Pathology scanning image clarity analysis method, system, electronic equipment and medium

Abstract

The present invention relates to the field of image processing technology, and specifically to a method, system, electronic device and medium for analyzing the clarity of a pathological scan image, including: splitting a pathological scan image into multiple local images, obtaining a first sub-image with the largest gradient of the local image; obtaining a low-frequency sub-band image of the first sub-image and several high-frequency sub-band images in different directions based on NSST; obtaining a score for each detail layer based on the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions, and obtaining a clarity score for the pathological scan image. The present invention obtains the first sub-image corresponding to each local image based on a gradient summation algorithm, and uses the score of the first sub-image as the score of the corresponding local image, thereby improving efficiency, enhancing the real-time performance of scoring each local image, and scoring based on NSST, so that the scoring is not affected by the image content, and improving the reliability of obtaining the score.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a method, system, electronic equipment and medium for analyzing the clarity of a pathological scanning image. Background Art

[0002] The development of modern medicine has greatly increased the demand for pathological slice scanning. During the scanning process, local scanning blur may occur due to internal and external factors, reducing the quality of the film. For large-scale scanning tasks, it is too cumbersome to manually find whether there is blur in each field of view. Therefore, an effective clarity evaluation method is needed to help scanners improve the efficiency of evaluating image quality, and automatically rescan the current slice when large areas of blur appear.

[0003] Most of the existing clarity evaluation algorithms are based on gradient calculations, such as Brenner gradient, Tenengrad gradient, Laplacian gradient, etc., or on grayscale variance. However, since they are all summation calculations, in the absence of a reference image, the evaluation score is easily affected by the image content. A blurred image with rich content may have a higher score than a clear image with less content, which is obviously inconsistent with the scoring goal. Summary of the invention

[0004] The purpose of the present invention is to overcome the problem of limitations of existing clarity evaluation algorithms and to provide a pathology scanning image clarity analysis method, system, electronic device and medium.

[0005] In order to achieve the above object, the present invention provides a method for analyzing the clarity of a pathological scanning image, comprising the following steps:

[0006] Acquire a pathological scan image, split the pathological scan image into w1×h1 pixels, and acquire a plurality of local images in which the effective pathological region area is larger than N×w1×h1 pixels;

[0007] Divide each local image into w2×h2 pixels to obtain multiple sub-images corresponding to each local image, calculate the multiple sub-images corresponding to the local image based on the gradient summation algorithm, and obtain the first sub-image with the largest gradient corresponding to each local image;

[0008] Perform the first-layer decomposition on each first sub-image to obtain a low-frequency sub-band image LG and p high-frequency sub-band images HG in different directions corresponding to each first sub-image. j (j=1,2,…p);

[0009] Calculate the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image, and calculate the clarity score of the first sub-image according to the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image. Use the clarity score of the first sub-image as the clarity score of the local image corresponding to the first sub-image, and obtain the clarity score of the local image corresponding to each first sub-image in turn to obtain the clarity score of the pathology scan image.

[0010] As a feasible implementation, the value of W1 is 2048, the value of h1 is 1024, the value of w2 is 256, and the value of h2 is 256.

[0011] As an implementable method, the format of the pathological scan image is a KFB format.

[0012] As an implementable method, the step of calculating the clarity score of the first sub-image according to the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image includes:

[0013] The average of several coefficient absolute value sums is calculated based on the coefficient absolute value sums of the high-frequency sub-band images in several directions of the first sub-image, and the clarity score of the first sub-image is calculated based on the average and the coefficient absolute value sum of the low-frequency sub-band image.

[0014] As an implementable method, the step of calculating the clarity score of the first sub-image according to the mean value and the absolute value of the coefficient of the low-frequency sub-band image includes:

[0015] The clarity score of the first sub-image is obtained by calculating the ratio of the mean value to the absolute value of the coefficient of the low-frequency sub-band image.

