Digital pathology slide color calibration method and apparatus

Abstract

The application provides a digital pathological section color calibration method and device, including the following steps: obtaining a standard color image and a color image to be calibrated, and extracting feature points of the standard color image and the color image to be calibrated; matching the standard color image and the color image to be calibrated based on the feature points to obtain a standard color matching image and a color matching image to be calibrated; taking the standard color matching image as a reference to histogram color calibration of the color matching image to be calibrated to obtain a first color calibration image; taking the standard color matching image as a reference to CCM color calibration of the first color calibration image to obtain a second color calibration image; and taking the standard color matching image as a reference to gamma correction of the color of the second color calibration image to obtain a color calibration completed image. The scheme fully utilizes all pixel information of a target image, quantizes a color calibration process to exclude artificial errors, and can significantly improve the efficiency and accuracy of digital pathological image color calibration.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of color correction, in particular to a digital pathological section color calibration method and device. BACKGROUND

[0002] A digital section scanner is a precision instrument integrating multiple disciplines such as optics, mechanics, electronics and computers. It controls the microscopic imaging system and the section movement, collects continuous high-resolution microscopic images and seamlessly splices them to generate a high-resolution whole section image (WSI). It has been widely used in the fields of pathological diagnosis, teaching and training, drug research and scientific research. With the expansion of applications, the scanner technology has been continuously improved, and various scanning technologies have emerged, which have put forward higher requirements for the color accuracy of the section images.

[0003] Tissue staining is a standard method for examining tissue cell structure in clinical pathology and life science research, but the traditional method requires complex sample preparation, professional laboratory facilities and well-trained tissue technicians, which makes it difficult to implement in resource-limited environments. There are now a variety of new technologies for digital pathological section color generation, among which deep learning technology digitally generates tissue staining images by training neural networks, providing a fast, cost-effective and accurate alternative to standard chemical staining methods. These methods have put higher requirements on the standardization of color, and the importance of color standardization has gradually increased.

[0004] Because different laboratories, equipment or staining methods can cause color differences in pathological sections, which in turn affect the diagnosis results, color standardization can reduce such differences and improve the accuracy and consistency of diagnosis, while facilitating image analysis, so it is urgent to quantitatively analyze the color differences of digital pathological images to complete color correction. SUMMARY

[0005] The embodiments of the present application provide a digital pathological section color calibration method and device, which quantifies the color correction process to eliminate human errors, can significantly improve the efficiency and accuracy of digital pathological image color correction, automatically provides color correction parameters and reduces color difference, and provides a standardized reference for the qualification of digital pathological images.

[0006] In a first aspect, the embodiments of the present application provide a digital pathological section color calibration method, which comprises:

[0007] Obtaining a standard color image and a to-be-corrected color image in the same field of view as the standard color image, and extracting feature points of the standard color image and feature points of the to-be-corrected color image;

[0008] The standard color image and the color-to-be-corrected image are matched based on the feature points of the standard color image and the feature points of the color-to-be-corrected image to obtain a standard color matching image and a color-to-be-corrected matching image, and the image contents of the standard color matching image and the color-to-be-corrected matching image are the same.

[0009] The color of the second color-corrected image is gamma-corrected based on the standard color matching image to obtain a color-corrected completed image.

[0010] In a second aspect, an embodiment of the present application provides a digital pathological section color calibration device, including:

[0011] The standard color image and the color-to-be-corrected image are matched based on the feature points of the standard color image and the feature points of the color-to-be-corrected image to obtain a standard color matching image and a color-to-be-corrected matching image, and the image contents of the standard color matching image and the color-to-be-corrected matching image are the same.

[0012] The color of the second color-corrected image is gamma-corrected based on the standard color matching image to obtain a color-corrected completed image.

[0013] The color of the second color-corrected image is gamma-corrected based on the standard color matching image to obtain a color-corrected completed image.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, including a memory and a processor, the memory stores a computer program, and the processor is configured to run the computer program to execute a digital pathological section color calibration method.

