Digital pathological image browsing method based on Depzom rule plug-in

Abstract

The invention provides a digital pathological image browsing method based on a Depzom rule plug-in. The digital pathological image browsing method comprises the following steps: receiving a browsing demand sent by a user side; determining a target digital pathological image according to a browsing demand, and performing image preprocessing on the target digital pathological image to obtain a preprocessed image; performing image cutting on the preprocessed image based on a preset cutting rule according to a Depzom rule plug-in to obtain a cut image, and displaying the cut image on a terminal interface; receiving a viewing demand for the cut image sent by the user side; determining a target image area needing to be viewed by a user in the cut image according to a viewing requirement; and displaying the target image area on a terminal interface to complete digital pathological image browsing. The technical effects that the digital pathological image is rapidly loaded and browsed, the universality of the browser is improved, and then the user experience feeling is improved are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a digital pathological image browsing method based on a deepzoom rule plug-in. Background Art

[0002] With the advancement of digital technology, digital pathology has gradually developed. Digital pathology images are stored and processed by converting tissue sections into digital images. This transformation has brought many benefits to the medical field, including convenient storage and sharing, remote access and collaboration, and the possibility of automated analysis and auxiliary diagnosis using computer algorithms. Therefore, how to efficiently realize the browsing of digital pathology images has become an important technical research direction.

[0003] In the prior art, application publication number: CN115942003A discloses a pathological image reading method based on a browser server architecture, including step S1, issuing a reading request from a browser front end to a server end to obtain a pathological image to be processed; step S2, a pathological image reading module performs format-setting processing on the pathological image to be processed to obtain the pathological image after format processing; step S3, according to the image coordinate parameters in the reading request, determining the pixel data stream corresponding to the image coordinate parameters of the format-processed pathological image; step S4, the server end transmits the output image obtained according to the pixel data stream to the browser front end for output display; the invention unifies the image types of different manufacturers, does not need to install clients of different manufacturers, and can use the browser based on the browser / server architecture to read the images. Although it has strong compatibility, it is time-consuming to unify the image types of different manufacturers.

[0004] Image browsing plug-ins are software tools or extensions used to display and manipulate images in a browser or other application. They usually provide image zooming, dragging, rotating, and other interactive functions so that users can quickly browse and analyze images. In the field of pathology images, image browsing plug-ins can provide efficient image browsing functions to help users browse and analyze pathology images at different levels and details.

[0005] However, when browsing digital pathology images, there are few such professional plug-ins on the market. Instead, there are some customized, private plug-ins or tools that may take a long time to load when processing large images, and the performance on mobile devices may be reduced. Some plug-ins may even have compatibility issues on specific browsers or devices, resulting in inconsistent user experience or limited functionality. As a result, there is a lack of such a popular, universal browser plug-in for browsing digital pathology images.

[0006] In summary, how to efficiently browse digital pathology images through appropriate plug-ins to solve the problem that digital pathology images take a long time to load and need to be browsed in a specific browser, resulting in poor user experience and limited browsing functions, is one of the urgent problems to be solved in the field of image processing technology. Summary of the invention

[0007] The present invention aims to solve the technical problems in the above-mentioned technologies to at least a certain extent. To this end, the purpose of the present invention is to provide a digital pathology image browsing method based on the deepzoom rule plug-in, by determining the target digital pathology image based on the browsing requirements and performing image processing, determining the target image area that the user needs to view; displaying the target image area on the terminal interface, completing the digital pathology image browsing, thereby achieving fast loading and browsing of digital pathology images, improving the universality of the browser, and thus improving the technical effect of user experience.

[0008] The present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, comprising:

[0009] Receive browsing requests sent by the user;

[0010] Determine a target digital pathology image according to a browsing requirement, perform image preprocessing on the target digital pathology image, and obtain a preprocessed image;

[0011] According to the deepzoom rule plug-in, the pre-processed image is segmented based on the preset segmentation rules, and the segmented image is obtained and displayed on the terminal interface;

[0012] Receiving a viewing request for a cut image sent by a user terminal;

[0013] According to the viewing requirements, determine the target image area that the user needs to view in the cut image;

[0014] The target image area is displayed on the terminal interface to complete the digital pathology image browsing.

[0015] Preferably, the digital pathology image browsing method based on the deepzoom rule plug-in has the following browsing requirements:

[0016] Image number of digital pathology images.

[0017] Preferably, the digital pathology image browsing method based on the deepzoom rule plug-in performs image preprocessing on the target digital pathology image to obtain a preprocessed image, including:

[0018] Cropping the edge of the target digital pathology image to obtain a first processed image;

[0019] Performing image noise reduction processing on the first processed image to obtain a second processed image;

[0020] Perform image enhancement processing on the second processed image to obtain a preprocessed image.