[0016] As an implementable method, the calculation formula for calculating the average of the sum of the absolute values ​​of the coefficients of the high-frequency sub-band images in several directions of the first sub-image is:

[0017] Wherein, p represents the number of high-frequency sub-band images obtained after the first-layer decomposition in the first sub-image, HGM represents the mean of the absolute value and coefficient of the high-frequency sub-band images in p directions, represents the absolute value sum of coefficients of the j-th high-frequency sub-band image obtained after performing the first-layer decomposition on the first sub-image, and w2 and h2 represent the length pixel and width pixel of the first sub-image respectively.

[0018] As an implementation method, the calculation formula for calculating the ratio of the mean value to the absolute value of the coefficient of the low-frequency sub-band image to obtain the clarity score of the first sub-image is: Wherein, w2 and h2 represent the length and width of the first sub-image in pixels, respectively. represents the absolute value sum of the coefficients of the low-frequency sub-band image obtained after the first-level decomposition of the first sub-image, and HGM(j,k) represents the mean value of the absolute value sum of the coefficients of the high-frequency sub-band images in several directions obtained after the first-level decomposition of the first sub-image.

[0019] Accordingly, the present invention also provides a pathology scanning image clarity analysis system, comprising the following modules:

[0020] A first acquisition module acquires a pathological scan image, splits the pathological scan image into w1×h1 pixels, and acquires a plurality of local images in which the effective pathological region area is larger than N×w1×h1 pixels;

[0021] A division module divides each local image into w2×h2 pixels to obtain multiple sub-images corresponding to each local image, and calculates the multiple sub-images corresponding to the local image based on a gradient summation algorithm to obtain a first sub-image with the largest gradient corresponding to each local image;

[0022] The second acquisition module performs the first-layer decomposition on each first sub-image to obtain a low-frequency sub-band image LG and p high-frequency sub-band images HG in different directions corresponding to each first sub-image. j (j=1,2,…p);

[0023] A scoring module calculates the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image, and obtains a clarity score of the first sub-image based on the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image, and uses the clarity score of the first sub-image as the clarity score of the local image corresponding to the first sub-image, and obtains the clarity score of each local image corresponding to the first sub-image in turn, thereby obtaining a clarity score of the pathology scan image.

[0024] Correspondingly, the present invention also provides an electronic device, comprising: at least one processor, and a memory communicatively connected to at least one of the processors; at least one of the processors is used to read the program in the memory to execute the above method.

[0025] Correspondingly, the present invention also provides an electronic device and a computer-readable storage medium, characterized in that instructions are stored on the medium, and when the instructions are executed on a computer, the computer executes the above method.

[0026] Beneficial effects of the present invention: The present invention discloses a method, system, electronic device and medium for analyzing the clarity of pathological scanning images. The present invention obtains a first sub-image corresponding to each local image based on a gradient summation algorithm, performs a first-layer decomposition on the first sub-image to obtain a corresponding low-frequency sub-band image and p high-frequency sub-band images in different directions, obtains a score for the first sub-image according to the absolute value of the coefficient corresponding to the low-frequency sub-band image and the absolute value of the coefficient of the p high-frequency sub-band images in different directions, and uses the score of the first sub-image as the score of the corresponding local image, thereby improving efficiency and enhancing the real-time performance of scoring each local image. Scoring is performed based on NSST, so that the scoring is not affected by the image content, thereby improving the reliability of obtaining the score. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Schematic diagram of the steps of a method for analyzing the clarity of a pathological scan image according to an embodiment of the present invention.

[0028] Figure 2 This is a schematic diagram of clarity scoring curves of different local images obtained by dividing the pathology scan image obtained by scanning the first type of slices according to -4 layers, 0 layers, and 5 layers respectively according to the pathology scan image clarity analysis method of an embodiment of the present invention.

