes:

[0033] The driving unit is used to drive the orthogonal linear beam and the sample to move relative to each other in the diagonal direction of the orthogonal linear beam, at a speed of ,in, This is the equivalent pixel size during tilted scanning. This refers to the single-frame exposure time of the detector;

[0034] The imaging unit uses a detector to scan N frames of images of the sample in a strip scanning manner;

[0035] The strip image acquisition unit selects the center row pixels of each frame as the image result of one frame, or the single frame processing result obtained after processing a single frame / adjacent multiple frames of images using a line illumination modulation algorithm as the image result of one frame. The image results of N frames under two orthogonal line beams are then stitched together in chronological order to form a strip image. , . Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the orthogonal line beam illuminating the sample according to the present invention;

[0037] Figure 2 This is a schematic diagram illustrating the principle of improving the spatial sampling rate in this invention (with...). (For example)

[0038] Figure 3 It is a strip image A diagram illustrating pixel repositioning;

[0039] Figure 4 This is a schematic diagram of image processing of mouse brain slices according to the present invention;

[0040] Figure 5 This is a comparison chart of simulation results of point confocal system, single-direction line scanning system and orthogonal line scanning system of the present invention in terms of three-dimensional spatial resolution and tomography capability. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0042] like Figure 1 As shown, this invention proposes an orthogonal line scanning imaging processing method, comprising the following steps:

[0043] S1, two orthogonal line beams with the same focal plane are used to illuminate the sample. The relative motion between the orthogonal line beams and the sample is controlled in the diagonal direction of the orthogonal line beams to acquire two strip images of the orthogonal line beams. , ;

[0044] S2, converting the two strip images , The pixels in the image are shifted and realigned to their original positions to obtain two realigned images. , ;

[0045] S3, reposition the two images , Perform fusion to obtain a fused image. .

[0046] Since the two orthogonal linear beams have the same focal plane (the focal plane of the objective lens) and their focusing directions are perpendicular to each other, they have a nearly identical distribution in three-dimensional space. Therefore, obtaining the strip images of the two orthogonal linear beams is straightforward. , Then, simply shift the pixels back to their true positions, and then combine the two repositioned images. , Fusion can overcome the problem of uneven spatial resolution in the lateral direction of imaging results in unidirectional line scanning mode, and improve tomographic capabilities, resulting in fused images. This refers to the focal plane image of the sample. The entire processing does not require additional image registration operations such as image rotation and scaling, and the amount of data processed is greatly reduced, which can effectively reduce the computation time.

[0047] In S1, strip images are acquired through continuous exposure using a detector. The acquired multi-frame images are then extracted and stitched together. Some calculations can be performed before stitching. Specifically, this can be done by: using a detector to scan the sample in a strip-scan manner to obtain N frames of images; then taking the center row pixels of each frame as the image result, or using a single frame or adjacent multi-frame images processed by a line illumination modulation algorithm as the image result. The N frames of images under orthogonal line beams are then stitched together in chronological order to form a strip image. , The line illumination modulation (LIMo) algorithm can be used for processing, employing existing LIMo optical tomography (LiMo) techniques or other LIMo algorithms. LiMo processes multiple adjacent frames to obtain a single-frame result, while other LIMo algorithms process a single frame to obtain a single-frame result. The processing results of all frames can be stitched together chronologically. In other words, this application can directly extract the center row pixels from N frames in S1 and stitch them together to form a strip image. , Alternatively, one can combine the existing line illumination modulation algorithm with the N frames of images in S1 to stitch the processed single frames into a strip image. , While incorporating a line illumination modulation algorithm increases the processing steps, it further enhances the system's imaging performance. More specifically, during scanning, two linear array detectors, each conjugate to one of the two orthogonal line beams, can be used for detection. One detector detects the signal light corresponding to a single-direction line beam within the orthogonal line beam, thus obtaining a strip image. The other probe detects a corresponding beam of signal light, obtaining a strip image. Alternatively, a single area detector paired with a cross-shaped mask can be used for detection, simultaneously acquiring two signal beams corresponding to two orthogonal line beams. The two signal beams are then separated to obtain separate strip images. , .

[0048] like Figure 2 As shown, regardless of whether one or two detectors are used, because the orthogonal line beam and the sample move relative to each other along the diagonal direction of the orthogonal line beam during detection, each line spot is tilted. Therefore, under this tilted scanning detection, the actual spatial sampling rate in the lateral direction is... , The size of the detector on the objective lens focal plane. This refers to the equivalent pixel size during the tilted scanning method described in this application. For example, the pixel size of the detector is... Using a magnification factor of If the objective lens is used for imaging, then the size of the detector on the focal plane of the objective lens, which is also the lateral spatial sampling rate of the system, is... In this tilted scanning mode, the actual lateral spatial sampling rate of the system is... .

