Cell image registration method, device, equipment, storage medium and program product

Abstract

The invention discloses a cell image registration method and device, electronic equipment, a storage medium and a computer program product. The method comprises the following steps: carrying out image acquisition on fluorescent microspheres for replacing a cell sample to obtain a bright field image and a fluorescent image; dividing the bright field image and the fluorescence image to obtain a plurality of area image pairs; wherein each area image pair comprises a bright field area image and a fluorescent area image which correspond to each other in position; performing centroid detection on the regional image pairs to obtain a bright field centroid point set and a fluorescent centroid point set of each regional image pair; performing centroid local matching on the bright field centroid point set and the fluorescent centroid point set to obtain local offset of each regional image pair; and carrying out fusion processing on the local offsets of all regional image pairs to obtain a global registration offset. According to the method, the problem that the accuracy and the stability of a registration result are affected due to matching errors or matching failures in the registration process can be solved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, specifically to a cell image registration method, apparatus, electronic device, readable storage medium, and computer program product. Background Technology

[0002] In cell biology and drug screening research, precise registration of bright-field and fluorescence images is often required to fuse cell structure and molecular marker information, thereby enabling comprehensive analysis of cell morphology, localization, and metabolic activities. Currently widely used registration methods primarily rely on feature-point-based methods. These methods extract significant features such as corners and edges from the images, establish correspondences between feature points, and calculate a global transformation matrix to achieve registration. However, due to significant differences in imaging mechanisms between bright-field and fluorescence images, their texture features and structural expressions often lack clear correspondence, making it difficult to extract a sufficient number of reliable feature points in multimodal scenarios. This limitation makes existing methods prone to matching errors or failures during registration, thus affecting the accuracy and stability of the registration results. Summary of the Invention

[0003] In view of the above problems, this application provides a cell image registration method, apparatus, electronic device, readable storage medium, and computer program product, which can solve the problem that matching errors or failures occur during the registration process, thereby affecting the accuracy and stability of the registration results.

[0004] In a first aspect, this application provides a cell image registration method, comprising: Images of fluorescent microspheres that served as substitute cell samples were acquired, yielding bright-field and fluorescence images; The bright-field image and the fluorescence image are divided to obtain multiple region image pairs; wherein each region image pair includes a bright-field region image and a fluorescence region image corresponding to the location; Centroid detection is performed on the region image pairs to obtain the bright field centroid set and fluorescence centroid set for each region image pair; Local centroid matching is performed on the bright-field centroid set and the fluorescence centroid set to obtain the local offset of each region image pair; The local offsets of all the image pairs in the region are fused to obtain the global registration offset.

[0005] In the above technical solution, this method can effectively solve the problems of difficult feature point extraction and unreliable matching caused by the difference in imaging mechanism between bright field images and fluorescence images. By using centroid detection with fluorescent microspheres as a unified reference and offset calculation logic that combines local and global approaches, the accuracy and stability of cell image registration results are significantly improved, providing reliable data support for the precise fusion analysis of cell structure information and molecular marker information in cell biology research and drug screening.

[0006] In some embodiments, the multiple bright-field regions in the bright-field image have the same image size; The multiple fluorescent regions in the fluorescence image have the same image size.

[0007] In the above technical solution, the method can ensure that the region division rules of the two types of images are consistent, providing a standardized comparison basis for the centroid detection and local offset calculation of the corresponding region image pairs, and avoiding the imbalance of local registration reference caused by the difference in region size.

[0008] In some embodiments, the centroid detection of the region image pairs to obtain the bright-field centroid set and fluorescence centroid set for each region image pair includes: Each of the aforementioned region image pairs is converted to grayscale to obtain a first processed region image pair; Binarize each first processing region image pair to obtain a second processing region image pair; Microsphere contour detection is performed on each of the second processing region image pairs to obtain microsphere contour image pairs including the outer contour of the microsphere; wherein, the microsphere contour image pair includes a bright field microsphere contour region image and a fluorescent microsphere contour region image corresponding to the position; Centroid detection is performed on the microsphere contour image pairs to obtain the bright-field centroid set and fluorescence centroid set for each region image pair.

[0009] In the above technical solution, this method can systematically eliminate redundant grayscale information and background noise interference in bright-field and fluorescence images through a progressive image processing flow of "grayscale conversion → binarization → microsphere contour detection → centroid detection," accurately separating the effective contour region of fluorescent microspheres. Specifically, grayscale conversion and binarization simplify the image information dimensions and highlight the brightness difference between the microspheres and the background; targeted microsphere contour detection can accurately locate the target region, avoiding interference from cell structure or other impurities on centroid identification, thereby ensuring that the bright-field centroid point set and the fluorescence centroid point set extracted from the contour image accurately correspond to the actual position of the fluorescent microspheres.

[0010] In some embodiments, the centroid detection of the microsphere contour image pairs to obtain the bright-field centroid set and fluorescence centroid set for each region image pair includes: Calculate the microsphere contour geometric features for each of the region image pairs; wherein, the microsphere contour geometric features include a first geometric feature of the outer contour of each microsphere in the bright-field microsphere contour region image, and a second geometric feature of the outer contour of each microsphere in the fluorescent microsphere contour region image, wherein both the first geometric feature and the second geometric feature include at least the zero-order moment and the first-order moment of the outer contour of the microsphere; Based on the first geometric feature, the centroid coordinates of the outer contour of each microsphere in the bright-field microsphere contour region image are calculated to obtain the bright-field centroid point set of the region image pair; Based on the second geometric feature, the centroid coordinates of the outer contour of each microsphere in the image of the fluorescent microsphere contour region are calculated to obtain the set of fluorescent centroid points of the region image pair.

[0011] In the above technical solution, the method can rely on the mathematical characteristics of the method of moments to transform the spatial morphological information of the microsphere contour into precise numerical parameters, avoiding the dependence on texture and edges in manual subjective judgment or traditional feature extraction, and fundamentally eliminating the interference of the difference between bright field and fluorescence image modes on centroid localization; at the same time, the combined calculation of the zero-order moment (reflecting the contour area / sum of gray levels) and the first-order moment (reflecting the centroid of the contour gray level distribution) can also accurately anchor the geometric center of the microsphere contour, thereby ensuring that the coordinate values ​​of the bright field centroid point set and the fluorescence centroid point set strictly correspond to the actual spatial position of the microsphere.