[0029] Figure 3 This is a schematic diagram of clarity scoring curves of different local images obtained by dividing the pathology scan image obtained by scanning the second type of slices according to -4 layers, 0 layers, and 5 layers respectively according to the pathology scan image clarity analysis method of an embodiment of the present invention.

[0030] Figure 4 This is a schematic diagram of clarity scoring curves of different local images obtained by dividing the pathology scan image obtained by scanning the third type of slices according to -4 layers, 0 layers, and 5 layers respectively according to the pathology scan image clarity analysis method of an embodiment of the present invention.

[0031] Figure 5 This is a schematic diagram of a local image with little content in an effective pathological area obtained by the pathological scanning image clarity analysis method according to an embodiment of the present invention.

[0032] Figure 6 This is a schematic diagram of a local image with a lot of content in an effective pathological area obtained by the pathological scanning image clarity analysis method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] See also Figure 1 This embodiment provides a technical solution: a method for analyzing the clarity of a pathological scanning image, comprising the following steps:

[0035] Step S100: Acquire a pathological scan image, split the pathological scan image into w1×h1 pixels, and acquire a plurality of local images in which the effective pathological region area is larger than N×w1×h1 pixels;

[0036] Step S200: Divide each local image into w2×h2 pixels to obtain multiple sub-images corresponding to each local image, and calculate the multiple sub-images corresponding to the local image based on the gradient summation algorithm to obtain the first sub-image with the largest gradient corresponding to each local image;

[0037] Step S300: Perform the first-level decomposition on each first sub-image to obtain a low-frequency sub-band image LG and p high-frequency sub-band images HG in different directions corresponding to each first sub-image. j (j=1,2,…p);

[0038] Step S400: Calculate the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image, calculate the clarity score of the first sub-image according to the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images of the detail layers in several directions in the first sub-image, use the clarity score of the first sub-image as the clarity score of the local image corresponding to the first sub-image, obtain the clarity score of the local image corresponding to each first sub-image in turn, and obtain the clarity score of the pathology scan image.

[0039] Execute step S100. In this embodiment, the value of W1 is set to 2048, the value of h1 is set to 1024, the value of w2 is set to 256, and the value of h2 is set to 256. That is to say, in this embodiment, a local image of 2048×1024 pixels and an effective pathological area larger than N×2048×1024 pixels is obtained. The value of N is determined according to the actual situation. In this embodiment, the value of N can be determined to be 0.2.

[0040] In this embodiment, the format of the pathological scan image is the KFB format.

[0041] Execute step S200, the gradient summation algorithm may be a Laplacian gradient summation algorithm, or may be other algorithms that can perform gradient summation calculations; in this embodiment, the gradient of each sub-image divided from the local image is calculated by the Laplacian gradient summation algorithm to obtain a first sub-image with the largest gradient, the scores of each layer of the first sub-image are calculated, and the scores of each layer of the first sub-image are used as the scores of each layer of the local image corresponding to the first sub-image, and the scores of each layer of the local image divided from the pathological scan image are used as the clarity scores of the pathological scan image; in this embodiment, the scores of each layer of the first sub-image are used as the scores of each layer of the local image corresponding to the first sub-image, thereby improving the efficiency of the clarity score calculation and improving the real-time performance.

[0042] Execute step S300, the step of processing each first sub-image to obtain a low-frequency sub-band image corresponding to each first sub-image and several high-frequency sub-band images in different directions in different detail layers is implemented based on NSST technology; this embodiment adopts NSST (non-subsampled shearlet transform), NSST can perform multi-scale and multi-directional transformation on the image to obtain a series of low- and high-frequency sub-band images. For example, if the image is decomposed in eight directions of the first layer, one low-frequency sub-band image and high-frequency sub-band images in eight directions of each layer can be obtained, that is, one direction corresponds to one high-frequency sub-band image, and there are a total of 8 high-frequency sub-band images, that is, the value of p is 8, wherein NSST technology is a prior art, and therefore, this embodiment will not be described in detail.