[0049] More specifically, the diagonal direction of orthogonal linear beams is Figure 1 In the X-direction, the relative motion between the orthogonal beam and the sample along the X-direction is used for multi-frame strip scanning imaging. This can be achieved by using a moving sample stage to move the sample while keeping the orthogonal beam stationary. The relative velocity between the orthogonal beam and the sample in S1... ,in, This refers to the exposure time per frame for the detector. Alternatively, a moving optical system can be used to move the orthogonal beam while keeping the sample stationary.

[0050] The pixel repositioning process in S2 is as follows: , ; , Where i represents the strip image The i-th pixel in the n-th row, where j represents the strip image. The j-th pixel in the n-th row of the image, where m is the total number of pixels in a single row of the image.

[0051] Figure 3 In the diagram (a), it indicates that the standard image (as a sample) commonly used in the field of image processing is scanned using the orthogonal line beam of this embodiment. The standard image moves along the x-direction, which is the direction in which the sample stage moves the sample. Figure 3 (b) represents the strip image. , Figure 3 (c) indicates that the strip image is... The repositioned image after pixel repositioning .

[0052] S1 strip image , The size of the image is m×N. The size of the image after the pixels are returned to their real positions in S2 is m×(m+N), where m represents the total number of pixels in a row in a single frame image and N represents the number of image frames detected by the detector.

[0053] In some embodiments, the two repositioned images in S3 , Multiplication and fusion are performed using a point spread function.

[0054] Specifically, the point spread function (PSF) of an optical system characterizes its spatial resolution. Single-direction line-beam imaging in orthogonal line beams, consistent with traditional line-scan imaging, has a PSF denoted as... It can be expressed as the illumination point spread function. and probe point spread function The product of, i.e.: ;

[0055] Wherein, the illumination point spread function It is the point spread function of linear array illumination. Point spread function of the objective lens The convolution, i.e.: ;

[0056] Among them, the probe point spread function It is the pixel unit size of the detector. Point spread function of the objective lens The convolution, i.e.: ;

[0057] therefore, ;

[0058] Similarly, the imaging result of the other line beam corresponding to the single-direction line beam in an orthogonal line beam is: , , , The superscript T indicates matrix transpose;

[0059] therefore, ;

[0060] Fusion: Furthermore, under ideal circumstances Since it is an impulse function, after simplification, .

[0061] In other embodiments, the two repositioned images in S3 , Fusion is performed using an intensity-based minimum projection method: This fusion method can effectively avoid background residue in the non-focus direction and combines the advantages of each line confocal imaging result in lateral resolution from both directions.

[0062] In other embodiments, the two repositioned images in S3 , Fusion is performed using a frequency-domain-based maximum projection method: , and These represent the two-dimensional discrete Fourier transform and the two-dimensional inverse discrete Fourier transform, respectively. This fusion method can preserve imaging information from both directions relatively well.

[0063] In other embodiments, the two repositioned images in S3 , Fusion is performed using a deconvolution-based method: , ]},in, and These represent the two-dimensional discrete Fourier transform and the inverse two-dimensional discrete Fourier transform, respectively. Represents the image of repositioning The estimated value at each iteration during the deconvolution process. Represents the image of repositioning In the deconvolution process, the estimated value for each iteration, in the first iteration... , In subsequent iterations, , , This represents the point spread function for imaging a line beam in the -45° direction within an orthogonal line beam. The point spread function represents the image formation of a line beam in the +45° direction within an orthogonal line beam, with the superscript T indicating matrix transpose. This fusion method, based on removing background residue and achieving lateral resolution homogenization, can further improve the resolution of the imaging results.

[0064] In the field of optical imaging, the lateral spatial resolution distribution of a single-direction line scanning system is uneven, meaning the resolution is higher in the lateral focusing direction than in the lateral non-focusing direction. This application uses two line beams with the same focal plane and perpendicular focusing directions for simultaneous imaging. Since the focusing directions of the two beams in orthogonal line beams are orthogonal, the problem of poor resolution in the non-focusing direction of a single-direction line beam can be optimized by utilizing the better resolution of the other beam in that direction. An algorithm fuses the line scanning results from the two directions, ultimately achieving a more uniform lateral resolution compared to the single-direction line beam method. Furthermore, the point confocal system under ideal conditions... It can be observed that the PSF of the line scan system in this embodiment is consistent with that of the point confocal system in the prior art, which further illustrates that the imaging method of this embodiment can achieve improved lateral resolution, homogenization and tomography capabilities, and achieve the same imaging quality as the point confocal system. Moreover, the imaging speed of the line scan system is faster than that of the point confocal system and it is suitable for large-size samples. Therefore, this embodiment can achieve fast and high-quality imaging of large-size samples.