[0021] Preferably, the digital pathology image browsing method based on the deepzoom rule plug-in performs image enhancement processing on the second processed image to obtain a preprocessed image, including:

[0022] Convert the preprocessed image from the RGB color space to the HSV color space to obtain a first converted image;

[0023] Obtain the brightness and chromaticity values ​​of each pixel of the first converted image in the HSV color space;

[0024] According to the brightness value of each pixel, determine the maximum brightness value and the minimum brightness value of the first converted image in the HSV color space;

[0025] The difference between the maximum brightness value and the minimum brightness value is used as the first coefficient; the value obtained by multiplying the maximum brightness value by the first adjustment parameter is used as the second coefficient; and the ratio of the second coefficient to the first coefficient is used as the brightness adjustment coefficient;

[0026] Select any pixel point, obtain the difference between the brightness value of the current pixel point and the minimum brightness value, and use the product of the difference and the brightness adjustment coefficient as the brightness adjustment value of the current pixel point;

[0027] According to the chromaticity value of each pixel, determine the maximum chromaticity value, the minimum chromaticity value and the chromaticity average value of the first converted image in the HSV color space;

[0028] The difference between the maximum chromaticity value and the minimum chromaticity value is used as the third coefficient; the value obtained by multiplying the chromaticity average value by the second adjustment parameter is used as the fourth coefficient; and the ratio of the third coefficient to the fourth coefficient is used as the chromaticity adjustment coefficient;

[0029] Select any pixel point, obtain the difference between the chromaticity value of the current pixel point and the minimum chromaticity value, and use the product of the difference and the chromaticity adjustment coefficient as the chromaticity adjustment value of the current pixel point;

[0030] Based on the brightness adjustment value and the chromaticity adjustment value, respectively, the current pixel point is adjusted for brightness and chromaticity; the above operations are performed on all the pixel points in the first conversion image to obtain a first adjusted image;

[0031] Matching the first converted image with the first adjusted image to determine angle change values ​​of corresponding pixels;

[0032] Obtaining the pixel value of each pixel in the first adjusted image, and determining whether the pixel value is less than a preset pixel value; treating the pixel values ​​less than the preset pixel value as first-category pixel points; treating the pixel values ​​greater than or equal to the preset pixel value as second-category pixel points;

[0033] Multiplying the pixel value of the first type of pixel by the reciprocal of the absolute value of the cosine value of the corresponding angle change value to obtain a first pixel adjustment value;

[0034] Multiplying the pixel value of the second type of pixel by the absolute value of the cosine value of the corresponding angle change value to obtain a second pixel adjustment value;

[0035] The first type of pixel points are adjusted according to the first pixel adjustment value, and the second type of pixel points are adjusted according to the second pixel adjustment value to obtain a second adjusted image, and the second adjusted image is used as the preprocessed image.

[0036] Preferably, in the digital pathology image browsing method based on the deepzoom rule plug-in, the preset cutting rule is: cutting the preprocessed image into a plurality of sub-images with a size of 256 pixels×256 pixels.

[0037] Preferably, the digital pathology image browsing method based on the deepzoom rule plug-in, after performing image segmentation on the preprocessed image based on the preset segmentation rule according to the deepzoom rule plug-in, obtaining the segmented image and displaying it on the terminal interface, storing the segmented image in the deepzoom rule plug-in, further comprises:

[0038] Generate several image vectors corresponding to several sub-images contained in the cut image;

[0039] Based on the deepzoom rule plug-in, several image storage groups are set, and several historical cutting images of the same type are stored in each image storage group. Each historical cutting image corresponds to a numbered folder, and each numbered folder stores several historical sub-images contained in the historical cutting images, and each historical sub-image corresponds to a historical image vector; several image vectors are added together to obtain a comprehensive image vector corresponding to the cutting image; the historical image vectors corresponding to several historical sub-images in the first numbered folder in each image storage group are added together to obtain the corresponding historical comprehensive image vector; the vector distance between the comprehensive image vector and each historical comprehensive image vector is calculated, and when the vector distance is less than the preset distance, it is determined that several sub-images contained in the cutting image belong to the current image storage group;

[0040] Acquire the image sizes of the plurality of sub-images contained in the cut image, and determine the number of sectors required to store the plurality of sub-images contained in the cut image according to the image size and a preset sector size;

[0041] Obtain the storage address of the last image in the current image storage group, and determine the target storage addresses of several sub-images contained in the cut image according to the number of sectors and the storage address of the last image;

[0042] Based on the target storage address, several sub-images contained in the cut image are stored, and the image numbers of the several sub-images contained in the cut image in the current storage group are determined, so as to complete the storage of the cut image in the deepzoom rule-based plug-in.

