Cell image splicing method and device, computer equipment and storage medium

Abstract

The invention provides a cell image splicing method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a first cell image and a second cell image, collected by a microscope, of a cell culture container in different visual fields, and according to the pixel size and magnification of the microscope and the scanning step length of the cell culture container, obtaining a first cell image and a second cell image of the cell culture container; and obtaining a first offset of the second cell image relative to the first cell image, and performing image splicing on the first cell image and the second cell image according to the first offset. Therefore, a splicing strategy based on equipment parameters is adopted, and the problem that image splicing cannot be carried out due to feature point deficiency is solved.

Description

Technical Field

[0001] This application relates to the field of biotechnology, and more specifically, to a method, apparatus, computer device, and storage medium for stitching cell images. Background Technology

[0002] In the fields of biopharmaceuticals, cell engineering, and medical microscopy, a small field-of-view acquisition and stitching strategy is often used to achieve large-field panoramic image reconstruction.

[0003] In related technologies, image stitching methods mainly rely on feature point matching strategies (such as SIFT, SURF, etc.) to perform registration by extracting salient features (such as cell contours, brightness changes, etc.) in the overlapping areas of the images, thereby completing geometric alignment and fusion processing.

[0004] However, in low-concentration cell culture scenarios, the sparse distribution of cells results in no obvious features between images in different fields of view, making it impossible to achieve feature point matching and subsequent stitching processing. Summary of the Invention

[0005] In view of this, embodiments of this application provide a cell image stitching method, apparatus, computer device, and storage medium to solve the problem that in low-concentration cell culture scenarios, the sparse distribution of cells results in no obvious features between small field-of-view images of adjacent fields of view, making feature point matching and subsequent stitching processing impossible.

[0006] In a first aspect, embodiments of this application provide a cell image stitching method, including: Acquire first and second cell images of cell culture vessels from different fields of view using a microscope; Based on the pixel size and magnification of the microscope and the scanning step size of the cell culture container, a first offset of the second cell image relative to the first cell image is obtained; Based on the first offset, the first cell image and the second cell image are stitched together.

[0007] In an optional implementation, the step of stitching the first cell image and the second cell image according to the first offset includes: Based on the first offset, the theoretical overlap region between the first cell image and the second cell image is determined; The theoretically overlapping regions are binarized to obtain binarized segmentation results; Based on the binarized segmentation results, the number of cells in the theoretically overlapping region is obtained; If the number of cells in the theoretically overlapping region is less than or equal to a preset number, then the first cell image and the second cell image are stitched together according to the theoretically overlapping region.

[0008] In an optional implementation, the step of stitching the first cell image and the second cell image according to the first offset further includes: If the number of cells in the theoretically overlapping region is greater than the preset number, then based on the first cell mask in the theoretically overlapping region of the first cell image and the second cell mask in the theoretically overlapping region of the second cell image, image registration is performed on the first cell image and the second cell image to obtain the second offset of the second cell image relative to the first cell image. Based on the second offset, the first cell image and the second cell image are stitched together.

[0009] In an optional implementation, the step of stitching the first cell image and the second cell image according to the second offset includes: Based on the second offset, the initial overlapping region of the first cell image and the second cell image is determined; A first image patch is determined from the initial overlapping region in the second cell image; Based on the first image patch, a matching image patch is obtained by matching the initial overlapping region in the first cell image; Based on the first image block and the matching image block, obtain the third offset of the second cell image relative to the first cell image; Based on the third offset, the first cell image and the second cell image are stitched together.

[0010] In an optional implementation, the step of matching the initial overlapping region in the first cell image based on the first image patch to obtain a matching image patch includes: From the initial overlapping region in the first cell image, determine an image block that meets a preset cell density as a second image block; Based on the first image block, a matching image block is obtained by matching the outer region of the second image block.