[0012] In some implementations, the step of performing centroid local matching on the bright-field centroid set and the fluorescence centroid set to obtain the local offset of each region image pair includes: Obtain preset matching algorithm parameters; wherein, the matching algorithm parameters include at least the maximum number of iterations and the interior point judgment threshold; Based on the matching algorithm parameters, local centroid matching is performed on the bright field centroid set and fluorescence centroid set of each region image pair to obtain the local offset of each region image pair.

[0013] In the above technical solution, this method can avoid inefficiency caused by getting stuck in local optima or redundant calculations in the matching process by setting the maximum number of iterations, ensuring that point set matching is completed within a reasonable time. It can also effectively filter out reliable matching pairs (interior points) in the bright field and fluorescence centroid point sets by setting an interior point judgment threshold, and eliminate erroneous matching pairs (exterior points) caused by noise or microsphere detection errors, reducing the interference of abnormal data on local offset calculations. Finally, relying on parameterized matching logic, the output stability of accurate local offsets for each region image pair can also be improved.

[0014] In some implementations, the step of performing local centroid matching on the bright-field centroid set and fluorescence centroid set of each region image pair according to the matching algorithm parameters to obtain the local offset of each region image pair includes: Initialize the global optimal inlier ratio for each region image pair; From the set of bright-field centroids and the set of fluorescent centroids of each region image pair, obtain the set of matching point pairs for each region image pair; Calculate the affine transformation matrix for each region image pair based on the matching point pairs; The coordinates of all centroids in the bright field centroid set are transformed according to the affine transformation matrix to obtain the transformed centroid set of each region image pair; Calculate the Euclidean distance between each transformed centroid in the transformed centroid set and its nearest neighbor in the fluorescent centroid set to obtain the Euclidean distance set for each region image pair; Based on the Euclidean distance set and the interior point determination threshold, determine the interior point set for each region image pair; The proportion of interior points in each region image pair is calculated based on the number of interior points in the interior point set and the total number of points in the transformed centroid point set. When the inlier ratio is greater than the currently stored global optimal inlier ratio, the inlier ratio is stored as the new global optimal inlier ratio for each region image pair, and the affine transformation matrix of the iteration corresponding to the inlier ratio is determined as the optimal affine transformation matrix for each region image pair. When the current iteration count reaches the maximum iteration count, the final optimal affine transformation matrix is ​​determined as the target affine transformation matrix for each region image pair; The local offset of each region image pair is generated based on the last column of the target affine transformation matrix.

[0015] In the above technical solution, this method can continuously filter out the affine transformation matrix that best reflects the true registration relationship by initializing and iteratively updating the globally optimal inlier ratio, thus avoiding random errors in single calculations. It can also achieve centroid coordinate mapping using the affine transformation matrix and filter inlier points using Euclidean distance and inlier thresholds, accurately eliminating outliers caused by microsphere detection bias and image noise, ensuring the reliability of matching point pairs. Simultaneously, generating local offsets based on the optimal affine transformation matrix at the maximum number of iterations not only ensures the convergence and efficiency of the matching process but also allows the local offsets to strictly conform to the actual positional correspondence of microspheres within the region, providing highly reliable local data for subsequent global registration offset fusion.

[0016] In some implementations, obtaining a set of matching point pairs for each region image pair from the set of bright-field centroids and the set of fluorescence centroids for each region image pair includes: A Gaussian distribution algorithm based on image center is used to randomly select point pairs from the bright field centroid set and the fluorescence centroid set of each region image pair to obtain the matching point pair set of each region image pair; Alternatively, multiple sets of nearest neighbor point pairs can be randomly selected from the bright-field centroid set and the fluorescence centroid set of each region image pair to obtain the matching point pair set of each region image pair.

[0017] In the above technical solution, the method can select centroid pairs that are closer to the core area of ​​the image and more representative based on the Gaussian distribution of the image center, reduce the interference of noise points in the edge area, and adapt to image scenes with relatively concentrated cell distribution; nearest neighbor selection can directly filter out the point pairs with the strongest positional correlation based on spatial distance, quickly establish a preliminary matching relationship, and adapt to scenes where the microspheres are relatively dispersed but have clear spatial correspondence.

[0018] In some implementations, the fusion of local offsets for all said region image pairs to obtain a global registration offset includes: Obtain the weight coefficients for each of the region image pairs; The global registration offset is obtained by weighted averaging and fusing the local offsets of all the image pairs in the region according to the weight coefficients.

[0019] In the above technical solution, the method can dynamically allocate weights based on factors such as image quality (e.g., microsphere clarity, noise level) and number of microspheres in different regions, so that the offset of regions with higher matching accuracy and stronger information reliability occupies a higher weight in the global calculation, reducing the interference of low-quality region data on the global result.

[0020] In some embodiments, the method further includes: Based on the global registration offset, the local offset of each region image pair is detected using feedback, and the fitness score of each local offset is obtained, and multiple fitness scores are output.

[0021] In the above technical solution, the method can achieve fine-grained monitoring of the registration results, thereby enhancing the interpretability and debuggability of the method, and helping to quickly locate and solve local registration anomalies in practical applications.

[0022] Secondly, this application provides a cell image registration device, comprising: The image acquisition unit is used to acquire images of fluorescent microspheres that are substitutes for cell samples, and to obtain bright-field images and fluorescence images; A region segmentation unit is used to segment the bright-field image and the fluorescence image to obtain multiple region image pairs; wherein each region image pair includes a bright-field region image and a fluorescence region image corresponding to the location; The centroid detection unit is used to perform centroid detection on the region image pairs to obtain the bright field centroid set and the fluorescence centroid set for each region image pair. A centroid local matching unit is used to perform centroid local matching on the bright field centroid point set and the fluorescence centroid point set to obtain the local offset of each region image pair; The fusion unit is used to fuse the local offsets of all the image pairs in the region to obtain the global registration offset.