[0043] Executing step S400: the step of calculating the clarity score of the first sub-image according to the coefficient absolute value sum of the low-frequency sub-band image and the coefficient absolute value sum of the high-frequency sub-band images in several directions of the first sub-image comprises:

[0044] The average of several coefficient absolute value sums is calculated based on the coefficient absolute value sums of the high-frequency sub-band images in several directions of the first sub-image, and the clarity score of the first sub-image is calculated based on the average and the coefficient absolute value sum of the low-frequency sub-band images.

[0045] The step of calculating the clarity score of the first sub-image according to the mean value and the absolute value of the coefficient of the low-frequency sub-band image comprises:

[0046] The clarity score of the first sub-image is obtained by calculating the ratio of the sum of the mean values ​​to the sum of the absolute values ​​of the coefficients of the low-frequency sub-band image. The calculation formula for calculating the mean of the sum of the absolute values ​​of the coefficients of the high-frequency sub-band image in several directions of the first sub-image is:

[0047] Wherein, p represents the number of high-frequency sub-band images obtained after the first-layer decomposition in the first sub-image, HGM represents the mean of the absolute value and coefficient of the high-frequency sub-band images in p directions, represents the absolute value sum of coefficients of the high-frequency subband image of the jth detail layer obtained after performing the first layer decomposition on the first sub-image, and w2 and h2 represent the length pixel and width pixel of the first sub-image respectively.

[0048] Specifically, by performing a first-layer decomposition on the first sub-image, high-frequency sub-band images in 8 directions and one low-frequency sub-band image are obtained, that is, the value of p in the above description is 8, and the absolute value sum of the coefficients of the high-frequency sub-band images in 8 directions are calculated respectively, and then the mean of the absolute value sum of the coefficients of the high-frequency sub-band images in 8 directions is calculated, and the ratio of the mean to the absolute value sum of the coefficients of the low-frequency sub-band image is calculated as the clarity score of the corresponding first sub-image.

[0049] This embodiment adopts a one-layer decomposition, that is, only one-layer NSST decomposition is performed on the first subgraph, which improves the real-time performance compared with the three-layer decomposition in the prior art.

[0050] The calculation formula for calculating the clarity score of the first sub-image by calculating the ratio of the mean value to the absolute value of the coefficient of the low-frequency sub-band image is: Wherein, w2 and h2 represent the length and width of the first sub-image in pixels, respectively. represents the absolute value sum of the coefficients of the low-frequency sub-band image obtained after the first-level decomposition of the first sub-image, and HGM(j,k) represents the mean value of the absolute value sum of the coefficients of the high-frequency sub-band images in several directions obtained after the first-level decomposition of the first sub-image.

[0051] Specifically, after the pathological scan image is split into multiple local images, a field of view sequence number is set for each local image, and after the clarity score of each layer corresponding to each local image is obtained, the clarity score of the pathological scan image is combined to form a combination; Figure 5 The first sub-image of one of the local images is shown in FIG. 1 . The gradient sum obtained based on the gradient summation algorithm is 88124.75, and the clarity score obtained by scoring is 0.1596. Figure 6 Shown is the first sub-image of another local image. The gradient sum obtained based on the gradient summation algorithm is 288580.03, and the clarity score obtained by scoring is 0.1096. It can be seen that the contents of the two images are different, and the clarity scores obtained are also different, and the scoring results will not be affected by the size of the effective pathological area.