[0043] Preferably, the digital pathology image browsing method based on the deepzoom rule plug-in has the following viewing requirements:

[0044] The sub-image number interval or sub-image number included in the cut image.

[0045] Preferably, the digital pathology image browsing method based on the deepzoom rule plug-in determines the target image area that the user needs to view in the cut image according to the viewing requirements, including:

[0046] The sub-image number interval or sub-image number sent by the user terminal is used as the viewing requirement;

[0047] A target sub-image to be displayed in the cut image is determined according to the viewing requirement, and a target image region to be viewed by the user in the cut image is determined according to the target sub-image.

[0048] Preferably, the digital pathology image browsing method based on the deepzoom rule plug-in displays the target image area on the terminal interface, including: displaying the target image area in a target image display window and a target image navigation window of the terminal interface.

[0049] Preferably, the digital pathology image browsing method based on the deepzoom rule plug-in, after displaying the target image area on the terminal interface, further comprises:

[0050] Receiving a first operation and a second operation of a user terminal;

[0051] Performing a first operation on a target image area displayed on a terminal interface to achieve image transformation; wherein the preset first operation includes at least one of image zooming in, image zooming out, and image rotation;

[0052] A second operation is performed on the target image area displayed on the terminal interface to implement image annotation; wherein the preset second operation includes at least one of image marking and image annotation.

[0053] The present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, including: receiving a browsing demand sent by a user end; determining a target digital pathology image according to the browsing demand, performing image preprocessing on the target digital pathology image to obtain a preprocessed image; performing image cutting on the preprocessed image based on a preset cutting rule according to the deepzoom rule plug-in to obtain a cut image and display it on a terminal interface; receiving a viewing demand for the cut image sent by a user end; determining a target image area in the cut image that the user needs to view according to the viewing demand; displaying the target image area on the terminal interface to complete the digital pathology image browsing. Thus, the technical effect of quickly loading and browsing digital pathology images, improving the universality of the browser, and thus enhancing the user experience is achieved.

[0054] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0055] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0057] Figure 1 It is a flow chart of a digital pathology image browsing method based on deepzoom rule plug-in in an embodiment of the present invention;

[0058] Figure 2 This is an optional display diagram of a digital pathology image processed based on a deepzoom rule plug-in in an embodiment of the present invention;

[0059] Figure 3 This is an optional digital pathology image display interface diagram in an embodiment of the present invention;

[0060] Figure 4 This is an optional digital pathology image navigation interface diagram in an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0062] Reference Figure 1 and Figure 2, an embodiment of the present invention provides a digital pathology image browsing method based on deepzoom rule plug-in, comprising:

[0063] Step S1, receiving a browsing request sent by a user terminal;

[0064] Step S2, determining a target digital pathology image according to a browsing requirement, and performing image preprocessing on the target digital pathology image to obtain a preprocessed image;

[0065] Step S3, performing image segmentation on the preprocessed image based on the preset segmentation rule according to the deepzoom rule plug-in, obtaining a segmented image and displaying it on the terminal interface;

[0066] Step S4, receiving a viewing request for the cut image sent by the user terminal;

[0067] Step S5, determining the target image area in the cut image that the user needs to view according to the viewing requirement;

[0068] Step S6, displaying the target image area on the terminal interface to complete the digital pathology image browsing.

[0069] In this embodiment, the browsing requirement is used to determine the pictures that the user needs to browse;

[0070] In this embodiment, the specific implementation method of performing image preprocessing on the target digital pathology image to obtain the preprocessed image may be: performing operations such as noise reduction, feature extraction, contour detection, cropping and enhancement on the target digital pathology image.

[0071] In this embodiment, the preset cutting rule may be cutting the image into a specified size, cutting the image into a specified number of sheets, etc.

[0072] In this embodiment, the terminal interface may be a computer interface, a tablet interface, or a mobile phone interface.

[0073] In this embodiment, the viewing requirement is used to determine which image area of ​​the cut image the user needs to browse.

[0074] In this embodiment, the cut image is obtained and displayed on the terminal interface as follows: Figure 2 shown.