[0023] In the above technical solution, the device can effectively solve the problems of difficult feature point extraction and unreliable matching caused by the difference in imaging mechanism between bright field images and fluorescence images. By using centroid detection with fluorescent microspheres as a unified reference and offset calculation logic that combines local and global approaches, the device can significantly improve the accuracy and stability of cell image registration results, and provide reliable data support for the precise fusion analysis of cell structure information and molecular marker information in cell biology research and drug screening.

[0024] Thirdly, this application provides an electronic device including a memory and a processor, the memory storing a computer program, and the processor running the computer program to cause the electronic device to perform the cell image registration method described in any one of the first aspects.

[0025] Fourthly, this application provides a readable storage medium storing a computer program, which, when executed by a processor, performs the cell image registration method described in any one of the first aspects.

[0026] Fifthly, this application provides a computer program product comprising a computer program that, when executed by a processor, performs the cell image registration method described in any one of the first aspects.

[0027] The beneficial effects of this application are as follows: it enables accurate extraction of microsphere centroids through morphological operations combined with contour analysis and geometric center calculation, without relying on traditional feature point detection algorithms. Furthermore, using independent binarization thresholds for each sub-region yields more accurate microsphere contours. Finally, performing RANSAC matching only within each corresponding sub-region significantly reduces computational cost, taking advantage of the fact that the number of centroids in local regions is much smaller than the total number of centroids in the entire image. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application 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.

[0029] Figure 1 This is a flowchart illustrating the cell image registration method in some embodiments of this application; Figure 2 This is a flowchart illustrating the cell image registration method in some embodiments of this application; Figure 3 This is a schematic diagram showing the positional deviation of microspheres under bright field and fluorescent environments in some embodiments of this application; Figure 4 Examples of image segmentation in some embodiments of this application; Figure 5 This is an example diagram of the bright field and fluorescence image processing flow in some embodiments of this application; Figure 6 Example diagrams of cell outline centroid detection in bright-field and fluorescence images in some embodiments of this application; Figure 7 This is a schematic diagram of the structure of the cell image registration device in some embodiments of this application; Figure 8 This is a schematic diagram of the structure of an electronic device in some embodiments of this application. Detailed Implementation

[0030] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0032] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more (including two), similarly, "multiple sets" refers to two or more sets (including two sets), and "multiple pieces" refers to two or more pieces (including two pieces) unless otherwise explicitly defined.

[0033] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0034] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0035] Due to physical offsets during the imaging process, bright-field and fluorescence images exhibit spatial positional discrepancies. Currently, to achieve accurate image registration, experiments typically incorporate fluorescent microspheres (which appear as regular circles or bright spots in the image and are visible in both bright-field and fluorescence images). By detecting the corresponding positions of the microspheres in the bright-field and fluorescence images, the spatial positional discrepancy between the two images is calculated, thus completing image registration. Existing microsphere registration methods mainly include manual annotation and image-based registration methods.

[0036] However, while manual annotation is simple and intuitive, requiring no complex algorithms, it is labor-intensive, inefficient, and easily influenced by the operator's subjectivity.

[0037] Among image-based registration methods, feature-point-based registration methods (such as the classic SIFT and ORB algorithms) often fail to detect feature points with good correspondence in multimodal images; while deep learning-based registration methods often require a large amount of training data for model training, and the trained models usually have poor interpretability.

[0038] To address the aforementioned technical issues, this application provides a microsphere registration method for bright-field and fluorescence images based on region segmentation and centroid detection. This method mainly includes the following four registration steps: region segmentation; centroid detection; local centroid matching; and matching result fusion.

[0039] Building upon this foundation, this method achieves accurate extraction of microsphere centroids through morphological operations combined with contour analysis and geometric center calculation, without relying on traditional feature point detection algorithms. Furthermore, using independent binarization thresholds for each sub-region yields even more accurate microsphere contours. Finally, performing RANSAC matching only within each corresponding sub-region significantly reduces computational cost, taking advantage of the fact that the number of centroids in local regions is much smaller than the total number of centroids in the entire image.

[0040] like Figure 1 As shown, some embodiments of this application provide a cell image registration method, which includes: S101. Image acquisition is performed on the fluorescent microspheres that replace the cell sample to obtain bright-field images and fluorescence images; S102. Divide the bright-field image and fluorescence image to obtain multiple region image pairs; wherein, each region image pair includes a bright-field region image and a fluorescence region image corresponding to the location; S103. Perform centroid detection on the region image pairs to obtain the bright field centroid set and fluorescence centroid set for each region image pair; S104. Perform local centroid matching on the bright field centroid set and the fluorescence centroid set to obtain the local offset of each region image pair; S105. The local offsets of all region image pairs are fused to obtain the global registration offset.

[0041] In some embodiments, fluorescent microspheres refer to tiny particles with fluorescent properties, uniform size, and stable shape (typically in the nanometer to micrometer range in diameter, appearing as regular circles or bright spots in images, visible in both bright-field and fluorescent images). Their core function is to be identifiable in bright-field images through morphological / grayscale differences and detected in fluorescent images through specific fluorescence signals, thereby bridging the gap between the two types of image modalities.

[0042] In some embodiments, a bright-field image refers to a conventional optical imaging image, which is generated by utilizing the difference in absorption and transmission of light as it passes through a cell sample. High-density structures within the cell (such as the nucleus) absorb more light, appearing as darker areas; while the background and low-density structures transmit more light, appearing as brighter areas. This image primarily reflects the overall morphology, outline, and other structural information of the cell.

[0043] In some embodiments, a fluorescence image refers to an image acquired using fluorescence imaging technology, which needs to be used in conjunction with a fluorescent label (such as the fluorescent microspheres in this method, or other molecularly specific fluorescent probes). The fluorescent label emits fluorescence when illuminated by excitation light of a specific wavelength, and the imaging system captures this fluorescence signal to generate an image. This image can specifically display the location of the label (such as the distribution of the fluorescent microspheres).

[0044] In some embodiments, a region image pair refers to a one-to-one combination of regions formed by dividing a bright-field image and a fluorescence image according to the same rules. Each combination includes "one bright-field region image" and "one fluorescence region image". For example, if the bright-field image is uniformly divided into 10×10 small regions, and the fluorescence image is also divided into 10×10 small regions with the same coordinate range and size, the "first row and first column region" of the bright-field image and the "first row and first column region" of the fluorescence image constitute a region image pair.