[0052] In this embodiment, the objective lens can be adjusted to scan different layers of the same slice -4 to 5, respectively, to obtain pathological scan images corresponding to the same slice but different layers, so as to effectively score the obtained pathological scan images. The scores for different types of slices are consistent and are not easily affected by the image content. Moreover, under the current parameter settings, a certain threshold can distinguish the image clarity of different layer scans (near layer 0 is a clearly focused image, -4 to 5 layers are scan images deviated from the corresponding layer number, and the layer height is 2mm); for example, Figure 2 It is a schematic diagram of the clarity score curve of different local images obtained by dividing the pathological scan image obtained by scanning the first type of slices according to -4 layers, 0 layers, and 5 layers, Figure 3 Schematic diagram of the clarity score curve of different local images obtained by dividing the pathological scan image obtained by scanning the second type of slices according to -4 layers, 0 layers, and 5 layers, Figure 4 The figure is a schematic diagram of the clarity score curve of different local images obtained by dividing the pathological scan image obtained by scanning the third type of slices according to -4 layers, 0 layers, and 5 layers respectively; wherein different field of view numbers correspond to different local images corresponding to the same pathological scan image. It can be seen that this embodiment improves the execution efficiency of the algorithm and ensures the accuracy of the results by performing NSST score calculation on the first sub-image of the local field of view and adopts one-layer decomposition and eight directions.

[0053] Based on the same inventive concept, an embodiment of the present invention further provides a pathological scanning image clarity analysis system. Regarding the components, please refer to the above related descriptions, and the repeated parts will not be repeated. The system includes the following steps:

[0054] A first acquisition module acquires a pathological scan image, splits the pathological scan image into w1×h1 pixels, and acquires a plurality of local images in which the effective pathological region area is larger than N×w1×h1 pixels;

[0055] A division module divides each local image into w2×h2 pixels to obtain multiple sub-images corresponding to each local image, and calculates the multiple sub-images corresponding to the local image based on a gradient summation algorithm to obtain a first sub-image with the largest gradient corresponding to each local image;

[0056] The second acquisition module performs the first layer decomposition on each first sub-image to obtain a low-frequency sub-band image LG and p high-frequency sub-band images HG of detail layers in different directions corresponding to each first sub-image. j (j=1,2,…p);

[0057] A scoring module calculates the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images of several directional detail layers in the first sub-image, and obtains a clarity score of the first sub-image based on the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images of several directional detail layers in the first sub-image. The clarity score of the first sub-image is used as the clarity score of the local image corresponding to the first sub-image, and the clarity score of the local image corresponding to each first sub-image is obtained in turn to obtain the clarity score of the pathological scan image.

[0058] Based on the same inventive concept, an embodiment of the present invention also provides an electronic device. Regarding the components, please refer to the above-mentioned related descriptions, and the repeated parts will not be repeated here. It includes: at least one processor, and a memory communicatively connected to the at least one processor; the at least one processor is used to read the program in the memory to execute the method described above.

[0059] Based on the same inventive concept, an embodiment of the present invention further provides a computer-readable storage medium. Regarding the components, please refer to the above-mentioned related descriptions, and the repeated parts will not be repeated. Instructions are stored on the medium. When the instructions are executed on the computer, the computer executes the method described above.

[0060] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for analyzing the clarity of a pathological scanning image, characterized in that: The following steps are involved: Acquire a pathological scan image, split the pathological scan image into w1×h1 pixels, and acquire a plurality of local images in which the effective pathological region area is larger than N×w1×h1 pixels; Divide each local image into w2×h2 pixels to obtain multiple sub-images corresponding to each local image, calculate the multiple sub-images corresponding to the local image based on the gradient summation algorithm, and obtain the first sub-image with the largest gradient corresponding to each local image, wherein w2 and h2 represent the length and width of the first sub-image in pixels, respectively; Perform the first-layer decomposition on each first sub-image to obtain a low-frequency sub-band image LG and p high-frequency sub-band images HG in different directions corresponding to each first sub-image. j (j=1,2,...,p); Calculate the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image, and calculate the clarity score of the first sub-image according to the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image. Use the clarity score of the first sub-image as the clarity score of the local image corresponding to the first sub-image, and obtain the clarity score of the local image corresponding to each first sub-image in turn to obtain the clarity score of the pathology scan image.

