CN115272362A - Method and device for segmenting effective area of digital pathology full-field image - Google Patents
Method and device for segmenting effective area of digital pathology full-field image Download PDFInfo
- Publication number
- CN115272362A CN115272362A CN202210710971.XA CN202210710971A CN115272362A CN 115272362 A CN115272362 A CN 115272362A CN 202210710971 A CN202210710971 A CN 202210710971A CN 115272362 A CN115272362 A CN 115272362A
- Authority
- CN
- China
- Prior art keywords
- image
- tissue
- area
- gray
- thumbnail
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000007170 pathology Effects 0.000 title claims abstract description 43
- 238000005457 optimization Methods 0.000 claims abstract description 27
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 238000012805 post-processing Methods 0.000 claims abstract description 21
- 230000002093 peripheral effect Effects 0.000 claims abstract description 18
- 230000002708 enhancing effect Effects 0.000 claims abstract description 15
- 230000000694 effects Effects 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 15
- 230000000877 morphologic effect Effects 0.000 claims description 14
- 230000008859 change Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 230000008520 organization Effects 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 6
- 238000013507 mapping Methods 0.000 abstract 1
- 238000004458 analytical method Methods 0.000 description 11
- 238000013473 artificial intelligence Methods 0.000 description 8
- 239000012535 impurity Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 230000001575 pathological effect Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010186 staining Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G06T3/04—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration by the use of histogram techniques
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
Abstract
The application relates to a method and a device for segmenting an effective region of a digital pathology full-field image. The method comprises the following steps: acquiring a thumbnail image of a digital pathology full-field image, and obtaining a scaling; removing noises such as characters, shadow, foreign matters and the like on the thumbnail; converting the thumbnail into a gray-scale image; enhancing the contrast of the gray level image by using a histogram equalization algorithm; after the gray level image with the enhanced contrast ratio is subjected to Gaussian blur, converting the image into a binary image by using a threshold value method; carrying out post-processing optimization on the generated binary image; and determining the peripheral outline coordinates of the effective area according to the gradient, and mapping the coordinates onto the full-field map according to the scaling obtained in the first step.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a method and an apparatus for segmenting an effective region of a digital pathology full-field image.
Background
With the development of digital pathology full-field image scanning technology, a digital scanner is appeared, pathological sections are scanned and imaged, and an ultrahigh-resolution full-field image file with more than one billion pixels is generated.
At present, the mainstream method for performing artificial intelligence analysis on a digital pathology whole-field image is to use a sliding window method to divide the digital pathology whole-field image into hundreds of small images with fixed sizes, then send the small images into a GPU once for calculation and analysis, and finally count and summarize the analysis results of all the small images to form the final analysis result of the digital pathology whole-field image.
However, there are a lot of blank and irrelevant areas on a full field image of digital pathology. The effective tissue area may only occupy one tenth or even less of the whole digital pathology whole-field image, and the whole digital pathology whole-field image is directly analyzed, so that a large amount of analysis time and calculation power are wasted in irrelevant areas, and impurities, dye liquor or character marks in the irrelevant areas cannot be effectively identified by an artificial intelligence algorithm, so that the accuracy of artificial intelligence analysis is also interfered.
Disclosure of Invention
In view of the above, it is necessary to provide a method for segmenting an effective region of a digital pathology full-field image.
A method for segmenting effective areas of a digital pathology full-field image, comprising the following steps:
acquiring a to-be-processed tissue thumbnail image processed by a digital pathology full-field image according to a preset scaling;
removing noise on the to-be-processed tissue thumbnail image to obtain a tissue effect image;
converting the tissue effect image into a gray-scale image, and enhancing the contrast of the gray-scale image by using a histogram equalization algorithm to obtain a tissue enhanced image;
performing Gaussian blur on the tissue enhanced image, and converting the image into a binary image by using a threshold value method;
carrying out post-processing optimization on the binary image according to the foreground and the background to obtain an organization area optimization image;
determining the peripheral contour coordinate of the target area according to the gray value gradient change of the tissue area optimization image, and converting the peripheral contour coordinate into the full-field image coordinate according to a preset scaling.
In one embodiment, acquiring a thumbnail image of a tissue to be processed, the thumbnail image being processed by a preset scaling of a digital pathology full-field image, further includes: and zooming the digital pathology full-field image into a tissue thumbnail image to be processed according to a preset zooming scale.
In one embodiment, the noise includes text noise, shadow noise, and foreign object noise;
removing noise on the thumbnail image of the tissue to be processed to obtain a tissue effect image, wherein the tissue effect image comprises:
converting the tissue thumbnail image to be processed from an RGB color space into an HSV color space to obtain an HSV tissue color image;
and for pixel points of which the H value and the S value in the HSV tissue color image are lower than the color threshold, covering the pixel points in white to obtain a tissue effect image.
In one embodiment, enhancing the contrast of the gray-scale image using a histogram equalization algorithm to obtain a tissue-enhanced image comprises:
dividing the gray level image into a preset number of image blocks;
and enhancing the contrast of each image block by using a histogram equalization algorithm, and combining to obtain a tissue enhanced image.
In one embodiment, the post-processing optimization of the generated binary image further includes: removing the holes in the foreground area of the binary image by adopting a morphological closing operation, and removing the fragmentary targets outside the foreground area by adopting a morphological opening operation; the foreground area is an effective tissue or cell area, and the fragmentary target comprises a fragmentary foreground area and a fragmentary error area.