[0075] The technical principle of the above technical solution is: receiving browsing requirements sent by the user end; determining the target digital pathology image according to the browsing requirements, performing image preprocessing on the target digital pathology image to obtain a preprocessed image; performing image cutting on the preprocessed image based on a preset cutting rule according to the deepzoom rule plug-in, obtaining a cut image and displaying it on the terminal interface; receiving the viewing requirements for the cut image sent by the user end; determining the target image area in the cut image that the user needs to view according to the viewing requirements; displaying the target image area on the terminal interface to complete the browsing of the digital pathology image.

[0076] The technical effect of the above technical solution is: by determining the target digital pathology image based on browsing needs and performing image processing, the target image area that the user needs to view is determined; the target image area is displayed on the terminal interface to complete the digital pathology image browsing, thereby achieving fast loading and browsing of digital pathology images, improving the universality of the browser, and thus enhancing the user experience.

[0077] The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, wherein the browsing requirement includes an image number of the digital pathology image.

[0078] In this embodiment, the image number may be the image name when the digital pathology image is stored, for example: B1044750-123456.

[0079] The technical principle of the above technical solution is: determining the browsing needs of users through image numbers.

[0080] The technical effect of the above technical solution is: the pictures that the user needs to view are determined through browsing needs, thereby meeting the user's expected effect.

[0081] The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, which performs image preprocessing on a target digital pathology image to obtain a preprocessed image, including:

[0082] Cropping the edge of the target digital pathology image to obtain a first processed image;

[0083] Performing image noise reduction processing on the first processed image to obtain a second processed image;

[0084] Perform image enhancement processing on the second processed image to obtain a preprocessed image.

[0085] In this embodiment, the specific implementation method of cropping the edge of the target digital pathology image to obtain the first processed image may be: using the Canny algorithm to perform edge detection on the target digital pathology image, and cropping the portion outside the outermost edge. The image after edge cropping is the first processed image.

[0086] In this embodiment, a specific implementation method of performing image noise reduction processing on the first processed image to obtain the second processed image may be: performing Gaussian filtering processing on all pixels in the first processed image, and using the filtered image as the second processed image.

[0087] In this embodiment, the specific implementation method of performing image enhancement processing on the second processed image to obtain the preprocessed image may be: select any pixel point in the second processed image as the central pixel point, determine a 3*3 area centered on the central pixel point, obtain the grayscale values ​​of all pixels in the area, select 6 pixels whose grayscale values ​​are closest to the grayscale value of the central pixel point, calculate the average grayscale value of the 6 pixels, and replace the grayscale value of the central pixel point with the average grayscale value, perform the above operation on all pixels in the second processed image to obtain the preprocessed image.

[0088] The technical principle and beneficial effects of the above technical solution are: cropping the edges of the target digital pathology image to reduce the redundant parts of the image; performing image noise reduction processing on the first processed image to obtain the second processed image; performing image enhancement processing on the second processed image to obtain a preprocessed image, thereby improving the clarity of the image and facilitating subsequent image browsing.

[0089] The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, which performs image enhancement processing on a second processed image to obtain a pre-processed image, including:

[0090] Convert the preprocessed image from the RGB color space to the HSV color space to obtain a first converted image;

[0091] Obtain the brightness and chromaticity values ​​of each pixel of the first converted image in the HSV color space;

[0092] According to the brightness value of each pixel, determine the maximum brightness value and the minimum brightness value of the first converted image in the HSV color space;

[0093] The difference between the maximum brightness value and the minimum brightness value is used as the first coefficient; the value obtained by multiplying the maximum brightness value by the first adjustment parameter is used as the second coefficient; and the ratio of the second coefficient to the first coefficient is used as the brightness adjustment coefficient;

[0094] Select any pixel point, obtain the difference between the brightness value of the current pixel point and the minimum brightness value, and use the product of the difference and the brightness adjustment coefficient as the brightness adjustment value of the current pixel point;

[0095] According to the chromaticity value of each pixel, determine the maximum chromaticity value, the minimum chromaticity value and the chromaticity average value of the first converted image in the HSV color space;

[0096] The difference between the maximum chromaticity value and the minimum chromaticity value is used as the third coefficient; the value obtained by multiplying the chromaticity average value by the second adjustment parameter is used as the fourth coefficient; and the ratio of the third coefficient to the fourth coefficient is used as the chromaticity adjustment coefficient;

[0097] Select any pixel point, obtain the difference between the chromaticity value of the current pixel point and the minimum chromaticity value, and use the product of the difference and the chromaticity adjustment coefficient as the chromaticity adjustment value of the current pixel point;

[0098] Based on the brightness adjustment value and the chromaticity adjustment value, respectively, the current pixel point is adjusted for brightness and chromaticity; the above operations are performed on all the pixel points in the first conversion image to obtain a first adjusted image;