[0011] In an optional implementation, the step of matching the initial overlapping region in the first cell image based on the first image patch to obtain a matching image patch includes: Based on the first image patch, the region with the highest correlation coefficient with the first image patch is determined by traversing the initial overlapping region in the first cell image as the matching image patch.

[0012] In an optional implementation, the method further includes: If the correlation coefficient between the first image block and the matching image block is greater than a preset coefficient threshold, then the image stitching is determined to be normal. If the correlation coefficient between the first image block and the matching image block is less than or equal to the preset coefficient threshold, an abnormal prompt message will be pushed to indicate that the image stitching is abnormal.

[0013] Secondly, embodiments of this application also provide a cell image stitching device, comprising: The acquisition module is used to acquire first and second cell images of cell culture containers taken under a microscope from different fields of view. The acquisition module is further configured to acquire a first offset of the second cell image relative to the first cell image based on the pixel size, magnification of the microscope and the scanning step size of the cell culture container; The stitching module is used to stitch the first cell image and the second cell image together according to the first offset.

[0014] Thirdly, embodiments of this application also provide a computer device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the method described in any of the first aspects.

[0015] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the method described in any of the first aspects.

[0016] This application provides a cell image stitching method, apparatus, computer device, and storage medium. The method includes: acquiring first and second cell images of a cell culture container from different fields of view using a microscope; obtaining a first offset of the second cell image relative to the first cell image based on the microscope's pixel size, magnification, and the scanning step size of the cell culture container; and stitching the first and second cell images together based on the first offset. This employs a stitching strategy based on device parameters, solving the problem of image stitching being impossible due to missing feature points. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 Flowchart of the cell image stitching method provided in the embodiments of this application Figure 1 ; Figure 2 A schematic diagram of cell image offset provided in the embodiments of this application. Figure 1 ; Figure 3 A schematic diagram of the first cell image provided for an embodiment of this application. Figure 1 ; Figure 4 A schematic diagram of the second cell image provided in the embodiments of this application. Figure 1 ; Figure 5 A schematic diagram of a stitched panoramic image provided in an embodiment of this application; Figure 6 Flowchart of the cell image stitching method provided in the embodiments of this application Figure 2 ; Figure 7 Flowchart of the cell image stitching method provided in the embodiments of this application Figure 3 ; Figure 8 A schematic diagram of cell image offset provided in the embodiments of this application. Figure 2 ; Figure 9 A schematic diagram of the second cell image provided in the embodiments of this application. Figure 2 ; Figure 10 A schematic diagram of the first cell image provided for an embodiment of this application. Figure 2 ; Figure 11 A schematic diagram of cell image offset provided in the embodiments of this application. Figure 3 ; Figure 12 Flowchart of the cell image stitching method provided in the embodiments of this application Figure 4 ; Figure 13 This is a schematic diagram of the structure of the cell image stitching device provided in the embodiments of this application; Figure 14 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] To address the problem that sparse cell distribution in low-concentration cell culture scenarios leads to a lack of distinct features between images in different fields of view, making feature point matching and subsequent stitching impossible, this application provides a cell image stitching method. This method solves the problem of missing feature points preventing image stitching, overcomes the bottleneck of stitching offset calculation failure in low-concentration cell blank scenes, and enables accurate counting and distribution analysis of cells throughout the low-concentration cell culture area. This provides reliable data support for single-cell clone identification, cell activity monitoring, and cell imaging analysis.

[0021] Figure 1 Flowchart of the cell image stitching method provided in the embodiments of this application Figure 1 In this embodiment, the executing entity can be a computer device.

[0022] like Figure 1 As shown, the method may include: S101. Obtain first and second cell images of the cell culture vessel from different fields of view using a microscope.

[0023] In this system, a cell culture container contains a low concentration of cells, and the container is placed on a stage. In one application scenario, a motor controls the movement of the stage, and during the movement of the stage, the microscope acquires cell images of the cell culture container from different fields of view.

[0024] In another application scenario, a motor controls the movement of a microscope, during which the microscope acquires images of cells in the cell culture container from different fields of view.