[0015] In a fourth aspect, an embodiment of the present application provides a readable storage medium, the readable storage medium stores a computer program, the computer program includes program codes for controlling a process to execute a process, and the process includes a digital pathological section color calibration method.

[0016] The main contributions and innovations of the present application are as follows:

[0017] The embodiment of the application can accurately capture key features under different scales through multi-scale space construction and Hessian matrix calculation when extracting feature points, fine details can be obtained under small scales, and macro features can be highlighted under large scales, laying an accurate foundation for image matching; feature point matching combined with geometric transformation parameter cutting ensures that the matching images obtained are the same in content, and guarantees the consistency of subsequent color correction objects; histogram color correction can quickly reduce the color difference between the image to be corrected and the standard image through RGB channel separation and cumulative distribution function mapping, and lay a good foundation for further color correction; CCM color correction optimizes the matrix with the minimum sum of color differences as the target, and constrains the sum of each row to be 1, which ensures the white balance unchanged while accurately adjusting the color; Gamma correction further refines the color correction by dynamic parameter optimization with the minimum color difference as the target, which finally significantly improves the efficiency and accuracy of digital pathology image color correction, and eliminates artificial errors to provide a standardized reference for image eligibility.

[0018] The details of one or more embodiments of the application are presented in the following drawings and description to make other features, objects and advantages of the application more apparent. BRIEF DESCRIPTION OF DRAWINGS

[0019] The drawings described herein are intended to provide further understanding of the application, and form a part of the application. The illustrative embodiments of the application and their description serve to explain the application without unduly limiting it. In the drawings:

[0020] Figure 1 is a flow chart of a digital pathology slice color calibration method according to an embodiment of the application;

[0021] Figure 2 is a schematic diagram of a standard color image and an image to be corrected according to an embodiment of the application;

[0022] Figure 3 is a comparison schematic diagram of a standard color matching image and a color correction completed image according to an embodiment of the application;

[0023] Figure 4 is a structural block diagram of a digital pathology slice color calibration device according to an embodiment of the application;

[0024] Figure 5 is a hardware structure schematic diagram of an electronic device according to an embodiment of the application. DETAILED DESCRIPTION

[0025] The exemplary embodiments will be described in detail herein with reference to the attached drawings. The description of the exemplary embodiments is intended to apply to all alternative embodiments, as would be understood by one skilled in the art. In other words, descriptions of some embodiments in the specification do not necessarily apply to all embodiments. Similarly, the description "comprises," "comprising," "containing" or "containing" throughout the disclosure can be understood in the context of this application as specifying the presence of stated steps or features, but not precluding the presence of one or more other steps or features, as would be understood by one of ordinary skill in the art.

[0026] It should be noted that the steps of the methods described in the other embodiments are not necessarily performed in the order described in this specification. In some other embodiments, the steps included in the methods can be more or less than described in this specification. In addition, a single step described in this specification can be broken down into multiple steps in other embodiments; and multiple steps described in this specification can be combined into a single step in other embodiments.

[0027] Embodiment one

[0028] The embodiments of the present application provide a digital pathology slice color calibration method, which fully utilizes all pixel information of the target image, quantizes the color calibration process to eliminate human error, can significantly improve the efficiency and accuracy of digital pathology image color correction, automatically provides color calibration parameters and reduces color difference, provides a standardized reference for digital pathology image qualification, and specifically Figure 1 , the method comprises:

[0029] Obtaining a standard color image and a color image to be calibrated in the same field of view as the standard color image, and extracting feature points of the standard color image and feature points of the color image to be calibrated;

[0030] Matching the standard color image and the color image to be calibrated based on the feature points of the standard color image and the feature points of the color image to be calibrated to obtain a standard color matching image and a color image to be calibrated, and the image content of the standard color matching image and the color image to be calibrated is the same;

[0031] Taking the standard color matching image as a reference, the color image to be calibrated is histogram calibrated to obtain a first color calibrated image, and the first color calibrated image is CCM calibrated based on the standard color matching image to obtain a second color calibrated image, and the color of the second color calibrated image is gamma corrected based on the standard color matching image to obtain a color calibrated image.