[0099] Matching the first converted image with the first adjusted image to determine angle change values ​​of corresponding pixels;

[0100] Obtaining the pixel value of each pixel in the first adjusted image, and determining whether the pixel value is less than a preset pixel value; treating the pixel values ​​less than the preset pixel value as first-category pixel points; treating the pixel values ​​greater than or equal to the preset pixel value as second-category pixel points;

[0101] Multiplying the pixel value of the first type of pixel by the reciprocal of the absolute value of the cosine value of the corresponding angle change value to obtain a first pixel adjustment value;

[0102] Multiplying the pixel value of the second type of pixel by the absolute value of the cosine value of the corresponding angle change value to obtain a second pixel adjustment value;

[0103] The first type of pixel points are adjusted according to the first pixel adjustment value, and the second type of pixel points are adjusted according to the second pixel adjustment value to obtain a second adjusted image, and the second adjusted image is used as the preprocessed image.

[0104] In this embodiment, the ratio of the maximum brightness value to the minimum brightness value may be used as the first adjustment parameter.

[0105] In this embodiment, the ratio of the maximum chromaticity value to the minimum chromaticity value may be used as the second adjustment parameter.

[0106] In this embodiment, the first conversion image is matched with the first adjustment image, and a specific implementation method for determining the angle change value of the corresponding pixel point can be: H, S, and V are used as the three coordinate axes of the three-dimensional coordinate system, and the origin of the coordinate axes is O; any pixel point A in the first conversion image is selected, and the H value, S value, and V value of the pixel point A in the first conversion image are obtained according to the channel value calculation formula of RGB to HSV, and the position of the pixel point A in the three-dimensional coordinate system is obtained, that is: A(H, S, V); the pixel point A and the coordinate origin O are connected to form a vector The coordinates are the same as A; determine the pixel point B corresponding to the pixel point A in the first adjusted image, obtain the H value, S value and V value of the pixel point B in the first converted image according to the channel value calculation formula of RGB to HSV, and obtain the position of the pixel point B in the three-dimensional coordinate system, that is: B(H, S, V); connect the pixel point B and the coordinate origin O to form a vector Coordinates are the same as B; calculation and the average of the angles of the three coordinate axes H, S, and V to obtain the first average angle; calculate and the average of the angles of the three coordinate axes H, S, and V to obtain a second average angle; the difference between the first average angle and the second average angle is used as the angle change value of the pixel A; the above operation is performed on all the pixels in the first converted image and the corresponding pixels in the first adjusted image to determine the angle change value of each pixel in the first converted image.

[0107] In this embodiment, the average pixel value of each pixel in the first adjusted image may be used as the preset pixel value.

[0108] In this embodiment, the specific implementation method of adjusting the first category of pixel points according to the first pixel adjustment value and adjusting the second category of pixel points according to the second pixel adjustment value can be: using the first pixel adjustment value corresponding to each first category pixel point as the current pixel value of the pixel point; using the second pixel adjustment value corresponding to each second category pixel point as the current pixel value of the pixel point.

[0109] The technical principle of the above technical solution is: converting the preprocessed image from the RGB color space to the HSV color space to obtain a first converted image; according to the maximum brightness value and the minimum brightness value of the first converted image in the HSV color space; determining the brightness adjustment coefficient according to the maximum brightness value and the minimum brightness value, and then determining the brightness adjustment value of the pixel point; based on the maximum chromaticity value, the minimum chromaticity value and the chromaticity average value of the first converted image in the HSV color space; and then determining the chromaticity adjustment coefficient; determining the chromaticity adjustment value of the pixel point according to the chromaticity adjustment coefficient; adjusting the brightness and chromaticity of the pixel point based on the brightness adjustment value and the chromaticity adjustment value to obtain a first adjusted image; matching the first converted image with the first adjusted image, determining the angle change value of the corresponding pixel point, classifying the pixel points in the first adjusted image according to the preset pixel value, and adjusting the pixel value of the pixel point based on the classification result and the angle change value to obtain the preprocessed image.

[0110] The technical effects of the above technical scheme are: by adjusting the brightness and chromaticity values ​​of the pixels, the preliminary enhancement processing of the image is completed; the angle change value of the pixel points is obtained through image matching, thereby eliminating the angle deviation of the pixel points caused by the preliminary noise reduction process; the pixels are classified and adjusted to obtain more accurate pixel adjustment values, thereby making the image enhancement effect more excellent and improving the user's viewing experience when viewing pictures.