[0025] The first cell image is a cell image of the cell culture container under the first field of view acquired by the microscope, and the second cell image is a cell image of the cell culture container under the second field of view acquired by the microscope. The first field of view and the second field of view are different small fields of view, used to indicate different spatial locations of the cell culture container, such as the central region, left side region, and right side region of the cell culture container.

[0026] S102. Based on the pixel size of the microscope, the magnification, and the scanning step size of the cell culture container, obtain the first offset of the second cell image relative to the first cell image.

[0027] The scanning step length of the cell culture container, also known as the sampling step length of the motor, refers to the distance the motor moves from acquiring the first cell image to acquiring the second cell image. In other words, the second cell image is the cell image acquired when the motor has moved the sampling step length with the first cell image as a reference.

[0028] The pixel size of the microscope is denoted as k (m×n, in μm / pixel), the magnification of the microscope is denoted as M, and the sampling step size of the motor is denoted as s (including the sampling step size in the x-direction and the sampling step size in the y-direction, in μm).

[0029] The first offset refers to the theoretical offset of the second cell image relative to the first cell image, determined based on the motor motion, with the first cell image as a reference. The first offset includes the first offset in the x direction and the first offset in the y direction. For example, the offset in the x direction and the offset in the y direction of the origin of the second pixel coordinate system (the upper left vertex of the second cell image) of the second cell image relative to the origin of the first pixel coordinate system (the upper left vertex of the first cell image) of the first cell image.

[0030] The first offset is expressed as: p = (s × M) / k, in pixels. Substitute the sampling step size in the x-direction into this formula to calculate the first offset in the x-direction. Substitute the sampling step size in the y-direction into this formula to calculate the first offset in the y-direction.

[0031] Figure 2 A schematic diagram of cell image offset provided in the embodiments of this application. Figure 1 ,like Figure 2 As shown, in the pixel coordinate system, the x-direction corresponds to the image height, the y-direction corresponds to the image width, the green box represents the first cell image, and the red direction represents the second cell image. The first cell image is the cell image acquired at time t0, and the second cell image is the cell image acquired at time t1. With time t0 as a reference, the arrows indicate the movement of the second cell image relative to the first cell image.

[0032] The width of the first cell image and the second cell image is W in pixels, and the height of the image is H in pixels.

[0033] For the x-direction, the sampling step size of the motor is dx, and the first offset in the x-direction is calculated as px1. The value is positive when the motor moves down and negative when the motor moves up.

[0034] For the y-direction, the sampling step size of the motor is dy, and the first offset in the y-direction is calculated as py1. The rightward movement of the motor is a positive value, and the leftward movement is a negative value.

[0035] It should be noted that if the motor moves smoothly and without error, then dy=0 when the motor moves along the x-direction, and similarly, dx=0 when the motor moves along the y-direction.

[0036] S103. Based on the first offset, stitch the first cell image and the second cell image together.

[0037] The theoretical overlap region between the first cell image and the second cell image is determined based on the first offset. See [link / reference] Figure 2 The width of the theoretically overlapping region is the image width W minus the first offset in the y direction, i.e., W-py1, and the height of the theoretically overlapping region is the image height H minus the first offset in the x direction, i.e., H-px1.

[0038] The theoretical overlapping region is the common area contained in the first cell image and the second cell image, which is determined based on the device parameters (sampling step size of the motor, pixel size of the microscope, and magnification). According to the theoretical overlapping region, the first cell image and the second cell image are registered and aligned, and the registered and aligned overlapping region is image fused. Then, the boundary region is filled with pixels to obtain the stitched panoramic image.

[0039] Figure 3 A schematic diagram of the first cell image provided for an embodiment of this application. Figure 1 , Figure 4 A schematic diagram of the second cell image provided in the embodiments of this application. Figure 1 , Figure 5 This is a schematic diagram of a stitched panoramic image provided in an embodiment of this application.