[0032] In some specific embodiments, the standard color image and the color image to be calibrated in the present scheme are both pathology slice images obtained using digital scanning slices, and a schematic diagram of the standard color image and the color image to be calibrated is as shown in Figure 2As shown, the standard color image is a pathological section image certified by experts in the field, and the standard color image can clearly see the details in the pathological section.

[0033] In some embodiments, the standard color image is subjected to multi-scale space construction to obtain a standard multi-scale space, and the Hessian matrix of each pixel point at each scale level in the standard multi-scale space is calculated to obtain a Hessian eigenvalue. If the Hessian eigenvalue of the current pixel point is the maximum value in the corresponding local region, the current pixel point is taken as a standard color image feature point. The corresponding local region includes a first region and a second region. The first region is a region including the current pixel point and adjacent pixel points centered on the current pixel point. The second region is a region corresponding to the first region at the adjacent scale level of the current scale level.

[0034] Similarly, the to-be-corrected color image is subjected to multi-scale space construction to obtain a to-be-corrected multi-scale space, and the Hessian matrix of each pixel point at each scale level in the to-be-corrected multi-scale space is calculated to obtain a Hessian eigenvalue. If the Hessian eigenvalue of the current pixel point is the maximum value in the corresponding local region, the current pixel point is taken as a to-be-corrected color image feature point. The corresponding local region includes a first region and a second region. The first region is a region including the current pixel point and adjacent pixel points centered on the current pixel point. The second region is a region corresponding to the first region at the adjacent scale level of the current scale level.

[0035] Specifically, the present scheme uses filters of different sizes to construct the standard multi-scale space and the to-be-corrected multi-scale space.

[0036] Specifically, the Hessian matrix of each pixel point is constructed, and the size of the Hessian matrix determinant is taken as the Hessian eigenvalue of the pixel point. The present scheme takes the extreme value point of the Hessian eigenvalue in the local region as the feature point, which can reflect the curvature change of the local region of the image. The extreme value points at different scales correspond to key features of different sizes. For example, in a pathological image, the extreme value points at a small scale may capture fine features such as texture details and small edge intersections; and the extreme value points at a large scale tend to highlight the object contour and the shape change of a relatively macro region.

[0037] Further, before constructing the multi-scale space of the standard color image and the to-be-corrected color image, the standard color image and the to-be-corrected color image are subjected to grayscale processing, and the features of the standard color image and the to-be-corrected color image are represented by integral images.

[0038] Specifically, the integral image is a feature representation of digital image processing, which is used to quickly calculate the sum of gray values of local area of the image. The present scheme uses the integral image to represent the features of the standard color image and the color correction image. In the subsequent image processing, the sum of gray values of any area can be obtained by simple addition and subtraction operation, which facilitates the subsequent multi-scale space construction, feature point matching and other operations.

[0039] In some embodiments, the standard color image feature points and the color correction image feature points with a vector distance less than a set threshold are matched feature points, the geometric transformation parameters of the standard color image and the color correction image are obtained, and the standard color matching image and the color correction matching image are obtained by cutting the standard color image and the color correction image based on the geometric transformation parameters and the matched feature points.

[0040] Specifically, the present scheme obtains the feature description vector of each feature point by calculating the Haar wavelet response of the area around each feature point in the standard color image and the color correction image, and then calculates the vector distance between each feature description vector in the standard color image and each feature description vector in the color correction image. In addition, the feature description vector includes the position, scale and direction information of the feature point.

[0041] Specifically, the present scheme uses a feature descriptor matrix to represent the matched feature points, and each row in the feature descriptor matrix includes two indexes, which are two matched feature points. The present scheme uses cosine similarity or Euclidean distance to match the feature points.