[0045] In some embodiments, centroid detection refers to the process of calculating the geometric center coordinates of the target (in this method, the outline of a fluorescent microsphere) using image algorithms. The core of this process is quantization calculation using "image moments" (such as the zeroth moment reflecting the target area and the first moment reflecting the centroid of the target's grayscale distribution). The specific process is as follows: first, the region image is grayscaled and binarized to highlight the microsphere outline; then, the microsphere's range is locked through outline detection; and finally, the centroid coordinates of this outline are calculated. The aim is to transform the "microsphere outline" into precise "point coordinates" (i.e., the centroid point), providing standardized feature points for subsequent point set matching.

[0046] In some embodiments, the bright-field centroid set refers to the set of centroid coordinates of all microspheres corresponding to the bright-field region image obtained by centroid detection in a single region image pair; the fluorescence centroid set refers to the set of centroid coordinates of all microspheres corresponding to the fluorescence region image obtained by centroid detection in a single region image pair.

[0047] In some embodiments, centroid local matching refers to the process of establishing a mapping relationship between corresponding points in the "bright field centroid set" and the "fluorescence centroid set" within a single region image pair using an algorithm (such as the RANSAC algorithm mentioned above). The core is to filter "valid matching point pairs" (inner points), eliminate "incorrect matching point pairs" (outer points, such as the centroids of false positive microspheres caused by noise), and calculate the positional offset relationship between the bright field and fluorescence images within the region based on the valid point pairs.

[0048] In some embodiments, local offset refers to the positional offset parameter (typically a two-dimensional coordinate offset, such as Δx, Δy, or affine transformation parameters including rotation and scaling) of a bright field image relative to a fluorescence image in a single region image pair obtained through "centroid local matching".

[0049] In some embodiments, global registration offset refers to the uniform offset relationship applicable to the entire brightfield image and fluorescence image obtained after weighted averaging and fusing the "local offsets" of all region image pairs.

[0050] In the above embodiments, this method can effectively solve the problems of difficult feature point extraction and unreliable matching caused by the difference in imaging mechanism between bright field images and fluorescence images. By using centroid detection with fluorescent microspheres as a unified reference and offset calculation logic that combines local and global approaches, the accuracy and stability of cell image registration results are significantly improved, providing reliable data support for the precise fusion analysis of cell structure information and molecular marker information in cell biology research and drug screening.

[0051] In some embodiments, multiple bright-field regions in a bright-field image have the same image size; Multiple fluorescent regions in the fluorescence image have the same image size.

[0052] In some embodiments, the method can divide the bright-field image and fluorescence image into multiple mutually independent regions of the same size before feature point detection. This allows subsequent registration operations to be performed only between corresponding images.

[0053] In some embodiments, because the method requires calculating the distance between each microsphere centroid in the bright-field image and each microsphere in the fluorescence image during the registration stage, the computational complexity of this step is O(n^2). 2 However, after region partitioning, this method only needs to calculate the distances between microspheres within the corresponding regions, thus greatly reducing the computational load. Furthermore, the registration process for each region can be performed in parallel, further accelerating the computational speed.

[0054] For example, the method was tested in practice for the above situation: when using the same registration algorithm (RANSAC), the total registration time without region division was 10.5 seconds, and the total registration time with region division (divided into 4 rows and 6 columns, a total of 24 independent regions) was 0.6 seconds, which reduced the time by about 94.3%.

[0055] In the above embodiments, the method can ensure that the region division rules of the two types of images are consistent, providing a standardized comparison basis for the centroid detection and local offset calculation of the corresponding region image pairs, and avoiding the imbalance of local registration reference caused by the difference in region size.

[0056] In some embodiments, centroid detection is performed on region image pairs to obtain a bright-field centroid set and a fluorescence centroid set for each region image pair, including: Each region image pair is converted to grayscale to obtain the first processed region image pair; Binarize each first processing region image pair to obtain a second processing region image pair; Microsphere contour detection is performed on each second processing region image pair to obtain a microsphere contour image pair including the outer contour of the microsphere; wherein, the microsphere contour image pair includes a bright field microsphere contour region image and a fluorescent microsphere contour region image corresponding to the position; Centroid detection was performed on the microsphere contour image pairs to obtain the bright-field centroid set and fluorescence centroid set for each region image pair.

[0057] In some embodiments, due to differences in imaging and texture structure between bright-field and fluorescence images, feature points extracted by methods such as SIFT and ORB lack stable correspondences and cannot achieve reliable matching. Therefore, this paper proposes a microsphere centroid detection method applicable to both bright-field and fluorescence images. Differential processing strategies are adopted to address the structural differences between the two types of images.

[0058] In some embodiments, the method first performs grayscale conversion on the bright-field image and the fluorescence image. In this case, the edges of the microspheres in the bright-field image are black, and the rest are white; the microsphere region in the fluorescence image is white, and the rest is black. Based on this, the method can perform a binarization inversion operation on the bright-field image (setting values ​​greater than a threshold as black, and values ​​less than a threshold as white), so that in the processed binarized image, the edges of the microspheres are converted to white, and the rest are black. Similarly, the fluorescence image is directly binarized (setting values ​​greater than a threshold as white, and values ​​less than a threshold as black), so that in the processed image, the microsphere region is converted to white, and the rest is black.

[0059] Based on this, this method can detect the external contour of microspheres using a contour detection algorithm on binarized bright-field and fluorescence images. Then, for the external contour of the microspheres, the zeroth moment (M) of the contour is calculated. 00 ) and first moment (M 10 and M 01 Find the centroid coordinates of the contour: .

[0060] In the above embodiments, this method can systematically eliminate redundant grayscale information and background noise interference in bright-field and fluorescence images through a progressive image processing flow of "grayscale conversion → binarization → microsphere contour detection → centroid detection," accurately separating the effective contour region of the fluorescent microspheres. Specifically, grayscale conversion and binarization simplify the image information dimensions and highlight the brightness difference between the microspheres and the background; targeted microsphere contour detection can also accurately locate the target region, avoiding interference from cell structure or other impurities on centroid identification, thereby ensuring that the bright-field centroid point set and the fluorescence centroid point set extracted from the contour image accurately correspond to the actual position of the fluorescent microspheres.