[0040] like Figures 3 to 5 As shown, the purple solid-line box represents the theoretical overlapping area of ​​the first cell image and the second cell image. After registering and aligning the first cell image and the second cell image, the overlapping area after registration and alignment is image fused, and then the boundary area is filled with black to obtain the stitched panoramic image.

[0041] Figure 6 Flowchart of the cell image stitching method provided in the embodiments of this application Figure 2 ,like Figure 6 As shown, in an optional embodiment, step S103 above, which involves stitching the first cell image and the second cell image together according to the first offset, may include: S201. Determine the theoretical overlap region of the first cell image and the second cell image based on the first offset.

[0042] S202. Binarize the theoretically overlapping regions to obtain the binarized segmentation results.

[0043] Binarization is performed on the theoretically overlapping region in either the first cell image or the second cell image to divide the pixels in the theoretically overlapping region into two categories: cells (e.g., pixel value 1) and non-cell background (e.g., pixel value 0), thus obtaining the binarized segmentation result.

[0044] S203. Based on the binarized segmentation results, obtain the number of cells in the theoretically overlapping region.

[0045] Based on the pixel values ​​in the binarized segmentation results, the number of cell connected regions in the binarized segmentation results is counted, which is taken as the number of cells in the theoretical overlapping region.

[0046] S204. If the number of cells in the theoretically overlapping region is less than or equal to the preset number, then the first cell image and the second cell image are stitched together according to the theoretically overlapping region.

[0047] If the number of cells in the theoretically overlapping area is less than or equal to the preset number, it indicates that the theoretically overlapping area may be a blank area. Then, based on the theoretically overlapping area, the first cell image and the second cell image are registered and aligned, and the registered and aligned overlapping area is image fused. Finally, the boundary area is filled with pixels to obtain a stitched panoramic image.

[0048] It should be noted that the preset quantity is denoted as n_th, and n_th is generally set to 0. If the precision requirement is low, n_th can be increased. The specific selection can be made according to the actual situation, and this embodiment does not make any special limitation on this.

[0049] In an optional implementation, step S103, which involves stitching the first cell image and the second cell image together based on the first offset, may further include: S205. If the number of cells in the theoretically overlapping region is greater than the preset number, then the first cell image and the second cell image are image registered according to the first cell mask in the theoretically overlapping region of the first cell image and the second cell mask in the theoretically overlapping region of the second cell image to obtain the second offset of the second cell image relative to the first cell image.

[0050] If the number of cells in the theoretically overlapping region is greater than a preset number, it indicates that there are sufficiently abundant cells in the theoretically overlapping region. Then, image registration is further performed based on the cell mask. The cell mask is used as a feature, and the edge feature matching registration method is used for image registration. The feature similarity between the first cell mask (mask) in the theoretically overlapping region of the first cell image and the second cell mask (mask) in the theoretically overlapping region of the second cell image is calculated. If the feature similarity exceeds a preset similarity threshold, it indicates that the first cell mask and the second cell mask may be masks of the same cell. Then, based on the position of the first cell mask in the first pixel coordinate system of the first cell image and the position of the second cell mask in the second pixel coordinate system of the second cell image, the relative offset between the second cell mask and the first cell mask is calculated. This relative offset is the position difference between the position of the second cell mask in the second pixel coordinate system and the position of the first cell mask in the first pixel coordinate system.

[0051] It should be noted that the first pixel coordinate system is a pixel coordinate system with the top left vertex of the first cell image as the origin, and the second pixel coordinate system is a pixel coordinate system with the top left vertex of the second cell image as the origin.

[0052] The relative offset between the second cell mask and the first cell mask is referred to as the second offset. The second offset refers to the amount by which the second cell image is offset relative to the first cell image based on the cell mask features, with the first cell image as a reference. It includes the second offset in the x-direction and the second offset in the y-direction.

[0053] S206. Based on the second offset, stitch the first cell image and the second cell image together.