[0065] During the aforementioned fusion process, it is possible to... It was found that the dynamic range of the reconstructed images was improved, for example, in strip images. and For a 16-bit image, after the fusion algorithm... Convert to a 32-bit image. A 32-bit image has a higher dynamic range, which is beneficial for displaying samples with rich details and complex structures. Of course, a 32-bit image also increases the amount of data. Therefore, this 32-bit image can be further processed into a 16-bit image using methods such as square root operations, which can avoid increasing the amount of data.

[0066] like Figure 4 As shown, the orthogonal line beams of this embodiment were used to visualize the brain of Thy1-eGFP-labeled mice. Imaging is performed on thick slices. Figure 4 The image shows the dendrites of neurons in a brain slice, where multiple dendritic spines are present. Figure 4 (a) shows the strip image acquired by the detector under illumination by the -45° line beam in the orthogonal line beams. , Figure 4 Image (b) shows the strip image acquired by the detector under illumination from the +45° line beam in the orthogonal line beams. , Figure 4 (c) represents the strip image. The repositioned image after pixel repositioning. Figure 4 (d) represents the strip image. The repositioned image after pixel repositioning. Figure 4 In the image (e), the image obtained by fusing the two repositioned images (c) and (d) is shown. This refers to the focal plane image of the sample. It can be observed that the dendritic spines on neurons in the mouse brain are on the order of micrometers. During imaging in an online scanning system, due to the non-uniform lateral resolution, their shape is easily altered, which is reflected in… Figure 1 In (c) and (d), the dendritic spines exhibit tailing in the non-focusing direction. However, this error can be avoided by reconstructing the image using an orthogonal line illumination imaging system. Furthermore, compared to the single-direction line scan imaging results (c) and (d), the final imaging result (e) with orthogonal line illumination also shows an improved signal-to-background ratio, indicating that this embodiment can enhance the tomographic capability of the line scan system.

[0067] The present invention also proposes an orthogonal line scanning imaging system, comprising:

[0068] The imaging module illuminates the sample with two orthogonal line beams having the same focal plane, and controls the relative movement of the orthogonal line beams and the sample along the diagonal direction of the orthogonal line beams to acquire two strip images of the orthogonal line beams. , ;

[0069] The repositioning module is used to reposition two strip images. , The pixels in the image are shifted and realigned to their original positions to obtain two realigned images. , ;

[0070] The fusion module is used to merge two repositioned images. , Perform fusion to obtain a fused image. .

[0071] The imaging module specifically includes:

[0072] The driving unit is used to drive the orthogonal linear beam and the sample to move relative to each other in the diagonal direction of the orthogonal linear beam, at a speed of ,in, This refers to the equivalent pixel size in the tilt scanning mode of this application. This refers to the single-frame exposure time of the detector;

[0073] The imaging unit uses a detector to scan N frames of images of the sample in a strip scanning manner;

[0074] The strip image acquisition unit selects the center row pixels of each frame as the image result of one frame, or the single frame processing result obtained after processing a single frame / adjacent multiple frames of images using a line illumination modulation algorithm as the image result of one frame. The image results of N frames under two orthogonal line beams are then stitched together in chronological order to form a strip image. , .

[0075] The drive unit may also include along Figure 5 The sample stage, which moves in the X direction, is used to carry the sample and move it at a speed. move.

[0076] The specific process of pixel translation and repositioning in the repositioning module is as follows: , Where i represents the strip image. The i-th pixel in the n-th row, where j represents the strip image. The j-th pixel in the n-th row of the image, where m is the total number of pixels in a single row of the image.

[0077] In some embodiments, the fusion module merges two repositioned images. , Multiplication and fusion are performed using a point spread function. That is:

[0078] Ideally Since it is an impulse function, after simplification, .

[0079] In other embodiments, the fusion module contains two repositioned images. , Fusion is performed using an intensity-based minimum projection method: This fusion method can effectively avoid background residue in the non-focus direction and combines the advantages of each line confocal imaging result in lateral resolution from the two directions.