[0042] Further, the present scheme uses the MSAC algorithm to obtain the geometric transformation parameters of the standard color image and the color correction image. Specifically, taking the standard color image as an example, a data set W is obtained by randomly averaging sampling in the standard color image, a minimum data set Q is randomly selected from the data set W without repetition for plane model fitting, a plane model S is obtained by fitting the minimum data set Q using the least squares method, the distances of all points in the data set W to the plane model S are calculated, the points within the allowable error threshold range are marked as inliers, and the rest of the points are outliers. The cost Ci of the plane model is calculated, the size of the cost Ci of the current model and the cost Cb of the previous best model is compared, the smaller one is recorded as the cost of the new best model, and the corresponding inliers and model parameters are recorded. Repeat the random average sampling in the standard color image and the calculation of the model parameters of the inliers until the iteration is completed. The best plane model parameters are obtained by fitting the plane model using all inlier data and the least squares method. Finally, the geometric transformation parameters are obtained using the best plane model parameters of the standard color image and the best plane model parameters of the color correction image.

[0043] Specifically, the cost function expression of the MSAC algorithm is as follows:

[0044]

[0045] wherein, when the distance e of the point p to the model is less than the threshold value T, the point is determined as an inlier and the weight is e; otherwise, as an outlier and the weight is T.

[0046] In some embodiments, the coinciding area of the standard color image and the image to be corrected is cut out as a standard color matching image and an image to be corrected, and the average color difference between the standard color matching image and the image to be corrected is 15.10419, which is much larger than the qualified value 5, so the image to be corrected needs to be corrected.

[0047] Specifically, CIEDE2000 is a formula for evaluating color difference, which is an improved color difference evaluation method proposed by the International Commission on Illumination (CIE) in 2000.

[0048] In some embodiments, in the histogram correction step, the standard color matching image is separated by RGB channels, and the histogram data distribution of each channel is calculated to obtain the standard image cumulative distribution function of each channel. The image to be corrected is separated by RGB channels, and the histogram data distribution of each channel is calculated to obtain the corrected image cumulative distribution function of each channel. Any pixel value of a pixel in the standard color matching image is taken as a first pixel value, and the cumulative probability of the first pixel value is found in the image to be corrected as a second pixel value according to the cumulative distribution function of the corresponding channel, and the first pixel value is mapped to the second pixel value. When the color difference between the image to be corrected and the standard matching image is less than a set threshold value, the histogram correction is ended to obtain a first corrected image.

[0049] Specifically, the cumulative distribution function in the present scheme represents the probability accumulation value of each pixel value, which reflects the cumulative pixel distribution of the image.

[0050] In some embodiments, in the CCM correction step, a CCM matrix is constructed, and the RGB channel matrix of the standard color matching image and the first corrected image is used as a data set. The sum of squares of color differences between the standard color matching image and the first corrected image is taken as an objective function to optimize the CCM matrix. After optimization, a second corrected image is obtained, wherein the sum of the values of each row is taken as a constraint of the CCM matrix.

[0051] Specifically, in the CCM correction step, the present scheme constructs a CCM matrix, and adjusts the RGB channel matrix of the first corrected image to minimize the sum of squares of color differences between the standard color matching image and the first corrected image to complete the CCM correction.

[0052] Specifically, the purpose of using the sum of the values ​​of each row to be 1 as the constraint of the CCM matrix is ​​to ensure that the white balance of the optimized image remains unchanged.

[0053] In some embodiments, in the gamma correction step, dynamic gamma correction parameters are set, and the second color-calibrated image is gamma-corrected using the dynamic gamma correction parameters to obtain a gamma-calibrated image. The color difference between the standard color matching image and the gamma-calibrated image is used as a cost function. When the cost function is minimized, the gamma calibration is completed, and the current gamma-calibrated image is used as the calibration completion image.

[0054] Specifically, the initial value of the dynamic gamma correction parameter is set to 0.85, the maximum value is 1.15, the optimization interval is 0.05, and during the gamma correction process, the dynamic gamma correction parameter is updated at an interval of 0.01.

[0055] Specifically, the comparison between the standard color matching image and the color calibration completed image is as follows: Figure 3 As shown by Figure 3 It can be clearly seen that there is basically no color difference between the color-calibrated image and the standard color-matching image. According to CIEDE2000, the color difference between the color-calibrated image and the standard color-matching image is calculated, and the color difference between the two is only 3.36373.