[0054] Since the second offset is determined based on the cell mask features, the target overlapping area between the first cell region and the second cell image is determined based on the second offset for the high-density cell region. Based on the target overlapping area, the first cell image and the second cell image are registered and aligned, and the overlapping area after registration and alignment is image fused. Then, the boundary area is filled with pixels to obtain the stitched panoramic image.

[0055] In this embodiment, the theoretically overlapping region is segmented, and scene judgment is completed based on the number of cells. In the processing of regions that do not contain cells, a first offset based on device parameters is used for image stitching. In the processing of regions that contain cells, the theoretically overlapping region is used as a feature region, and a second offset based on cell mask features is used for image stitching. Since the second offset is based on cell mask features, it can effectively compensate for geometric deviations caused by factors such as motor movement, lens distortion, or stage drift, thus significantly improving the accuracy of image stitching. Therefore, the stitching strategy based on device parameters and dynamic scene adaptation can handle the stitching of sparse foregrounds, solve the problem that the stitching offset cannot be calculated due to missing feature points, and reduce the amount of computation and improve the calculation speed by only using the theoretically overlapping region for image stitching. Figure 7 Flowchart of the cell image stitching method provided in the embodiments of this application Figure 3 ,like Figure 7 As shown, in an optional embodiment, step S204 above, which involves stitching the first cell image and the second cell image together according to the second offset, may include: S301. Determine the initial overlapping area of ​​the first cell image and the second cell image based on the second offset.

[0056] Figure 8 A schematic diagram of cell image offset provided in the embodiments of this application. Figure 2 ,like Figure 8 As shown, the width of the initial overlapping region is the image width W minus the second offset in the y direction, that is, W-py2, and the height of the initial overlapping region is the image height H minus the second offset in the x direction, that is, H-px2.

[0057] S302. Determine the first image block from the initial overlapping region in the second cell image.

[0058] From the initial overlapping region in the second cell image, determine the image block that meets the preset cell density as the first image block.

[0059] Among them, the image block that meets the preset cell density can be the local region with the highest cell density in the initial overlapping region of the second cell image, that is, the first image block is the image region with the highest cell density in the initial overlapping region of the second cell image.

[0060] Figure 9 A schematic diagram of the second cell image provided in the embodiments of this application. Figure 2 ,like Figure 9 As shown, the solid purple box represents the initial overlapping area in the second cell image, and the dashed red box represents the first image patch with the highest cell density, i.e., the ROIA region.

[0061] S303. Based on the first image block, match the initial overlapping region in the first cell image to obtain a matching image block.

[0062] Matching is performed within the initial overlapping region of the first cell image, using a sliding search within this region to find the image block most relevant to the first image block. The matching image block and the first image block are of the same size.

[0063] In some embodiments, template matching strategies, phase correlation methods, or optical flow methods can also be used to match the initial overlapping regions in the first cell image to obtain a matching image block.

[0064] In an optional implementation, step S303 above, which involves matching the initial overlapping region in the first cell image based on the first image patch to obtain a matching image patch, may include: Based on the first image patch, the region with the highest correlation coefficient with the first image patch is determined by traversing the initial overlapping region in the first cell image.

[0065] Using the first image patch as a template, a template matching strategy is employed to traverse the initial overlapping region in the first cell image and calculate the correlation coefficient between the traversed image patch and the first image patch. The region with the highest correlation coefficient is then selected as the matching image patch.

[0066] The larger the correlation coefficient, the stronger the correlation; the smaller the correlation coefficient, the weaker the correlation. The correlation coefficient can be, for example, the normalized cross-correlation (NCC).

[0067] In an optional implementation, step S303 above, which involves matching the initial overlapping region in the first cell image based on the first image patch to obtain a matched image patch, includes: From the initial overlapping region in the first cell image, determine an image block that meets the preset cell density as the second image block; Based on the first image block, a matching image block is obtained by matching the outer region of the second image block.

