Registration method, device, electronic device and computer readable storage medium

Abstract

The embodiment of the present invention provides a registration method, device, electronic device and computer-readable storage medium, which relate to the field of image processing technology. The method includes obtaining an original image of a pathological section staining image and a target image of the pathological section staining image, and using an optical flow method to register the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule. Thereby improving the accuracy of the registration of the pathological section staining image.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a registration method, device, electronic device and computer-readable storage medium. Background Art

[0002] Nowadays, medical image registration technology is increasingly used in the medical field. Among them, the registration of pathological section staining images is mainly based on B-spline curves, affine transformations, etc. Using this registration method, the accuracy of pathological section staining image registration needs to be improved. Summary of the invention

[0003] In view of this, the present invention provides a registration method, device, electronic device and computer-readable storage medium.

[0004] In order to achieve the above purpose, the technical solution adopted by the embodiment of the present invention is as follows:

[0005] In a first aspect, an embodiment of the present invention provides a registration method, comprising:

[0006] Acquire an original image of a pathological section staining image and a target image of a pathological section staining image;

[0007] The optical flow method is used to align the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule.

[0008] In an optional implementation, the step of registering the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule comprises:

[0009] Divide the original image of the pathological section staining image into multiple scales in the frequency domain according to the frequency spectrum information;

[0010] For the original image of the pathological section stained image and the target image of the pathological section stained image, the gradient descent method is used to optimize at each divided scale, and the optical flow deformation field is fitted respectively, and the original image of the pathological section stained image and the target image of the pathological section stained image are aligned.

[0011] In an optional embodiment, the step of aligning the original image of the pathological section stained image and the target image of the pathological section stained image includes: according to the optical flow deformation field, using a dense optical flow method to align the original image of the pathological section stained image and the target image of the pathological section stained image pixel by pixel.

[0012] In an optional embodiment, the method further includes: using a bilinear interpolation method to assist the dense optical flow method to register the original image of the pathological section staining image and the target image of the pathological section staining image.

[0013] In an optional embodiment, before using the optical flow method to align the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules, the method also includes: using affine transformation to pre-align the original image of the pathological section stained image with reference to the target image of the pathological section stained image.

[0014] In an optional embodiment, the step of pre-registering the original image of the pathological section staining image with reference to the target image of the pathological section staining image using affine transformation comprises:

[0015] Estimate an affine transformation matrix based on pre-selected paired points;

[0016] The original image of the pathological section staining image is pre-registered with reference to the target image of the pathological section staining image according to the affine transformation matrix.

[0017] In an optional implementation, before registering the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule using an optical flow method, the method further includes:

[0018] grayscale the original image of the pathological section staining image and the target image of the pathological section staining image;

[0019] The original image of the pathological section staining image after grayscale processing and the target image of the pathological section staining image are filled to the same size.

[0020] In an optional embodiment, the method further comprises: outputting the image after the registration is completed; and / or

[0021] Output the optical flow deformation field.

[0022] In a second aspect, an embodiment of the present invention provides a registration device, including:

[0023] An image acquisition module, used to acquire an original image of a pathological section staining image and a target image of a pathological section staining image;

[0024] The image registration module is used to register the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule by using an optical flow method.

[0025] In an optional implementation, the image registration module is used to register the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule through the following steps:

[0026] Divide the original image of the pathological section staining image into multiple scales in the frequency domain according to the frequency spectrum information;

[0027] For the original image of the pathological section stained image and the target image of the pathological section stained image, the gradient descent method is used to optimize at each divided scale, and the optical flow deformation field is fitted respectively, and the original image of the pathological section stained image and the target image of the pathological section stained image are aligned.

[0028] In an optional implementation, the image registration module is used to register the original image of the pathological section staining image and the target image of the pathological section staining image pixel by pixel using a dense optical flow method according to the optical flow deformation field.

[0029] In an optional embodiment, the registration device also includes a pre-registration module, which is used to pre-register the original image of the pathological section stained image with reference to the target image of the pathological section stained image using affine transformation before using the optical flow method to align the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules.

[0030] In an optional embodiment, the registration device also includes a preprocessing module for graying the original image of the pathological section stained image and the target image of the pathological section stained image before registering the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules using an optical flow method, and filling the original image of the pathological section stained image and the target image of the pathological section stained image after the graying process to the same size.