[0080] In other embodiments, the fusion module contains two repositioned images. , Fusion is performed using a frequency-domain-based maximum projection method: , and These represent the two-dimensional discrete Fourier transform and the two-dimensional inverse discrete Fourier transform, respectively. This fusion method can preserve imaging information from both directions relatively well.

[0081] In other embodiments, the fusion module contains two repositioned images. , Fusion is performed using a deconvolution-based method: = , ]},in, and These represent the two-dimensional discrete Fourier transform and the inverse two-dimensional discrete Fourier transform, respectively. Represents the image of repositioning The estimated value at each iteration during the deconvolution process. Represents the image of repositioning In the deconvolution process, the estimated value for each iteration, in the first iteration... , In subsequent iterations, , , This represents the point spread function for imaging a line beam in the -45° direction within an orthogonal line beam. The point spread function represents the image formation of a line beam in the +45° direction within an orthogonal line beam, with the superscript T indicating matrix transpose. This fusion method, based on removing background residue and achieving lateral resolution homogenization, can further improve the resolution of the imaging results.

[0082] The effectiveness of the orthogonal line scanning system in this invention will be verified below.

[0083] Figure 5 This is a comparison chart of simulation results for the point confocal system, the unidirectional line scanning system, and the orthogonal line confocal system of this embodiment in terms of three-dimensional spatial resolution and tomography capability. Figure 5In the diagram, (a), (b), and (c) represent the point spread functions of the point confocal system, the unidirectional line scanning system, and the orthogonal line scanning system on the objective lens focal plane, respectively. It can be observed that the unidirectional line scanning system suffers from uneven resolution distribution in both the focusing and non-focusing directions, while the orthogonal line scanning system effectively solves this problem.

[0084] Figure 5 In (d), (e), (f), and (g), ○ represents the simulation results of the point confocal system (LSCM), △ represents the simulation results of the single-direction line scanning system (LS), and the solid line — represents the simulation results of the orthogonal line scanning system (cLS). The simulation conditions are: the numerical aperture of the objective lens is [value missing]. Magnification is The excitation wavelength is The wavelength of the signal light emission is The detector pixel size is .

[0085] Figure 5 In Figures (d), (e), and (f), the point spread function curves of the point confocal system, the unidirectional line scanning system, and the orthogonal line scanning system in the X, Y, and Z directions of three-dimensional space are shown, respectively. The full width at half maximum (FWHM) of the curve represents the spatial resolution in that dimension; the narrower the FWHM, the higher the spatial resolution in that dimension. It can be observed that... Figure 5 In (d), the three imaging systems have almost identical spatial resolution in the X direction; Figure 5 In (e), the single-direction line scan imaging system has poor spatial resolution in the Y direction, while the orthogonal line scan imaging system can achieve the same level of spatial resolution in the Y direction as the point confocal system; and Figure 5 In Figure (f), the orthogonal line scanning system has better spatial resolution in the Z direction than the single-direction line scanning system, and the single-direction line scanning system has better spatial resolution than the point confocal system. Comparing Figures (d) and (e), it can be found that LS-x = 0.32 μm and LS-y = 0.42 μm in the single-direction line scanning system, which are not equal, while LSCM-x = 0.32 μm and LSCM-y = 0.32 μm in the point scanning system, which are equal. That is, the point confocal system has a uniform resolution distribution in the lateral direction, while the single-direction line scanning system has an uneven resolution distribution in the lateral direction. In the orthogonal line scanning system of this application, cLS-x = cLS-y = 0.32 μm, indicating that it can overcome the problem of uneven resolution in the lateral direction of the line scanning system. At the same time, regarding the axial resolution capability, referring to Figure (f), the cross-z, representing the axial resolution capability of the orthogonal line scanning system, is 0.54 μm, which has the narrowest full width at half maximum (FWHM) and the best axial resolution capability.

[0086] ​Figure (g) compares the tomographic capabilities of the point confocal system, the unidirectional line scanning system, and the orthogonal line scanning system. The narrower the full width at half maximum (FWHM) of the curve, the better its tomographic capability. It can be observed that the tomographic capability of the orthogonal line scanning system is comparable to or even better than that of the point confocal system.

[0087] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for orthogonal line scanning imaging processing, characterized in that, Includes the following steps: S1, two orthogonal linear beams with the same focal plane are used to illuminate the sample. The relative motion between the orthogonal linear beams and the sample is controlled in the diagonal direction of the orthogonal linear beams. At the same time, two signal beams corresponding to the two orthogonal linear beams are acquired to obtain two strip images of the orthogonal linear beams. , ; S2, converting the two strip images , The pixels in the image are shifted and realigned to their original positions to obtain two realigned images. , ; S3, reposition the two images , Perform fusion to obtain a fused image. .