[0111] The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, wherein a preset cutting rule is: cutting a preprocessed image into a plurality of sub-images of a size of 256 pixels×256 pixels.

[0112] The technical principle of the above technical solution is: cutting the pre-processed image.

[0113] The technical effect of the above technical solution is: realizing on-demand loading of images, which is beneficial to improving the speed of image loading.

[0114] The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, wherein after performing image segmentation on a preprocessed image based on a preset segmentation rule according to the deepzoom rule plug-in, obtaining a segmented image and displaying it on a terminal interface, the segmented image is stored in a deepzoom rule plug-in, and further comprising:

[0115] Generate several image vectors corresponding to several sub-images contained in the cut image;

[0116] Based on the deepzoom rule plug-in, several image storage groups are set. Each image storage group stores several historical cutting images of the same type. Each historical cutting image corresponds to a numbered folder. Each numbered folder stores several historical sub-images contained in the historical cutting image. Each historical sub-image corresponds to a historical image vector.

[0117] Adding a number of image vectors to obtain a comprehensive image vector corresponding to the cut image; adding a number of historical image vectors corresponding to the historical sub-images in the first numbered folder in each image storage group to obtain a corresponding historical comprehensive image vector; calculating the vector distance between the comprehensive image vector and each historical comprehensive image vector, and when the vector distance is less than a preset distance, determining that a number of sub-images contained in the cut image belong to the current image storage group;

[0118] Acquire the image sizes of the plurality of sub-images contained in the cut image, and determine the number of sectors required to store the plurality of sub-images contained in the cut image according to the image size and a preset sector size;

[0119] Obtain the storage address of the last image in the current image storage group, and determine the target storage addresses of several sub-images contained in the cut image according to the number of sectors and the storage address of the last image;

[0120] Based on the target storage address, several sub-images contained in the cut image are stored, and the image numbers of the several sub-images contained in the cut image in the current storage group are determined, so as to complete the storage of the cut image in the deepzoom rule-based plug-in.

[0121] In this embodiment, a feature extraction algorithm of a histogram of oriented gradients is used to extract features from an image, and a feature vector corresponding to the image is obtained, and the feature vector is used as an image vector corresponding to the image.

[0122] In this embodiment, the comprehensive image vector is determined as Determine the historical comprehensive image vector as The vector distance is

[0123] In this embodiment, the sector size may be 128KB.

[0124] In this embodiment, a specific implementation method for determining the number of sectors required to store several sub-images contained in a cut image based on the image size and the preset sector size may be: adding the image sizes of several sub-images to obtain the total image size, dividing the total image size by the sector size as the number of sectors, and rounding up if the result is not an integer.

[0125] In this embodiment, the specific implementation method of determining the target storage addresses of several sub-images contained in the cut image based on the number of sectors and the storage address of the last image can be: multiplying the number of sectors by the sector size to obtain the memory size occupied by the sector; determining the storage address range corresponding to the memory occupied by the sector; adding the storage address of the last image to the storage address range to obtain the target storage addresses of the several sub-images contained in the cut image.

[0126] The technical principle of the above technical solution is: generate several image vectors corresponding to several sub-images contained in the cut image; add several image vectors to obtain the comprehensive image vector corresponding to the cut image; add the historical image vectors corresponding to several historical sub-images in the first numbered folder in each image storage group to obtain the corresponding historical comprehensive image vector; calculate the vector distance between the comprehensive image vector and each historical comprehensive image vector, and when the vector distance is not greater than the preset distance, determine that the several sub-images contained in the cut image belong to the current image storage group; obtain the image size of the several sub-images contained in the cut image, and determine the number of sectors required to store the several sub-images contained in the cut image according to the image size and the preset sector size; obtain the storage address of the last image in the current image storage group, and determine the target storage address of the several sub-images contained in the cut image according to the number of sectors and the storage address of the last image; store the several sub-images contained in the cut image based on the target storage address, and determine the image numbers of the several sub-images contained in the cut image in the current storage group, and complete the storage of the cut image in the deepzoom rule-based plug-in.

[0127] The technical effect of the above technical solution is: by determining the distance between image vectors to group and store images, it is beneficial to search for digital pathology images and improves the loading efficiency of digital pathology images.

[0128] The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, wherein the viewing requirement includes a sub-image number interval or a sub-image number contained in a cut image.

[0129] In this embodiment, the sub-image number interval may be image 01 to image 07;

[0130] In this embodiment, the sub-image number may be image 02;

[0131] The technical principle of the above technical solution is: receiving the sub-image number interval or sub-image number sent by the user terminal, and determining the viewing requirement according to the sub-image number interval or sub-image number.