[0031] In a third aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor can execute the machine executable instructions to implement the method described in any one of the aforementioned implementations.

[0032] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements a method as described in any one of the aforementioned embodiments.

[0033] The registration method, device, electronic device and computer-readable storage medium provided in the embodiments of the present invention cleverly apply the optical flow method to the registration of pathological section stained images. By using the optical flow method to register the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules, the accuracy of the pathological section stained image registration is improved.

[0034] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A block diagram of an electronic device provided by an embodiment of the present invention is shown.

[0037] Figure 2 A schematic flow chart of a registration method provided by an embodiment of the present invention is shown.

[0038] Figure 3 A schematic diagram of a process of S120 provided in an embodiment of the present invention is shown.

[0039] Figure 4 A schematic flow chart of a registration method in an exemplary scene provided by an embodiment of the present invention is shown.

[0040] Figure 5 A CK staining image provided by an embodiment of the present invention is shown.

[0041] Figure 6 A PDL1 staining image provided by an embodiment of the present invention is shown.

[0042] Figure 7 A pre-registered image provided by an embodiment of the present invention is shown.

[0043] Figure 8 A CK staining image after being grayed and filled to the same size provided by an embodiment of the present invention is shown.

[0044] Fig. 9 A PDL1 dyeing image after being grayed and filled to the same size provided by an embodiment of the present invention is shown.

[0045] Fig.10An image after registration provided by an embodiment of the present invention is shown.

[0046] Fig.11 A block diagram of a registration device provided by an embodiment of the present invention is shown.

[0047] Icon: 100 - electronic device; 110 - memory; 120 - processor; 130 - communication module; 140 - registration device; 141 - image acquisition module; 142 - image registration module; 143 - pre-registration module; 144 - pre-processing module; 145 - information output module. DETAILED DESCRIPTION

[0048] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0049] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0050] It should be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0051] Please refer to Figure 1 , is a block diagram of an electronic device 100. The electronic device 100 in the embodiment of the present invention may be an image processing server, an image processing platform, etc. capable of performing image registration. The electronic device 100 includes a memory 110, a processor 120, and a communication module 130. The memory 110, the processor 120, and the communication module 130 are electrically connected to each other directly or indirectly to achieve data transmission or interaction. For example, these components may be electrically connected to each other via one or more communication buses or signal lines.

[0052] The memory 110 is used to store programs or data. The memory 110 may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc.

[0053] The processor 120 is used to read and write data or programs stored in the memory 110 and execute corresponding functions.

[0054] The communication module 130 is used to establish a communication connection between the electronic device 100 and other communication terminals through the network, and to send and receive data through the network.

[0055] It should be understood that Figure 1 The structure shown is only a schematic diagram of the structure of the electronic device 100. The electronic device 100 may also include Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown. Figure 1 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0056] See also Figure 2 , is a flow chart of a registration method provided by an embodiment of the present invention, and the registration method can be applied to Figure 1 The electronic device 100 shown may be implemented by the processor 120 in the electronic device 100. The registration method includes S110 and S120.

[0057] S110, obtaining an original image of the pathological section staining image and a target image of the pathological section staining image.

[0058] S120, using an optical flow method, registering the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule.

[0059] Optical flow is the instantaneous speed of the pixel movement of a moving object in space on the observation imaging plane. The optical flow field is a collection of two-dimensional optical flow vectors, which can describe the grayscale change trend (motion trend) of the corresponding point group on the image. Using the optical flow method for registration can support pixel-level registration. Applying the optical flow method to the registration of tissue pathology section staining images can improve the accuracy of registration between the original image of the pathology section staining image and the target image of the pathology section staining image.

[0060] In one implementation, in order to improve the accuracy of the registration, before executing S120, the original image of the pathological section staining image and the target image of the pathological section staining image are registered according to a set rule using an optical flow method, the original image of the pathological section staining image and the target image of the pathological section staining image may be preprocessed and pre-registered. The method of preprocessing and pre-registering the original image of the pathological section staining image and the target image of the pathological section staining image may be flexibly selected.

[0061] For example, the preprocessing process may include: graying the original image of the pathological section staining image and the target image of the pathological section staining image, and filling the original image of the pathological section staining image and the target image of the pathological section staining image after the graying process to the same size.