[0068] Among them, the image block that meets the preset cell density can be the local region with the highest cell density in the initial overlapping region of the first cell image, that is, the second image block is the image region with the highest cell density in the initial overlapping region of the first cell image.

[0069] The outer region of the second image block can be a region formed by extending a preset number of pixels outward from the center of the second image block. For example, the preset number of pixels can be 5 pixels. This embodiment does not make any special limitation on this.

[0070] Based on the first image block, a matching is performed in the outer region of the second image block to determine the matching image block from the outer region.

[0071] Figure 10 A schematic diagram of the first cell image provided for an embodiment of this application. Figure 2 ,like Figure 10 As shown, the solid purple box represents the initial overlapping area in the first cell image, the dashed red box represents the second image patch with the highest cell density, and the dashed blue box represents the outward expansion area, i.e., the ROIB region.

[0072] It should be noted that, based on the first image block, the region with the highest correlation coefficient with the first image block can also be determined by traversing the outer region of the first cell image as the matching image block.

[0073] In this embodiment, template matching is performed within the initial overlapping area of ​​the first cell image based on the first image block to determine the matching image block, or template matching is performed within the outer expansion area of ​​the first cell image based on the first image block to determine the matching image block. Since the outer expansion area is within the initial overlapping area, the search range is small and the computational load is small.

[0074] S304. Based on the first image block and the matching image block, obtain the third offset of the second cell image relative to the first cell image.

[0075] Obtain the position of the first image block in the second pixel coordinate system where the second cell image is located, and the position of the matching image block in the first pixel coordinate system where the first cell image is located. Use the position difference between the position of the first image block in the second pixel coordinate system where the second cell image is located and the position of the matching image block in the first pixel coordinate system where the first cell image is located as the third offset.

[0076] The third offset is the optimal matching offset obtained through correlation matching, including the third offset in the x-direction and the third offset in the y-direction.

[0077] S305. Based on the third offset, stitch the first cell image and the second cell image together.

[0078] Based on the third offset, the actual overlapping area between the first cell image and the second cell image is determined. Then, the first cell image and the second cell image are registered and aligned according to the actual overlapping area. The registered and aligned overlapping area is then image-fused. Finally, the boundary area is filled with pixels to obtain the stitched panoramic image.

[0079] Figure 11 A schematic diagram of cell image offset provided in the embodiments of this application. Figure 3 ,like Figure 11 As shown, the width of the actual overlapping region is the image width W minus the third offset in the y direction, that is, W-py3, and the height of the actual overlapping region is the image height H minus the third offset in the x direction, that is, H-px3.

[0080] In this embodiment, in the processing of the cell region, the second offset is first determined based on the cell mask features, then the initial overlapping region corresponding to the second offset is determined, and then the initial overlapping region in the first cell image is used as the matching region for fine registration to obtain the third offset. The third offset is more accurate than the second offset, which further improves the image stitching accuracy.

[0081] Image stitching is performed using a second offset based on cell mask features. Since the second offset is based on cell mask features, it can effectively compensate for geometric deviations caused by factors such as motor motion, lens distortion, or stage drift, thus significantly improving the accuracy of image stitching. Based on the stitching strategy adapted to device parameters and dynamic scene, it can handle the stitching of sparse foregrounds and solve the problem of not being able to calculate the stitching offset due to missing feature points. At the same time, image stitching is performed only based on theoretically overlapping areas, which reduces the amount of computation and improves the computation speed.

[0082] Figure 12 Flowchart of the cell image stitching method provided in the embodiments of this application Figure 4 ,like Figure 12 As shown, in an optional implementation, the method may further include: S401. If the correlation coefficient between the first image block and the matching image block is greater than the preset coefficient threshold, then the image stitching is determined to be normal.

[0083] S402. If the correlation coefficient between the first image block and the matching image block is less than or equal to a preset coefficient threshold, an abnormal prompt message is pushed to indicate that the image stitching is abnormal.