2. The method for analyzing the clarity of pathological scanning images according to claim 1, characterized in that: The value of w1 is 2048, the value of h1 is 1024, the value of w2 is 256, and the value of h2 is 256.

3. The method for analyzing the clarity of pathological scanning images according to claim 1, characterized in that: The format of the pathological scan image is KFB format.

4. The method for analyzing the clarity of pathological scanning images according to claim 1, characterized in that: The step of calculating the clarity score of the first sub-image according to the coefficient absolute value sum of the low-frequency sub-band image and the coefficient absolute value sum of the high-frequency sub-band images in several directions in the first sub-image comprises: The average of several coefficient absolute value sums is calculated based on the coefficient absolute value sums of the high-frequency sub-band images in several directions of the first sub-image, and the clarity score of the first sub-image is calculated based on the average and the coefficient absolute value sum of the low-frequency sub-band image.

5. The method for analyzing the clarity of pathological scanning images according to claim 4, characterized in that: The step of calculating the clarity score of the first sub-image according to the mean value and the absolute value of the coefficient of the low-frequency sub-band image comprises: The clarity score of the first sub-image is obtained by calculating the ratio of the mean value to the absolute value of the coefficient of the low-frequency sub-band image.

6. The method for analyzing the clarity of pathological scanning images according to claim 4, characterized in that: The calculation formula for calculating the mean of the sum of the absolute values ​​of the coefficients of the high-frequency sub-band images in several directions of the first sub-image is: Wherein, p represents the number of high-frequency sub-band images obtained after the first-layer decomposition in the first sub-image, HGM represents the mean of the absolute value and coefficient of the high-frequency sub-band images in p directions, represents the absolute value sum of coefficients of the j-th high-frequency sub-band image obtained after performing the first-layer decomposition on the first sub-image, and w2 and h2 represent the length pixel and width pixel of the first sub-image respectively.

7. The method for analyzing the clarity of pathological scanning images according to claim 5, characterized in that: The calculation formula for calculating the clarity score of the first sub-image by calculating the ratio of the mean value to the absolute value of the coefficient of the low-frequency sub-band image is: Wherein, w2 and h2 represent the length and width of the first sub-image in pixels, respectively. represents the absolute value sum of the coefficients of the low-frequency sub-band image obtained after the first-level decomposition of the first sub-image, and HGM(j,k) represents the mean value of the absolute value sum of the coefficients of the high-frequency sub-band images in several directions obtained after the first-level decomposition of the first sub-image.

8. A pathology scanning image clarity analysis system, characterized in that: Includes the following modules: A first acquisition module acquires a pathological scan image, splits the pathological scan image into w1×h1 pixels, and acquires a plurality of local images in which the effective pathological region area is larger than N×w1×h1 pixels; A division module divides each local image into w2×h2 pixels to obtain multiple sub-images corresponding to each local image, and calculates the multiple sub-images corresponding to the local image based on a gradient summation algorithm to obtain a first sub-image with the largest gradient corresponding to each local image, wherein w2 and h2 represent the length and width of the first sub-image in pixels, respectively; The second acquisition module performs the first-layer decomposition on each first sub-image to obtain a low-frequency sub-band image LG and p high-frequency sub-band images HG in different directions corresponding to each first sub-image. j (j=1,2,...,p); A scoring module calculates the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image, and obtains a clarity score of the first sub-image based on the absolute value sum of the coefficients of the low-frequency sub-band image and the absolute value sum of the coefficients of the high-frequency sub-band images in several directions of the first sub-image, and uses the clarity score of the first sub-image as the clarity score of the local image corresponding to the first sub-image, and obtains the clarity score of each local image corresponding to the first sub-image in turn, thereby obtaining a clarity score of the pathology scan image.

9. An electronic device, characterized in that: include: at least one processor, and a memory communicatively coupled to at least one of the processors; At least one of the processors is used to read the program in the memory to execute the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The medium stores instructions, and when the instructions are executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 7.

Images