[0062] For example, the pre-registration process may include: pre-registering the original image of the pathological section staining image with reference to the target image of the pathological section staining image. In one implementation, the original image of the pathological section staining image may be pre-registered with reference to the target image of the pathological section staining image using an affine transformation. An embodiment of the present invention provides an optional implementation of pre-registering the original image of the pathological section staining image with reference to the target image of the pathological section staining image using an affine transformation: estimating an affine transformation matrix based on pre-selected paired points, and pre-registering the original image of the pathological section staining image with reference to the target image of the pathological section staining image according to the affine transformation matrix. In other implementations, the original image of the pathological section staining image may also be pre-registered with reference to the target image of the pathological section staining image using a frequency domain estimation method, a center estimation method, a histogram pairing method, or the like. The embodiment of the present invention does not limit this.

[0063] Among them, the pre-processing process can be executed after the pre-registration process, and the original image of the pathological section staining image and the target image of the pathological section staining image are pre-registered and then grayed and filled to the same size, thereby improving the accuracy and efficiency of subsequent registration. It is understandable that in other embodiments, only one of the pre-processing and pre-registration processes can be executed. For example, the registration operation in S120 is executed after the original image of the pathological section staining image and the target image of the pathological section staining image are pre-processed. For another example, the registration operation in S120 is executed after the original image of the pathological section staining image and the target image of the pathological section staining image are pre-registered. For another example, the pre-processing and pre-registration processes can be omitted, and the registration operation in S120 can be directly executed. The embodiments of the present invention do not limit this.

[0064] In order to quickly optimize the optical flow deformation field of the original image of the pathological section stained image for the target image of the pathological section stained image, and thus obtain a more accurate registration image, in an embodiment of the present invention, the original image of the pathological section stained image can be divided into multiple scales, and the optical flow deformation field can be fitted at different scales to improve the fitting speed and accuracy of the deformation field.

[0065] Please refer to Figure 3 In S120, the optical flow method is used to align the original image of the pathological section staining image and the target image of the pathological section staining image according to the set rules, which can be achieved through S121 and S122.

[0066] S121, dividing the original image of the pathological section staining image into multiple scales in the frequency domain according to the frequency spectrum information.

[0067] S122, for the original image of the pathological section staining image and the target image of the pathological section staining image, a gradient descent method is used to optimize them at each divided scale, and optical flow deformation field is fitted respectively, so as to align the original image of the pathological section staining image and the target image of the pathological section staining image.

[0068] In the embodiment of the present invention, the optional levels of scale division include but are not limited to linear type (such as 2, 4, 6, 8), exponential type (such as 2, 4, 8, 16), a specific level sequence determined according to frequency domain information, etc.

[0069] In view of the fact that the optical flow field can be divided into dense optical flow and sparse optical flow according to the number of sampling points in the optical flow field, wherein the dense optical flow can record the offset of each point of the entire image, and can be aligned at the pixel level with high accuracy, so in order to further improve the accuracy of the alignment, the step of aligning the original image of the pathological section stained image and the target image of the pathological section stained image in S122 may include: according to the optical flow deformation field, using the dense optical flow method to align the original image of the pathological section stained image and the target image of the pathological section stained image pixel by pixel.

[0070] In one implementation, a bilinear interpolation method may be used to assist the dense optical flow method in registering the original image of the pathological section staining image and the target image of the pathological section staining image, so as to improve the registration efficiency.

[0071] After the image registration is implemented using the above method, at least one of the registered image and the optical flow deformation field can be output.

[0072] Compared with the traditional registration technology, the registration method in the embodiment of the present invention uses the dense optical flow method for image registration, which can obtain more accurate registration images, make the spatial correspondence of the same tissue on different slices stronger, and effectively reduce the phenomenon of tissue dislocation. By dividing the image into multiple scales, multi-stage optical flow deformation field fitting is realized, which has higher computational efficiency than directly fitting the entire image in a single stage. At the same time, the local optimal result in the optimization process can be effectively ignored, thereby improving the accuracy of registration.