[0084] If the correlation coefficient between the first image block and the matching image block is greater than the preset coefficient threshold, for example, NCC > th, it means that the registration process and calculation parameters are qualified and the image stitching is normal. If the correlation coefficient between the first image block and the matching image block is less than or equal to the preset coefficient threshold, for example, NCC ≤ th, an abnormal prompt message will be pushed to indicate that the image stitching is abnormal.

[0085] Among them, the abnormal prompts may indicate that there may be errors in the movement of the stage or abnormalities in image acquisition, prompting the operator to manually review the stitching effect and perform system maintenance if necessary, such as calibrating the stage movement accuracy and checking whether the microscope is working properly.

[0086] In this embodiment, by setting a coefficient threshold, the calculation quality of the third offset is verified. If it is unqualified, it is returned in time and prompts for manual review and system maintenance, which improves the reliability of the splicing process, avoids deviation of subsequent cell analysis results due to splicing errors, and reduces the monitoring risks in biopharmaceutical research and production.

[0087] Figure 13 This is a schematic diagram of the structure of the cell image stitching device provided in the embodiments of this application. The device can be integrated into a computer device.

[0088] like Figure 13 As shown, the device may include: The acquisition module 501 is used to acquire first and second cell images of cell culture containers taken by a microscope from different fields of view; The acquisition module 501 is also used to acquire a first offset of the second cell image relative to the first cell image based on the pixel size of the microscope, the magnification, and the scanning step size of the cell culture container. The stitching module 502 is used to stitch together the first cell image and the second cell image according to the first offset.

[0089] In an optional implementation, the splicing module 502 is specifically used for: Based on the first offset, the theoretical overlap region of the first cell image and the second cell image is determined; The theoretically overlapping regions are binarized to obtain the binarized segmentation results; Based on the binarized segmentation results, the number of cells in the theoretically overlapping region is obtained; If the number of cells in the theoretically overlapping region is less than or equal to the preset number, then the first cell image and the second cell image are stitched together according to the theoretically overlapping region.

[0090] In an optional implementation, the splicing module 502 is specifically used for: If the number of cells in the theoretically overlapping region is greater than the preset number, then the first cell image and the second cell image are image registered according to the first cell mask in the theoretically overlapping region of the first cell image and the second cell mask in the theoretically overlapping region of the second cell image to obtain the second offset of the second cell image relative to the first cell image. Based on the second offset, the first cell image and the second cell image are stitched together.

[0091] In an optional implementation, the splicing module 502 is specifically used for: Based on the second offset, the initial overlapping region of the first cell image and the second cell image is determined; Determine the first image patch from the initial overlapping region in the second cell image; Based on the first image patch, the initial overlapping region in the first cell image is matched to obtain the matching image patch; Based on the first image block and the matching image block, obtain the third offset of the second cell image relative to the first cell image; Based on the third offset, the first cell image and the second cell image are stitched together.

[0092] In an optional implementation, the splicing module 502 is specifically used for: From the initial overlapping region in the first cell image, determine an image block that meets the preset cell density as the second image block; Based on the first image block, a matching image block is obtained by matching the outer region of the second image block.

[0093] In an optional implementation, the splicing module 502 is specifically used for: Based on the first image patch, the region with the highest correlation coefficient with the first image patch is determined by traversing the initial overlapping region in the first cell image.

[0094] In an optional embodiment, the device further includes: The determination module 503 is used to determine that the image stitching is normal if the correlation coefficient between the first image block and the matching image block is greater than a preset coefficient threshold. The push module 504 is used to push an abnormal prompt message if the correlation coefficient between the first image block and the matching image block is less than or equal to a preset coefficient threshold, so as to indicate that the image stitching is abnormal.

[0095] The processing flow of each module in the device and the interaction flow between each module can be referred to the relevant descriptions in the above method embodiments, and will not be detailed here.