[0056] Example 2

[0057] Based on the same concept, refer to Figure 4 , this application also proposes a digital pathology slide color calibration device, comprising:

[0058] An acquisition module is used to acquire a standard color image and an image to be calibrated that is within the same field of view as the standard color image, and to extract feature points of the standard color image and the image to be calibrated;

[0059] a feature point matching module, which matches the standard color image with the image to be calibrated based on the feature points of the standard color image and the feature points of the image to be calibrated to obtain a standard color matching image and an image to be calibrated, wherein the image content of the standard color matching image and the image to be calibrated are the same;

[0060] The color correction module is used to perform histogram correction on the image to be color-corrected based on a standard color matching image to obtain a first color-corrected image, perform CCM color correction on the first color-corrected image based on the standard color matching image to obtain a second color-corrected image, and perform gamma correction on the color of the second color-corrected image based on the standard color matching image to obtain a color-corrected image.

[0061] Example 3

[0062] This embodiment also provides an electronic device, referring to Figure 4The computer program product, comprising a memory 404 having stored therein a computer program and a processor 402 arranged to execute the computer program to perform the steps of any of the above method embodiments.

[0063] In particular, the processor 402 can include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or can be configured as one or more integrated circuits to implement the embodiments of the present application.

[0064] The memory 404 can include a mass storage that stores data or instructions. For example, and without limitation, the memory 404 can include a Hard Disk Drive (HDD), a floppy disk drive, a Solid State Drive (SSD), a flash drive, a Compact Disc Read Only Memory (CD-ROM), a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. The memory 404 can be removable and / or non-removable (or fixed) as appropriate. The memory 404 can be internal or external as appropriate. In particular embodiments, the memory 404 is a Non-Volatile memory. In particular embodiments, the memory 404 includes a Read-Only Memory (ROM) and a Random Access Memory (RAM). The ROM can be a mask-programmed ROM, a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically EPROM (EEPROM), an Electrically Alterable ROM (EAROM), or a FLASH memory, or a combination of two or more of these, as appropriate. The RAM can be a Static Random-Access Memory (SRAM) or a Dynamic Random Access Memory (DRAM), which can be a Fast Page Mode Dynamic Random Access Memory (FPMDRAM), an Extended Data Output Dynamic Random Access Memory (EDODRAM), a Synchronous Dynamic Random-Access Memory (SDRAM), or the like, as appropriate.

[0065] The memory 404 can be used to store or cache various data files needed for processing and / or communication, and possible computer program instructions executed by the processor 402.

[0066] The processor 402 implements any of the digital pathology slice color calibration methods described above by reading and executing the computer program instructions stored in the memory 404.

[0067] Optionally, the electronic device described above can further include a transmission device 406 connected to the processor 402 and an input / output device 408 connected to the processor 402.

[0068] The transmission device 406 can be used to receive or send data via a network. Specific examples of the network can include wired or wireless networks provided by a communication provider of the electronic device. In one example, the transmission device includes a network adapter (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In one example, the transmission device 406 can be a radio frequency (RF) module for communicating with the Internet in a wireless manner.

[0069] The input / output device 408 is used to input or output information. In the present embodiment, the input information can be standard color images, color calibration images, etc., and the output information can be color calibration completed images, etc.

[0070] Optionally, in the present embodiment, the processor 402 can be configured to perform the following steps by computer program:

[0071] Obtain standard color images and color calibration images in the same field of view as the standard color images, and extract feature points of the standard color images and feature points of the color calibration images;

[0072] Match the standard color images and the color calibration images based on the feature points of the standard color images and the feature points of the color calibration images to obtain standard color matching images and color calibration matching images, and the image contents of the standard color matching images and the color calibration matching images are the same;

[0073] Perform histogram color calibration on the color calibration matching images based on the standard color matching images to obtain first color calibration images, perform CCM color calibration on the first color calibration images based on the standard color matching images to obtain second color calibration images, and perform gamma correction on the colors of the second color calibration images based on the standard color matching images to obtain color calibration completed images.