[0073] In order to more clearly illustrate the implementation scheme in the embodiment of the present invention, taking the original image as a CK (cytokeratinCK) stained image and the target image as a PDL1 (Programmed cell death 1 ligand 1) stained image as an example, the implementation process of using the optical flow method to align the original image and the target image is illustrated.

[0074] like Figure 4 As shown, the registration process is as follows:

[0075] The CK staining image of the pathological section is taken as the original image, and the PDL1 staining image is taken as the target image for registration. Figure 5 As shown, PDL1 staining images are shown Figure 6 shown.

[0076] The CK stained image is pre-registered with the PDL1 stained image using the affine transformation method based on the pre-selected paired points to optimize and improve the accuracy of optical flow registration. Figure 7 shown.

[0077] The pre-registered CK staining image and PDL1 staining image are grayed and filled to the same size. The grayed and filled CK staining image is as follows: Figure 8 As shown in Figure 2, the PDL1 staining image after grayscale and filling to the same size is as follows Fig. 9 shown.

[0078] According to the spectrum information of the CK staining image, the grayscaled and filled CK staining image is scaled. In this scenario, the six scales are used as an example for illustration. The results of the six scales are shown in Table 1.

[0079]

[0080]

[0081] Table 1

[0082] The optical flow deformation field is fitted to the CK staining image and the PDL1 staining image at six scales, and the optical flow deformation field is obtained. The CK staining image is registered to the PDL1 staining image using the dense optical flow method to obtain the registered image (using the dense optical flow method, the registered image is obtained according to the CK staining image and the optical flow deformation field). The registered image is shown in Fig.10 shown.

[0083] In order to demonstrate the effectiveness of the registration method in the embodiment of the present invention, the similarity and correlation between the original image and the target image before registration were tested, and the similarity and correlation between the image after registration and the target image were tested. The mean absolute error (MAE) between the data of the original image that was grayed and filled to the same size and the data of the target image was about 0.08582131, and the mean absolute error between the data of the image registered using the optical flow method and the data of the target image was about 0.0710295. It can be seen that the mean absolute error between the data of the original image and the data of the target image before registration using the optical flow method is greater than the mean absolute error between the data of the image registered using the optical flow method and the data of the target image, that the registration method in the embodiment of the present invention improves the similarity of the pathological section staining image. The Pearson correlation coefficient between the data of the original image and the data of the target image is measured to be about 0.73915153, while the Pearson correlation coefficient between the data of the image registered using the optical flow method and the data of the target image is about 0.79480496. It can be seen from the fact that the Pearson correlation coefficient between the data of the image registered using the optical flow method and the data of the target image is greater than the Pearson correlation coefficient between the data of the original image and the data of the target image that the registration method in the embodiment of the present invention improves the correlation of the pathological section staining image. The data are shown in Table 2.

[0084]

[0085]

[0086] Table 2

[0087] See also Fig.11 In order to execute the corresponding steps in the above embodiments and various possible methods, an implementation method of the registration device 140 is given below. The registration device 140 can be applied to the above Figure 1 The electronic device 100 shown. Further, Fig.11 This is a functional module diagram of a registration device 140 provided in an embodiment of the present invention. It should be noted that the basic principle and technical effect of the registration device 140 provided in this embodiment are the same as those of the above method embodiment. For the sake of brief description, for the parts not mentioned in this embodiment, reference can be made to the corresponding contents in the above method embodiment. The registration device 140 includes: an image acquisition module 141 and an image registration module 142.

[0088] The image acquisition module 141 is used to acquire the original image of the pathological section staining image and the target image of the pathological section staining image.

[0089] The image registration module 142 is used to register the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule by using an optical flow method.

[0090] In one implementation, the image registration module 142 is used to align the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules through the following steps: dividing the original image of the pathological section stained image into multiple scales in the frequency domain according to spectral information; optimizing the original image of the pathological section stained image and the target image of the pathological section stained image using a gradient descent method at each divided scale, fitting the optical flow deformation field respectively, and aligning the original image of the pathological section stained image and the target image of the pathological section stained image.

[0091] In one implementation, the image registration module 142 is used to register the original image of the pathological section staining image and the target image of the pathological section staining image pixel by pixel using a dense optical flow method according to the optical flow deformation field.

[0092] In one implementation, the registration device 140 also includes a pre-registration module 143, which is used to pre-register the original image of the pathological section stained image with reference to the target image of the pathological section stained image using affine transformation before using the optical flow method to align the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules.