[0061] In some embodiments, centroid detection is performed on the microsphere contour image pairs to obtain the bright-field centroid set and fluorescence centroid set for each region image pair, including: Calculate the microsphere contour geometric features for each region image pair; wherein, the microsphere contour geometric features include the first geometric features of the outer contour of each microsphere in the bright field microsphere contour region image, and the second geometric features of the outer contour of each microsphere in the fluorescent microsphere contour region image, wherein both the first geometric features and the second geometric features include at least the zero-order moment and the first-order moment of the outer contour of the microsphere; Based on the first geometric feature, the centroid coordinates of the outer contour of each microsphere in the bright-field microsphere contour region image are calculated to obtain the bright-field centroid point set of the region image pair; Based on the second geometric feature, the centroid coordinates of the outer contour of each microsphere in the image of the fluorescent microsphere contour region are calculated to obtain the set of fluorescent centroid points of the region image pair.

[0062] In the above embodiments, this method can rely on the mathematical characteristics of the method of moments to transform the spatial morphological information of the microsphere contour into precise numerical parameters, avoiding the dependence on texture and edges in manual subjective judgment or traditional feature extraction, and fundamentally eliminating the interference of the difference between bright field and fluorescence image modes on centroid localization. At the same time, the combined calculation of the zero-order moment (reflecting the contour area / sum of gray levels) and the first-order moment (reflecting the centroid of the contour gray level distribution) can also accurately anchor the geometric center of the microsphere contour, thereby ensuring that the coordinate values ​​of the bright field centroid point set and the fluorescence centroid point set strictly correspond to the actual spatial position of the microsphere.

[0063] In some embodiments, local centroid matching is performed on the bright-field centroid set and the fluorescence centroid set to obtain the local offset of each region image pair, including: Obtain the preset matching algorithm parameters; wherein the matching algorithm parameters include at least the maximum number of iterations and the interior point judgment threshold; Based on the matching algorithm parameters, local centroid matching is performed on the bright field centroid set and fluorescence centroid set of each region image pair to obtain the local offset of each region image pair.

[0064] In some embodiments, the method can use the RANSAC (Random Sample Consensus) algorithm to match the centroid of microspheres in bright-field and fluorescence images.

[0065] In the above embodiments, this method can avoid inefficiency caused by getting stuck in local optima or redundant calculations during the matching process by setting the maximum number of iterations, ensuring that point set matching is completed within a reasonable time. It can also effectively filter out reliable matching pairs (interior points) in the bright field and fluorescence centroid point sets by setting an interior point judgment threshold, and eliminate erroneous matching pairs (exterior points) caused by noise or microsphere detection errors, reducing the interference of abnormal data on local offset calculations. Finally, relying on parameterized matching logic, the output stability of accurate local offsets for each region image pair can also be improved.

[0066] In some embodiments, local centroid matching is performed on the bright-field centroid set and fluorescence centroid set of each region image pair according to the matching algorithm parameters to obtain the local offset of each region image pair, including: Initialize the globally optimal inlier ratio for each region image pair; From the bright field centroid set and the fluorescence centroid set of each region image pair, obtain the set of matching point pairs for each region image pair; Calculate the affine transformation matrix for each region image pair based on the matching point pairs; Based on the affine transformation matrix, the coordinates of all centroids in the bright field centroid set are transformed to obtain the transformed centroid set for each region image pair. Calculate the Euclidean distance between each transformed centroid in the transformed centroid set and its nearest neighbor in the fluorescent centroid set to obtain the set of Euclidean distances for each region image pair; Based on the Euclidean distance set and the interior point judgment threshold, determine the interior point set of each region image pair; Calculate the proportion of inliers for each region image pair based on the number of inliers in the inlier set and the total number of points in the transformed centroid set. When the inlier ratio is greater than the currently stored global optimal inlier ratio, the inlier ratio is stored as the new global optimal inlier ratio for each region image pair, and the affine transformation matrix corresponding to the inlier ratio in the iteration round is determined as the optimal affine transformation matrix for each region image pair. When the current iteration count reaches the maximum iteration count, the final optimal affine transformation matrix is ​​determined as the target affine transformation matrix for each region image pair; Based on the last column of the target affine transformation matrix, generate the local offset for each region image pair.

[0067] Exemplarily, assume that the set of centroid points extracted from the bright-field image is P, and the set of centroid points extracted from the fluorescence image is Q. Set the parameters of the RANSAC algorithm, set the maximum number of iterations N and the inlier judgment threshold t (the threshold t is used to determine whether the distance between the matching points is close enough to be recognized as an inlier. This value needs to be set according to the specific size of the microspheres, etc. In this method, 2 pixels are used).

[0068] Based on this, in each round of iteration, the method randomly selects 3 pairs of nearest-neighbor matching point pairs from the bright-field point set P and the fluorescence image point set Q to form a subset {p, q} for calculating the affine transformation matrix. Among them, 3 pairs of matching points are selected because at least three non-collinear points are required to calculate the affine transformation matrix in a two-dimensional plane; at the same time, this step can also randomly select point pairs based on the Gaussian distribution centered on the image, making the points in the central region of the image more likely to be selected (due to the deviation between the bright-field and fluorescence images, the points in the edge region of the image may be missing, and the centroid points in the central region are more likely to have a stable corresponding relationship).

[0069] Then, the method uses the selected 3 pairs of points to calculate the affine transformation matrix A, which satisfies q≈A·p, and A can transform all points in the bright-field image into the coordinate system of the fluorescence image.

[0070] For all points in the bright-field image point set P, use the matrix A for transformation to obtain the point set . For each point in, search for the nearest neighbor point in the fluorescence image point set Q and calculate the Euclidean distance d. If d < t, then this point is considered an "inlier".

[0071] Count the number of inliers N inlier in this round, and calculate the proportion of its total number of points: .