[0096] Figure 14A schematic diagram of the structure of the computer device provided in the embodiments of this application, such as... Figure 14 As shown, the device may include a processor 601, a memory 602, and a bus 603. The memory 602 stores machine-readable instructions that can be executed by the processor 601. When the computer device is running, the processor 601 communicates with the memory 602 through the bus 603, and the processor 601 executes the machine-readable instructions to perform the above-described method.

[0097] This application also provides a computer-readable storage medium storing a computer program, which is executed by a processor to perform the above-described method.

[0098] In this embodiment, the computer program, when run by the processor, can also execute other machine-readable instructions to perform other methods as described in the embodiments. For details on the specific execution steps and principles, please refer to the description of the embodiments, which will not be repeated here.

[0099] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0100] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0101] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0102] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0103] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0104] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for stitching cell images, characterized in that, include: Acquire first and second cell images of cell culture vessels from different fields of view using a microscope; Based on the pixel size and magnification of the microscope and the scanning step size of the cell culture container, a first offset of the second cell image relative to the first cell image is obtained; Based on the first offset, the first cell image and the second cell image are stitched together.

2. The method according to claim 1, characterized in that, The step of stitching the first cell image and the second cell image together according to the first offset includes: Based on the first offset, the theoretical overlap region between the first cell image and the second cell image is determined; The theoretically overlapping regions are binarized to obtain binarized segmentation results; Based on the binarized segmentation results, the number of cells in the theoretically overlapping region is obtained; If the number of cells in the theoretically overlapping region is less than or equal to a preset number, then the first cell image and the second cell image are stitched together according to the theoretically overlapping region.

3. The method according to claim 2, characterized in that, The step of stitching the first cell image and the second cell image together according to the first offset further includes: If the number of cells in the theoretically overlapping region is greater than the preset number, then based on the first cell mask in the theoretically overlapping region of the first cell image and the second cell mask in the theoretically overlapping region of the second cell image, image registration is performed on the first cell image and the second cell image to obtain the second offset of the second cell image relative to the first cell image. Based on the second offset, the first cell image and the second cell image are stitched together.

4. The method according to claim 3, characterized in that, The step of stitching the first cell image and the second cell image together according to the second offset includes: Based on the second offset, the initial overlapping region of the first cell image and the second cell image is determined; A first image patch is determined from the initial overlapping region in the second cell image; Based on the first image patch, a matching image patch is obtained by matching the initial overlapping region in the first cell image; Based on the first image block and the matching image block, obtain the third offset of the second cell image relative to the first cell image; Based on the third offset, the first cell image and the second cell image are stitched together.

5. The method according to claim 4, characterized in that, The step of matching the initial overlapping region in the first cell image based on the first image patch to obtain a matched image patch includes: From the initial overlapping region in the first cell image, determine an image block that meets a preset cell density as a second image block; Based on the first image block, a matching image block is obtained by matching the outer region of the second image block.

6. The method according to claim 4, characterized in that, The step of matching the initial overlapping region in the first cell image based on the first image patch to obtain a matched image patch includes: Based on the first image patch, the region with the highest correlation coefficient with the first image patch is determined by traversing the initial overlapping region in the first cell image as the matching image patch.

7. The method according to claim 6, characterized in that, The method further includes: If the correlation coefficient between the first image block and the matching image block is greater than a preset coefficient threshold, then the image stitching is determined to be normal. If the correlation coefficient between the first image block and the matching image block is less than or equal to the preset coefficient threshold, an abnormal prompt message will be pushed to indicate that the image stitching is abnormal.

8. A cell image stitching device, characterized in that, include: The acquisition module is used to acquire first and second cell images of cell culture containers taken under a microscope from different fields of view. The acquisition module is further configured to acquire a first offset of the second cell image relative to the first cell image based on the pixel size, magnification of the microscope and the scanning step size of the cell culture container; The stitching module is used to stitch the first cell image and the second cell image together according to the first offset.

9. A computer device, characterized in that, include: The computer device includes a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the method according to any one of claims 1 to 7.

Images