[0093] In one implementation, the registration device 140 also includes a preprocessing module 144, which is used to grayscale the original image of the pathological section stained image and the target image of the pathological section stained image before aligning the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules using the optical flow method, and fill the original image of the pathological section stained image and the target image of the pathological section stained image after the grayscale processing to the same size.

[0094] In one implementation, the registration device 140 further includes an information output module 145 for outputting at least one of a registered image and an optical flow deformation field.

[0095] Optionally, the above units can be stored in the form of software or firmware. Figure 1 The memory 110 shown in the figure may be fixed in the operating system (OS) of the electronic device 100 and may be Figure 1 Meanwhile, data and program codes required for executing the above units may be stored in the memory 110.

[0096] The registration method, device, electronic device and computer-readable storage medium provided in the embodiments of the present invention cleverly apply the optical flow method to the registration of pathological section stained images, combine pre-registration, preprocessing and other processes, use dense optical flow to perform optical flow deformation field fitting at multiple sizes, and improve the accuracy of pathological section stained image registration.

[0097] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0098] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0099] If the functions are implemented in the form of software function modules 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 the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0100] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A registration method, characterized in that: include: Acquire an original image of a pathological section staining image and a target image of a pathological section staining image; Using the optical flow method, the original image of the pathological section staining image and the target image of the pathological section staining image are registered according to a set rule, including: Divide the original image of the pathological section staining image into multiple scales in the frequency domain according to the frequency spectrum information; For the original image of the pathological section stained image and the target image of the pathological section stained image, the gradient descent method is used to optimize at each divided scale, and the optical flow deformation field is fitted respectively; according to the optical flow deformation field, the dense optical flow method is used to align the original image of the pathological section stained image and the target image of the pathological section stained image pixel by pixel, and the bilinear interpolation method is used to assist the alignment.

2. The registration method according to claim 1, characterized in that: Before using the optical flow method to align the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules, the method also includes: using affine transformation to pre-align the original image of the pathological section stained image with reference to the target image of the pathological section stained image.

3. The registration method according to claim 2, characterized in that: The step of pre-registering the original image of the pathological section staining image with reference to the target image of the pathological section staining image by using affine transformation comprises: Estimate an affine transformation matrix based on pre-selected paired points; The original image of the pathological section staining image is pre-registered with reference to the target image of the pathological section staining image according to the affine transformation matrix.

4. The registration method according to claim 1, characterized in that: Before registering the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule using the optical flow method, the method further includes: grayscale the original image of the pathological section staining image and the target image of the pathological section staining image; The original image of the pathological section staining image after grayscale processing and the target image of the pathological section staining image are filled to the same size.

5. The registration method according to claim 1, characterized in that: The method further comprises: outputting the image after the registration; and / or Output the optical flow deformation field.

6. A registration device, characterized in that: include: An image acquisition module, used to acquire an original image of a pathological section staining image and a target image of a pathological section staining image; An image registration module is used to register the original image of the pathological section staining image and the target image of the pathological section staining image according to a set rule using an optical flow method, including: dividing the original image of the pathological section staining image into multiple scales in the frequency domain according to spectrum information; For the original image of the pathological section stained image and the target image of the pathological section stained image, the gradient descent method is used to optimize at each divided scale, and the optical flow deformation field is fitted respectively; according to the optical flow deformation field, the dense optical flow method is used to align the original image of the pathological section stained image and the target image of the pathological section stained image pixel by pixel, and the bilinear interpolation method is used to assist the alignment.

7. The registration device according to claim 6, characterized in that: The registration device also includes a pre-registration module, which is used to pre-register the original image of the pathological section stained image with reference to the target image of the pathological section stained image using affine transformation before using the optical flow method to align the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules.

8. The registration device according to claim 6, characterized in that: The registration device also includes a preprocessing module, which is used to grayscale the original image of the pathological section stained image and the target image of the pathological section stained image before aligning the original image of the pathological section stained image and the target image of the pathological section stained image according to set rules using the optical flow method, and fill the original image of the pathological section stained image and the target image of the pathological section stained image after the grayscale processing to the same size.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor can execute the machine executable instructions to implement the method according to any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images