[0072] If the inlier proportion of the current model is greater than that of the optimal model, update the current affine matrix to the optimal result.

[0073] After completing all iterations, output the affine transformation matrix with the highest inlier proportion as the final matching result, and extract the last column of the affine transformation matrix, which is the offset offset.

[0074] In some embodiments, extracting the last column of the affine transformation matrix essentially only retains the translation part and no longer retains other transformation information such as rotation, scaling, and shearing that may be included in the affine transformation.

[0075] In the above embodiments, this method can continuously filter out the affine transformation matrix that best reflects the true registration relationship by initializing and iteratively updating the globally optimal inlier ratio, thus avoiding random errors in single calculations. It can also achieve centroid coordinate mapping using the affine transformation matrix and filter inlier points using Euclidean distance and inlier thresholds, accurately eliminating outlier points caused by microsphere detection bias and image noise, ensuring the reliability of matching point pairs. Simultaneously, generating local offsets based on the optimal affine transformation matrix that achieves the maximum number of iterations not only ensures the convergence and efficiency of the matching process but also allows the local offsets to strictly conform to the actual positional correspondence of microspheres within the region, providing highly reliable local data for subsequent global registration offset fusion.

[0076] In some embodiments, obtaining a set of matching point pairs for each region image pair from the bright-field centroid set and the fluorescence centroid set of each region image pair includes: A Gaussian distribution algorithm based on image center is used to randomly select point pairs from the bright field centroid set and the fluorescence centroid set of each region image pair to obtain the matching point pair set of each region image pair; Alternatively, multiple sets of nearest neighbor point pairs can be randomly selected from the bright-field centroid set and the fluorescence centroid set of each region image pair to obtain the matching point pair set for each region image pair.

[0077] In the above embodiments, the method can select centroid pairs that are closer to the core area of ​​the image and more representative based on the Gaussian distribution of the image center, reduce the interference of noise points in the edge area, and adapt to image scenes with relatively concentrated cell distribution; nearest neighbor selection can directly filter out the point pairs with the strongest positional correlation based on spatial distance, quickly establish a preliminary matching relationship, and adapt to scenes where the microspheres are relatively dispersed but have clear spatial correspondence.

[0078] In some embodiments, the local offsets of all region image pairs are fused to obtain the global registration offset, including: Obtain the weight coefficients for each region image pair; The global registration offset is obtained by weighted averaging and fusing the local offsets of all region image pairs based on the weight coefficients.

[0079] For example, suppose that in the above steps the original image is divided into m×n sub-regions, the RANSAC centroid registration process is performed on each sub-region, and the local offset of that region is obtained respectively. (i,j) Therefore, the following matrix can be obtained: offset(1,1);offset(2,1)......offset(l,m); offset(2,1);offset(2,2)...offset(2,m); ………………………………………… offset(m,1);offset(m,2)...ofset(n,m).

[0080] At this point, to generate the global registration offset for the entire image, a weighted average method is used to fuse the offsets from each region. The specific calculation method is as follows: .

[0081] By default, all weights can be set to 1, meaning that each region is considered to contribute equally. If the image has slight distortion, the weight of the central region can be set to be larger, and the weight of the edge region to be smaller.

[0082] In the above embodiments, the method can dynamically allocate weights based on factors such as image quality (e.g., microsphere clarity, noise level) and number of microspheres in different regions, so that the offset of regions with higher matching accuracy and stronger information reliability occupies a higher weight in the global calculation, reducing the interference of low-quality region data on the global result.

[0083] In some embodiments, the method further includes: Based on the global registration offset, the local offset of each region image pair is detected using feedback, and the fitness score of each local offset is obtained, and multiple fitness scores are output.

[0084] In some embodiments, a quantization system may be introduced in the method to score the local offset in a feedback manner.

[0085] In the above embodiments, the method can achieve fine-grained monitoring of the registration results, thereby enhancing the interpretability and debuggability of the method and helping to quickly locate and solve local registration anomalies in practical applications.

[0086] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below. In some embodiments, such as Figure 2 As shown, the cell image registration method includes: Figure 3 A schematic diagram of microsphere positional deviation under bright field and fluorescent environments is shown.

[0087] S201. Divide the bright field image and fluorescence image into regions respectively, and divide the original image into multiple equal and independent regions.

[0088] Please refer to Figure 4 , Figure 4 An example of image segmentation is shown.

[0089] S202. Different morphological detection methods are used to obtain the microsphere contours of the bright field image and the fluorescence image, respectively. Then, the centroid coordinates of the microspheres are obtained by calculating the image moments of the contours and used as registration feature points.

[0090] Please refer to Figure 5 , Figure 5 An example diagram of a bright-field and fluorescence image processing workflow is shown.

[0091] Please refer to Figure 6 , Figure 6 An example image of cell contour centroid detection using bright-field and fluorescence images is shown.

[0092] S203. Within each sub-region, based on the centroid set extracted from the bright field and fluorescence images, the RANSAC algorithm is used for local registration estimation.

[0093] S204. Merge the local registration results of each region and calculate the overall registration relationship of the entire map.

[0094] In some embodiments, bright-field images and fluorescence images exhibit significant differences in their imaging mechanisms. This method employs differentiated microsphere detection approaches for these two image types. For bright-field images, an inversion binarization + contour detection strategy is used. For fluorescence images, a binarization + contour detection strategy is used. This approach successfully detects the contour edges of microspheres against both background backgrounds.

[0095] In some embodiments, the method first applies a more targeted adaptive binarization threshold to each independent region, resulting in more accurate binarization and facilitating subsequent contour detection. Then, during the matching process, matching is performed on corresponding sub-regions, significantly reducing computation time and improving detection efficiency. Finally, the results from multiple regions are fused, with different weights assigned to each region based on the specific imaging conditions.

[0096] Figure 7 A schematic diagram of a cell image registration device is shown. It should be understood that this device is related to... Figure 1 The method executed in the middle corresponds to the steps involved in the aforementioned method. The specific functions and effects of the device can be found in the description above. To avoid repetition, detailed descriptions are omitted here.