2. The orthogonal line scanning imaging processing method according to claim 1, characterized in that, The specific steps for obtaining the strip image in S1 include: The detector scans N frames of images of the sample in a strip scanning manner; Then, the center row pixels of each frame are taken as the image result of one frame, or the single frame processing result obtained after processing a single frame / adjacent multiple frames of images with the line illumination modulation algorithm is taken as the image result of one frame. The image results of N frames under two orthogonal line beams are stitched together in chronological order to form a strip image. , .

3. The orthogonal line scanning imaging processing method according to claim 2, characterized in that, The velocity of the relative motion between the orthogonal linear beam and the sample in S1 ,in, This is the equivalent pixel size during tilted scanning. This is the exposure time for a single frame of the detector.

4. The orthogonal line scanning imaging processing method according to claim 2, characterized in that, The pixel repositioning process in S2 is as follows: ， ； , Where i represents the strip image The i-th pixel in the n-th row, where j represents the strip image. The j-th pixel in the n-th row of the image, where m is the total number of pixels in a single row of the image.

5. The orthogonal line scanning imaging processing method according to claim 4, characterized in that, The strip image obtained by splicing the images in time sequence in S1 , The size is m×N, and the repositioned image in S2 , The size is m×(m+N), where m represents the total number of pixels in a row in a single frame image, and N represents the number of image frames detected by the detector.

6. The orthogonal line scanning imaging processing method according to claim 1, characterized in that, Two repositioning images in S3 , Multiplication and fusion are performed using a point spread function.

7. The orthogonal line scanning imaging processing method according to claim 6, characterized in that, The point spread function of the unidirectional line beam imaging in the orthogonal line beam is: , ;in Let be the diffusion function of the illumination point. The diffusion function of the probe point; ,in Let be the point spread function for linear array illumination. Let be the point spread function of the objective lens; ,in Indicates the pixel unit size of the detector; The point spread function of the image of the other line beam corresponding to the unidirectional line beam in the orthogonal line beam is: , The superscript T indicates matrix transpose; The point spread function after fusion is After simplification, .

8. The orthogonal line scanning imaging processing method according to claim 1, characterized in that, Two repositioning images in S3 , Fusion is performed using an intensity-based minimum projection method: ; Alternatively, fusion can be performed using a frequency-domain-based maximum projection method: , and These represent the two-dimensional discrete Fourier transform and the inverse two-dimensional discrete Fourier transform, respectively. Alternatively, fusion can be performed using a deconvolution-based method: , ]},in, and These represent the two-dimensional discrete Fourier transform and the inverse two-dimensional discrete Fourier transform, respectively. Represents the image of repositioning The estimated value at each iteration during the deconvolution process. Represents the image of repositioning In the deconvolution process, the estimated value for each iteration, in the first iteration... , In subsequent iterations, , , This represents the point spread function for imaging a line beam in the -45° direction within an orthogonal line beam. This represents the point spread function for imaging a line beam in the +45° direction within an orthogonal line beam, with the superscript T indicating matrix transpose.

9. An orthogonal line scanning imaging system, characterized in that, include: The imaging module illuminates the sample with two orthogonal linear beams that have the same focal plane, controls the relative motion between the orthogonal linear beams and the sample along the diagonal direction of the orthogonal linear beams, and simultaneously acquires two signal beams corresponding to the two orthogonal linear beams to obtain two strip images of the orthogonal linear beams. , ; The repositioning module is used to reposition two strip images. , The pixels in the image are shifted and realigned to their original positions to obtain two realigned images. , ; The fusion module is used to merge two repositioned images. , Perform fusion to obtain a fused image. .

10. The orthogonal line scanning imaging system according to claim 9, characterized in that, The imaging module specifically includes: The driving unit is used to drive the orthogonal linear beam and the sample to move relative to each other in the diagonal direction of the orthogonal linear beam, at a speed of ,in, This is the equivalent pixel size during tilted scanning. This refers to the single-frame exposure time of the detector; The imaging unit uses a detector to scan N frames of images of the sample in a strip scanning manner; The strip image acquisition unit selects the center row pixels of each frame as the image result of one frame, or the single frame processing result obtained after processing a single frame / adjacent multiple frames of images using a line illumination modulation algorithm as the image result of one frame. The image results of N frames under two orthogonal line beams are then stitched together in chronological order to form a strip image. , .

Images