[0132] The technical effect of the above technical solution is: determining the viewing requirements through the sub-image number interval or the sub-image number, realizing on-demand loading, and improving the loading rate and user experience.

[0133] The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, which determines a target image area in a cut image that a user needs to view according to viewing requirements, including:

[0134] The sub-image number interval or sub-image number sent by the user terminal is used as the viewing requirement;

[0135] A target sub-image to be displayed in the cut image is determined according to the viewing requirement, and a target image region to be viewed by the user in the cut image is determined according to the target sub-image.

[0136] The technical principle of the above technical solution is: determine the viewing requirements based on the sub-image number interval or sub-image number sent by the user end; determine the target sub-image that needs to be displayed in the cut image based on the viewing requirements, and determine the target image area that the user needs to view in the cut image based on the target sub-image.

[0137] The technical effect of the above technical solution is: determining the target sub-image to be displayed in the cut image according to the sub-image number interval or the sub-image number, and then determining the target image area, thereby realizing on-demand loading of the image, improving the loading rate and user experience.

[0138] Reference Figure 3 and Figure 4 The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, which displays a target image area on a terminal interface, including: displaying the target image area on a target image display window and a target image navigation window of the terminal interface.

[0139] In this embodiment, the target image display window is as follows: Figure 3 The display window on the right.

[0140] In this embodiment, the target image navigation window is as follows: Figure 4 The arrow points to the display window.

[0141] The technical principle of the above technical solution is: displaying the target image area in the target image display window and the target image navigation window of the terminal interface.

[0142] The technical effects of the above technical solution are: displaying the target image area in the target image display window of the terminal interface is beneficial to observing the details of the image; displaying the target image area in the target image navigation window can quickly select the image area and improve the image browsing efficiency.

[0143] The embodiment of the present invention provides a digital pathology image browsing method based on a deepzoom rule plug-in, which, after displaying the target image area on a terminal interface, further includes:

[0144] Receiving a first operation and a second operation of a user terminal;

[0145] Performing a first operation on a target image area displayed on a terminal interface to achieve image transformation; wherein the first operation includes at least one of image zooming in, image zooming out, and image rotation;

[0146] A second operation is performed on the target image area displayed on the terminal interface to implement image labeling; wherein the second operation includes at least one of image labeling and image annotation.

[0147] The technical principle of the above technical solution is: performing image transformation and image annotation on a plurality of sub-images contained in a target image area displayed on a terminal interface.

[0148] The technical effect of the above technical solution is: by performing image transformation on the image, various parts of the image can be flexibly viewed to improve browsing efficiency; by performing image annotation on the image, the key parts of the image can be clarified, which is beneficial to improving the browsing efficiency of the subsequent browsing process.

[0149] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A digital pathology image browsing method based on deepzoom rule plug-in, characterized in that: include: Receive browsing requests sent by the user; Determine a target digital pathology image according to a browsing requirement, and perform image preprocessing on the target digital pathology image to obtain a preprocessed image; Performing image segmentation on the pre-processed image based on a preset segmentation rule according to the deepzoom rule plug-in, obtaining a segmented image and displaying it on a terminal interface; Receiving a viewing request for a cut image sent by a user terminal; According to the viewing requirements, determine the target image area that the user needs to view in the cut image; The target image area is displayed on the terminal interface to complete the digital pathology image browsing.

2. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 1, characterized in that: The browsing requirement includes the image number of the digital pathology image.

3. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 1, characterized in that: Performing image preprocessing on the target digital pathology image to obtain a preprocessed image includes: Cropping the edge of the target digital pathology image to obtain a first processed image; Performing image noise reduction processing on the first processed image to obtain a second processed image; Perform image enhancement processing on the second processed image to obtain a preprocessed image.

4. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 3, characterized in that: Performing image enhancement processing on the second processed image to obtain a preprocessed image includes: Convert the preprocessed image from the RGB color space to the HSV color space to obtain a first converted image; Obtaining the brightness value and chromaticity value of each pixel of the first converted image in the HSV color space; Determine the maximum brightness value and the minimum brightness value of the first converted image in the HSV color space according to the brightness value of each pixel; The difference between the maximum brightness value and the minimum brightness value is used as the first coefficient; the value obtained by multiplying the maximum brightness value by the first adjustment parameter is used as the second coefficient; and the ratio of the second coefficient to the first coefficient is used as the brightness adjustment coefficient; Select any pixel point, obtain the difference between the brightness value of the current pixel point and the minimum brightness value, and use the product of the difference and the brightness adjustment coefficient as the brightness adjustment value of the current pixel point; Determine the maximum chromaticity value, the minimum chromaticity value and the average chromaticity value of the first converted image in the HSV color space according to the chromaticity value of each pixel; The difference between the maximum chromaticity value and the minimum chromaticity value is used as the third coefficient; the value obtained by multiplying the chromaticity average value by the second adjustment parameter is used as the fourth coefficient; and the ratio of the third coefficient to the fourth coefficient is used as the chromaticity adjustment coefficient; Select any pixel point, obtain the difference between the chromaticity value of the current pixel point and the minimum chromaticity value, and use the product of the difference and the chromaticity adjustment coefficient as the chromaticity adjustment value of the current pixel point; Based on the brightness adjustment value and the chromaticity adjustment value, respectively, the current pixel point is adjusted for brightness and chromaticity; the above operations are performed on all the pixel points in the first conversion image to obtain a first adjusted image; Matching the first converted image with the first adjusted image to determine angle change values ​​of corresponding pixels; Obtaining the pixel value of each pixel in the first adjusted image, and determining whether the pixel value is less than a preset pixel value; treating the pixel values ​​less than the preset pixel value as first-category pixel points; treating the pixel values ​​greater than or equal to the preset pixel value as second-category pixel points; Multiplying the pixel value of the first type of pixel by the reciprocal of the absolute value of the cosine value of the corresponding angle change value to obtain a first pixel adjustment value; Multiplying the pixel value of the second type of pixel by the absolute value of the cosine value of the corresponding angle change value to obtain a second pixel adjustment value; The first type of pixel points are adjusted according to the first pixel adjustment value, and the second type of pixel points are adjusted according to the second pixel adjustment value to obtain a second adjusted image, and the second adjusted image is used as the preprocessed image.

5. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 1, characterized in that: The preset cutting rule is: cutting the preprocessed image into a plurality of sub-images with a size of 256 pixels×256 pixels.

6. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 1, characterized in that: After performing image segmentation on the preprocessed image based on the preset segmentation rule according to the deepzoom rule plug-in, obtaining the segmented image and displaying it on the terminal interface, storing the segmented image in the deepzoom rule plug-in, further comprising: Generate several image vectors corresponding to several sub-images contained in the cut image; Based on the deepzoom rule plug-in, several image storage groups are set. Each image storage group stores several historical cutting images of the same type. Each historical cutting image corresponds to a numbered folder. Each numbered folder stores several historical sub-images contained in the historical cutting image. Each historical sub-image corresponds to a historical image vector. Adding the plurality of image vectors to obtain a comprehensive image vector corresponding to the cut image; adding the historical image vectors corresponding to the plurality of historical sub-images in the first numbered folder in each image storage group to obtain a corresponding historical comprehensive image vector; calculating the vector distance between the comprehensive image vector and each historical comprehensive image vector, and when the vector distance is less than a preset distance, determining that the plurality of sub-images contained in the cut image belong to the current image storage group; Acquire the image sizes of the plurality of sub-images contained in the cut image, and determine the number of sectors required to store the plurality of sub-images contained in the cut image according to the image size and a preset sector size; Acquire the storage address of the last image in the current image storage group, and determine the target storage addresses of a plurality of sub-images included in the cut image according to the number of sectors and the storage address of the last image; The plurality of sub-images contained in the cut image are stored based on the target storage address, and the image numbers of the plurality of sub-images contained in the cut image in the current storage group are determined, so as to complete the storage of the cut image in the deepzoom rule-based plug-in.

7. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 1, characterized in that: The viewing requirement includes a sub-image number interval or a sub-image number included in the cut image.

8. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 1, characterized in that: According to the viewing requirements, determine the target image area that the user needs to view in the cut image, including: The sub-image number interval or sub-image number sent by the user terminal is used as the viewing requirement; A target sub-image to be displayed in the cut image is determined according to the viewing requirement, and a target image region to be viewed by the user in the cut image is determined according to the target sub-image.

9. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 1, characterized in that: Displaying the target image area on a terminal interface includes: displaying the target image area in a target image display window and a target image navigation window of the terminal interface.

10. The digital pathology image browsing method based on deepzoom rule plug-in according to claim 1, characterized in that: After displaying the target image area on the terminal interface, the method further includes: Receiving a first operation and a second operation of a user terminal; Performing a first operation on a target image area displayed on a terminal interface to achieve image transformation; wherein the first operation includes at least one of image zooming in, image zooming out, and image rotation; A second operation is performed on the target image area displayed on the terminal interface to implement image annotation; wherein the preset second operation includes at least one of image marking and image annotation.

Images