[0074] It should be noted that the specific examples in the present embodiment can refer to the examples described in the above embodiments and optional implementation manners, and the present embodiment will not be described here again.

[0075] Generally, the various embodiments can be implemented in hardware or special-purpose circuits, software, logic or any combination thereof. Some aspects of the application can be implemented in hardware, while other aspects can be implemented in firmware or software which can be executed by a controller, microprocessor or other computing device, but the application is not limited thereto. While various aspects of the application can be illustrated and described as block diagrams, flow charts, or using some other pictorial representation, it is well understood that these blocks, apparatus, systems, techniques or methods described herein can be implemented in, as non-limiting examples, hardware, software, firmware, special-purpose circuits or logic, general purpose hardware or controler or other computing devices, or some combination thereof.

[0076] Embodiments of the application can be implemented by computer software executable by a data processor of the mobile device such as in the processor entity, or by hardware, or by a combination of software and hardware. Computer software or program, also called program product, including software routines, applets and / or macros, can be stored in any apparatus-readable data storage medium and they include program instructions to implement specific tasks. The program product can include one or more computer-executable components. The one or more computer-executable components can be at least one software code or portions thereof. Further, in this regard it should be noted that any flows described herein can be embodied in computer-executable instructions, which can be used to program computers or other processors to implement aspects of the present application as methods, or as combinations of modules. Figure 5 Any block in the logical flow of the method described herein can represent a module, segment, or portion of code which comprises one or more executable instructions for implementing specific logical functions or steps in the process. The software may

[0077] It should be understood by those skilled in the art that any combination of the technical features of the above embodiments can be made, and in order to make the description concise, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered that it is within the scope of the description.

[0078] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the present application. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A color calibration method for digital pathology slides, characterized in that: The following steps are involved: A standard color image and an image to be calibrated in the same field of view as the standard color image are obtained, and feature points of the standard color image and the image to be calibrated are extracted, wherein a multi-scale space is constructed for the standard color image to obtain a standard multi-scale space, and a Hessian matrix is ​​calculated for each pixel at each scale level in the standard multi-scale space to obtain a Hessian eigenvalue. If the Hessian eigenvalue of the current pixel is the maximum value in the corresponding local area, the current pixel is used as the feature point of the standard color image, and the corresponding local area includes a first area and a second area. The first area is centered on the current pixel and includes the current pixel and pixels adjacent to the current pixel. The second area The region is a region corresponding to the first region at an adjacent scale level of the current scale level; a multi-scale space is constructed for the image to be calibrated to obtain a multi-scale space to be calibrated, and a Hessian matrix is ​​calculated for each pixel at each scale level in the multi-scale space to be calibrated to obtain a Hessian eigenvalue; if the Hessian eigenvalue of the current pixel is the maximum value in the corresponding local region, the current pixel is used as a feature point of the image to be calibrated, and the corresponding local region includes a first region and a second region, the first region is centered on the current pixel and includes the current pixel and pixels adjacent to the current pixel, and the second region is a region corresponding to the first region at an adjacent scale level of the current scale level; Matching the standard color image with the image to be color-calibrated based on feature points of the standard color image and feature points of the image to be color-calibrated to obtain a standard color matching image and an image to be color-calibrated, wherein the standard color matching image and the image to be color-calibrated have the same image content; The image to be color-calibrated is subjected to histogram color correction based on the standard color matching image to obtain a first color-calibrated image, the first color-calibrated image is subjected to CCM color correction based on the standard color matching image to obtain a second color-calibrated image, and the color of the second color-calibrated image is subjected to gamma correction based on the standard color matching image to obtain a color-calibrated image.

2. A digital pathology slide color calibration method according to claim 1, characterized in that: The feature points of the standard color image and the feature points of the image to be calibrated whose vector distance is less than a set threshold are used as matching feature points, and the geometric transformation parameters of the standard color image and the image to be calibrated are obtained. Based on the geometric transformation parameters and the matching feature points, the standard color image and the image to be calibrated are cropped to obtain the standard color matching image and the image to be calibrated.