[0097] The cell image registration device includes: Image acquisition unit 310 is used to acquire images of fluorescent microspheres that are substitute cell samples, and obtain bright-field images and fluorescence images; The region segmentation unit 320 is used to segment the bright-field image and the fluorescence image to obtain multiple region image pairs; wherein, each region image pair includes a bright-field region image and a fluorescence region image corresponding to the location; Centroid detection unit 330 is used to perform centroid detection on region image pairs to obtain the bright field centroid set and fluorescence centroid set for each region image pair; Centroid local matching unit 340 is used to perform centroid local matching on the bright field centroid point set and the fluorescence centroid point set to obtain the local offset of each region image pair; The fusion unit 350 is used to fuse the local offsets of all region image pairs to obtain the global registration offset.

[0098] In some embodiments, multiple bright-field regions in a bright-field image have the same image size; Multiple fluorescent regions in the fluorescence image have the same image size.

[0099] In some embodiments, the centroid detection unit 330 includes: Grayscale unit 331 is used to perform grayscale processing on each region image pair to obtain a first processed region image pair; Binarization unit 332 is used to perform binarization processing on each first processing region image pair to obtain a second processing region image pair; The contour detection unit 333 is used to perform microsphere contour detection on each second processing region image pair to obtain a microsphere contour image pair including the outer contour of the microsphere; wherein, the microsphere contour image pair includes a bright field microsphere contour region image and a fluorescent microsphere contour region image corresponding to the position. The centroid detection unit 334 is used to perform centroid detection on the microsphere contour image pairs to obtain the bright field centroid set and the fluorescence centroid set for each region image pair.

[0100] In some embodiments, the centroid detection unit 334 is specifically used to calculate the microsphere contour geometric features of each region image pair; wherein, the microsphere contour geometric features include a first geometric feature of the outer contour of each microsphere in the bright field microsphere contour region image and a second geometric feature of the outer contour of each microsphere in the fluorescent microsphere contour region image, and both the first geometric feature and the second geometric feature include at least the zero-order moment and the first-order moment of the outer contour of the microsphere; Based on the first geometric feature, the centroid coordinates of the outer contour of each microsphere in the bright-field microsphere contour region image are calculated to obtain the bright-field centroid point set of the region image pair; Based on the second geometric feature, the centroid coordinates of the outer contour of each microsphere in the image of the fluorescent microsphere contour region are calculated to obtain the set of fluorescent centroid points of the region image pair.

[0101] In some embodiments, the centroid local matching unit 340 includes: The first acquisition subunit 341 is used to acquire preset matching algorithm parameters; wherein, the matching algorithm parameters include at least the maximum number of iterations and the interior point judgment threshold; The local matching subunit 342 is used to perform local centroid matching on the bright field centroid set and fluorescence centroid set of each region image pair according to the matching algorithm parameters, so as to obtain the local offset of each region image pair.

[0102] In some embodiments, the local matching subunit 342 is specifically used to initialize the globally optimal inlier ratio for each region image pair; From the bright field centroid set and the fluorescence centroid set of each region image pair, obtain the set of matching point pairs for each region image pair; Calculate the affine transformation matrix for each region image pair based on the matching point pairs; Based on the affine transformation matrix, the coordinates of all centroids in the bright field centroid set are transformed to obtain the transformed centroid set for each region image pair. Calculate the Euclidean distance between each transformed centroid in the transformed centroid set and its nearest neighbor in the fluorescent centroid set to obtain the set of Euclidean distances for each region image pair; Based on the Euclidean distance set and the interior point judgment threshold, determine the interior point set of each region image pair; Calculate the proportion of inliers for each region image pair based on the number of inliers in the inlier set and the total number of points in the transformed centroid set. When the inlier ratio is greater than the currently stored global optimal inlier ratio, the inlier ratio is stored as the new global optimal inlier ratio for each region image pair, and the affine transformation matrix corresponding to the inlier ratio in the iteration round is determined as the optimal affine transformation matrix for each region image pair. When the current iteration count reaches the maximum iteration count, the final optimal affine transformation matrix is ​​determined as the target affine transformation matrix for each region image pair; Based on the last column of the target affine transformation matrix, generate the local offset for each region image pair.

[0103] In some embodiments, the local matching subunit 342 is further configured to use a Gaussian distribution algorithm based on the image center to randomly select point pairs from the bright field centroid set and the fluorescence centroid set of each region image pair to obtain a set of matching point pairs for each region image pair. Alternatively, multiple sets of nearest neighbor point pairs can be randomly selected from the bright-field centroid set and the fluorescence centroid set of each region image pair to obtain the matching point pair set for each region image pair.

[0104] In some embodiments, the fusion unit 350 includes: The second acquisition subunit 351 is used to acquire the weight coefficients of each region image pair; The fusion subunit 352 is used to perform weighted average fusion calculation on the local offsets of all region image pairs according to the weight coefficients to obtain the global registration offset.

[0105] In some embodiments, the cell image registration device further includes: The detection unit 360 is used to perform feedback detection on the local offset of each region image pair based on the global registration offset, obtain the fitness score of each local offset, and output multiple fitness scores.

[0106] like Figure 8 As shown, this application provides an electronic device 400, which includes a processor 401 and a memory 402. The processor 401 and the memory 402 are interconnected and communicate with each other through a communication bus 403 and / or other forms of connection mechanism (not shown). The memory 402 stores a computer program that can be executed by the processor 401. When the computing device is running, the processor 401 executes the computer program to perform the method in any of the aforementioned optional implementations.

[0107] This application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the method in any of the aforementioned optional implementations.

[0108] The computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0109] This application provides a computer program product, which includes a computer program that, when run by a processor, executes the method in any of the aforementioned optional implementations.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application. In particular, as long as there is no conflict, the various technical features mentioned in the embodiments can be combined in any way. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method of cell image registration, characterized by, The method comprises the following steps: image acquisition of fluorescent microspheres in a substitute cell sample to obtain a bright field image and a fluorescent image; division of the bright field image and the fluorescent image to obtain a plurality of region image pairs; wherein each region image pair comprises a bright field region image and a fluorescent region image corresponding in position; centroid detection of the region image pairs to obtain a bright field centroid point set and a fluorescent centroid point set of each region image pair; centroid local matching of the bright field centroid point set and the fluorescent centroid point set to obtain a local offset of each region image pair; fusion processing of the local offsets of all region image pairs to obtain a global registration offset.