3. The color calibration method for digital pathology sections according to claim 1, characterized in that: In the histogram color correction step, the standard color matching image is separated into RGB channels, and the histogram data distribution under each channel is calculated to obtain the standard image cumulative distribution function under each channel. The color matching image to be corrected is separated into RGB channels, and the histogram data distribution under each channel is calculated to obtain the color correction image cumulative distribution function under each channel. The pixel value of any pixel in the standard color matching image is used as the first pixel value, and the pixel value with the closest cumulative probability to the first pixel value is found in the color matching image to be corrected as the second pixel value according to the cumulative distribution function of the corresponding channel. The first pixel value is used to map the second pixel value. When the color difference between the color matching image to be corrected and the standard matching image is less than the set threshold, the histogram color correction is ended to obtain the first color correction image.

4. A digital pathology slide color calibration method according to claim 1, characterized in that: In the CCM color correction step, a CCM matrix is ​​constructed. The standard color matching image and the RGB channel matrix of the first color correction image are used as data sets. The CCM matrix is ​​optimized with the objective function of minimizing the sum of the squares of the color differences between the standard color matching image and the first color correction image. After the optimization is completed, a second color correction image is obtained. The sum of the values ​​in each row is 1, which is used as a constraint on the CCM matrix.

5. The method for color calibration of digital pathology sections according to claim 1, wherein: In the gamma correction step, dynamic gamma correction parameters are set, and the second color-calibrated image is gamma-corrected using the dynamic gamma correction parameters to obtain a gamma-calibrated image. The color difference between the standard color matching image and the gamma-calibrated image is used as a cost function. When the cost function is minimized, the gamma calibration is completed, and the current gamma-calibrated image is used as the calibration completion image.

6. A digital pathology slide color calibration device, characterized in that: include: An acquisition module is used to acquire a standard color image and an image to be calibrated that is within the same field of view as the standard color image, and extract feature points of the standard color image and the image to be calibrated, wherein a multi-scale space is constructed for the standard color image to obtain a standard multi-scale space, and a Hessian matrix is ​​calculated for each pixel at each scale level in the standard multi-scale space to obtain a Hessian eigenvalue. If the Hessian eigenvalue of the current pixel is the maximum value in the corresponding local area, the current pixel is used as a feature point of the standard color image, and the corresponding local area includes a first area and a second area. The first area is centered on the current pixel and includes the current pixel and pixels adjacent to the current pixel. The second region is a region corresponding to the first region at an adjacent scale level of the current scale level; a multi-scale space is constructed for the image to be calibrated to obtain a multi-scale space to be calibrated, and a Hessian matrix is ​​calculated for each pixel at each scale level in the multi-scale space to be calibrated to obtain a Hessian eigenvalue; if the Hessian eigenvalue of the current pixel is the maximum value in the corresponding local region, the current pixel is used as a feature point of the image to be calibrated, and the corresponding local region includes a first region and a second region, the first region is centered on the current pixel and includes the current pixel and pixels adjacent to the current pixel, and the second region is a region corresponding to the first region at an adjacent scale level of the current scale level; a feature point matching module, which matches the standard color image with the image to be calibrated based on the feature points of the standard color image and the feature points of the image to be calibrated to obtain a standard color matching image and an image to be calibrated, wherein the image content of the standard color matching image and the image to be calibrated are the same; The color correction module is used to perform histogram correction on the image to be color-corrected based on a standard color matching image to obtain a first color-corrected image, perform CCM correction on the first color-corrected image based on the standard color matching image to obtain a second color-corrected image, and perform gamma correction on the color of the second color-corrected image based on the standard color matching image to obtain a color-corrected image.

7. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to run the computer program to execute the digital pathology slide color calibration method according to any one of claims 1 to 5.

8. A readable storage medium, characterized in that: The readable storage medium stores a computer program, wherein the computer program includes a program code for controlling a process to execute a process, wherein the process includes a digital pathology slide color calibration method according to any one of claims 1 to 5.

Images