2. The cell image registration method according to claim 1, wherein, The bright field region images in the bright field image are of consistent size; The fluorescent region images in the fluorescent image are of consistent size.

3. The cell image registration method of claim 1, wherein, The centroid detection of the region image pairs to obtain a bright field centroid point set and a fluorescent centroid point set of each region image pair comprises the following steps: gray scale processing of each region image pair to obtain a first processed region image pair; binary processing of each first processed region image pair to obtain a second processed region image pair; microsphere contour detection of each second processed region image pair to obtain a microsphere contour image pair comprising a microsphere external contour; wherein the microsphere contour image pair comprises a bright field microsphere contour region image and a fluorescent microsphere contour region image corresponding in position; centroid detection of the microsphere contour image pair to obtain a bright field centroid point set and a fluorescent centroid point set of each region image pair.

4. The cell image registration method according to claim 3, wherein, The centroid detection of the microsphere contour image pair to obtain a bright field centroid point set and a fluorescent centroid point set of each region image pair comprises the following steps: calculation of a microsphere contour geometric feature of each region image pair; wherein the microsphere contour geometric feature comprises a first geometric feature of each microsphere external contour in the bright field microsphere contour region image and a second geometric feature of each microsphere external contour in the fluorescent microsphere contour region image, and the first geometric feature and the second geometric feature each comprise at least a zeroth moment and a first moment of the microsphere external contour; calculation of a centroid coordinate of each microsphere external contour in the bright field microsphere contour region image according to the first geometric feature to obtain a bright field centroid point set of the region image pair; calculation of a centroid coordinate of each microsphere external contour in the fluorescent microsphere contour region image according to the second geometric feature to obtain a fluorescent centroid point set of the region image pair.

5. The cell image registration method of claim 1, wherein, The centroid local matching of the bright field centroid point set and the fluorescent centroid point set to obtain a local offset of each region image pair comprises the following steps: acquisition of preset matching algorithm parameters; wherein the matching algorithm parameters comprise at least a maximum iteration number and an inlier judgment threshold value; centroid local matching of the bright field centroid point set and the fluorescent centroid point set of each region image pair according to the matching algorithm parameters to obtain a local offset of each region image pair.

6. The cell image registration method according to claim 5, wherein, The centroid local matching of the bright field centroid point set and the fluorescent centroid point set of each of the region image pairs according to the matching algorithm parameter obtains a local offset of each of the region image pairs, and includes the following steps: initializing a global optimal inlier ratio of each of the region image pairs; obtaining a matching point pair set of each of the region image pairs from the bright field centroid point set and the fluorescent centroid point set of each of the region image pairs; calculating an affine transformation matrix of each of the region image pairs according to the matching point pair; performing coordinate transformation on all centroid points in the bright field centroid point set according to the affine transformation matrix to obtain a transformed centroid point set of each of the region image pairs; calculating the Euclidean distance between each transformed centroid point in the transformed centroid point set and its nearest neighbor point in the fluorescent centroid point set to obtain a Euclidean distance set of each of the region image pairs; determining an inlier set of each of the region image pairs according to the Euclidean distance set and the inlier judgment threshold; calculating an inlier ratio of each of the region image pairs according to the inlier number of the inlier set and the total point number of the transformed centroid point set; when the inlier ratio is greater than the currently stored global optimal inlier ratio, storing the inlier ratio as a new global optimal inlier ratio of each of the region image pairs, and determining the affine transformation matrix corresponding to the iteration round of the inlier ratio as an optimal affine transformation matrix of each of the region image pairs; when the current iteration number reaches the maximum iteration number, determining the finally determined optimal affine transformation matrix as a target affine transformation matrix of each of the region image pairs; generating a local offset of each of the region image pairs according to the last column of the target affine transformation matrix.

7. The cell image registration method of claim 6, wherein, The method further includes the following steps: obtaining a matching point pair set of each of the region image pairs from the bright field centroid point set and the fluorescent centroid point set of each of the region image pairs; obtaining a matching point pair set of each of the region image pairs from the bright field centroid point set and the fluorescent centroid point set of each of the region image pairs; 8. The cell image registration method of claim 1, wherein, The method further includes the following steps: obtaining a matching point pair set of each of the region image pairs from the bright field centroid point set and the fluorescent centroid point set of each of the region image pairs; obtaining a matching point pair set of each of the region image pairs from the bright field centroid point set and the fluorescent centroid point set of each of the region image pairs; 9. The cell image registration method of claim 1, wherein, The method further includes the following steps: performing feedback detection on the local offset of each of the region image pairs based on the global registration offset to obtain an adaptability score of each of the local offsets, and outputting a plurality of the adaptability scores.

10. A cell image registration apparatus characterized by comprising: The cell image registration device includes: an image acquisition unit configured to acquire images of fluorescent microspheres that replace a cell sample to obtain a bright field image and a fluorescent image; The region division unit is configured to divide the bright-field image and the fluorescent image to obtain a plurality of region image pairs; each of the region image pairs comprises a bright-field region image and a fluorescent region image corresponding in position; The centroid detection unit is configured to perform centroid detection on the region image pairs to obtain a bright-field centroid point set and a fluorescent centroid point set of each of the region image pairs; The centroid local matching unit is configured to perform centroid local matching on the bright-field centroid point set and the fluorescent centroid point set to obtain a local offset of each of the region image pairs; The fusion unit is configured to perform fusion processing on the local offsets of all the region image pairs to obtain a global registration offset.

11. An electronic device, comprising: The electronic device comprises a memory and a processor, the memory is configured to store a computer program, and the processor is configured to run the computer program to enable the electronic device to perform the cell image registration method in any one of claims 1 to 9.

12. A readable storage medium, characterized by, The readable storage medium stores a computer program, and the computer program is configured to be run by a processor to perform the cell image registration method in any one of claims 1 to 9.

13. A computer program product, characterised in that, The computer program product comprises a computer program, and the computer program is configured to be run by a processor to perform the cell image registration method in any one of claims 1 to 9.

Images