In the generated binary image, the foreground is an effective tissue or cell area, and the background is an irrelevant area. In the generated original binary image, fragmentary background holes partially in the effective region and fragmentary foreground regions partially outside the tissue region exist, and the fragmentary wrong positions need to be removed through post-processing optimization. In the invention, the original binary image is subjected to morphological closing operation to remove holes in the foreground region, and morphological opening operation is used to remove fragmented objects dissociating outside a large foreground region. After the errors are removed, the foreground of the binary image is expanded to a certain extent, so that the range of the foreground can completely include the effective tissue or cell area. And (4) according to the gray value gradient change of the binary image, calculating the edge contour coordinate of the foreground, and multiplying by the scaling of the first step to obtain the contour coordinate of the effective area on the full-field image original image.
A digital pathology full-field image active area segmentation device, the device comprising:
the thumbnail acquisition module is used for acquiring a thumbnail image of the tissue to be processed, which is processed by the digital pathology full-field image according to a preset scaling;
the noise processing module is used for removing noise on the to-be-processed tissue thumbnail image to obtain a tissue effect image;
the contrast enhancement module is used for converting the tissue effect image into a gray image and enhancing the contrast of the gray image by using a histogram equalization algorithm to obtain a tissue enhancement image;
the binary image conversion module is used for carrying out Gaussian blur on the tissue enhancement image and converting the image into a binary image by using a threshold value method;
the post-processing module is used for carrying out post-processing optimization on the binary image according to the foreground and the background to obtain an organization area optimized image;
and the contour coordinate calculation module is used for determining the peripheral contour coordinate of the target region according to the gray value gradient change of the tissue region optimization image, and converting the peripheral contour coordinate into the full-field image coordinate according to a preset scaling.
A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
acquiring a to-be-processed tissue thumbnail image of a digital pathology full-field image processed according to a preset scaling;
removing noise on the to-be-processed tissue thumbnail image to obtain a tissue effect image;
converting the tissue effect image into a gray image, and enhancing the contrast of the gray image by using a histogram equalization algorithm to obtain a tissue enhanced image;
performing Gaussian blur on the tissue enhanced image, and converting the image into a binary image by using a threshold value method;
carrying out post-processing optimization on the binary image according to the foreground and the background to obtain an optimized image of the tissue region;
determining the peripheral contour coordinate of the target area according to the gray value gradient change of the tissue area optimization image, and converting the peripheral contour coordinate into the full-field image coordinate according to a preset scaling.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring a to-be-processed tissue thumbnail image processed by a digital pathology full-field image according to a preset scaling;
removing noise on the tissue thumbnail image to be processed to obtain a tissue effect image;
converting the tissue effect image into a gray-scale image, and enhancing the contrast of the gray-scale image by using a histogram equalization algorithm to obtain a tissue enhanced image;
carrying out Gaussian blur on the tissue enhanced image, and converting the image into a binary image by using a threshold value method;
carrying out post-processing optimization on the binary image according to the foreground and the background to obtain an organization area optimization image;
determining the peripheral contour coordinate of the target area according to the gray value gradient change of the tissue area optimization image, and converting the peripheral contour coordinate into the full-field image coordinate according to a preset scaling.
According to the method and the device for segmenting the effective region of the digital pathology full-field image, the effective region can be rapidly positioned and segmented to obtain the effective tissue or cell region through rapid focusing of the effective region, so that an artificial intelligence algorithm only analyzes the effective region, a large amount of analysis time and calculation power are saved, and meanwhile the robustness and accuracy of artificial intelligence analysis are improved.
Drawings
FIG. 1 is a flowchart illustrating a method for segmenting an effective region of a full-field digital pathology image according to an embodiment;
FIG. 2 is a schematic illustration of the effect of processing in one embodiment;
FIG. 3 is a schematic structural diagram of an embodiment of a device for segmenting an effective region of a full-field digital pathology image;
FIG. 4 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.
In one embodiment, as shown in fig. 1, there is provided a method for segmenting an effective region of a digital pathology full-field image, comprising the following steps:
step S110, acquiring a to-be-processed tissue thumbnail image processed by the digital pathology full-field image according to a preset scaling.
Wherein, the effective tissue or cell area is determined on the digital pathology full field image, firstly, the global visual field of the pathological section is needed to be obtained. On stained and slice-processed sections, the region of the effective tissue can be clearly distinguished even in a lower-magnification field of view. Under the premise of fully balancing the calculation time saved by the low-power visual field and the loss of edge precision, the magnification of the thumbnail acquired by the method is 0.625. And after the thumbnail is obtained, remapping the effective area outline coordinates calculated on the thumbnail back to the full-field original image according to the scaling. In the present invention, a thumbnail magnification of 1.25 times is obtained, and if a digital pathology whole field image is scanned using a scanner with a magnification of 40 times, the scaling is 40/0.625=64. It can be seen that the length of a side of a thumbnail is sixty-fourth of the original full field map.
And step S120, removing noise on the to-be-processed tissue thumbnail image to obtain a tissue effect image.
The thumbnail is converted from an RGB (R: red, G: green and B: blue) color space into an HSV (H: hue, S: saturation and V: brightness) color space, experiments are carried out by using a large number of slices to know that the values of H and S channels of the noise on the thumbnail are obviously lower than a normal tissue area, so that the values of the H and S channels are subjected to threshold processing, when the values of H and S at a certain position are lower than a threshold value, the position can be determined to be positioned on the noise interferent, and then the positions are directly covered with white, so that the effect of removing the noise is achieved.
Step S130, converting the tissue effect image into a gray scale image, and enhancing the contrast of the gray scale image by using a histogram equalization algorithm to obtain a tissue enhanced image.
Wherein, the section design with lighter staining is aimed at. The method is realized by adaptive histogram equalization (CLAHE) which limits the contrast. The method is to divide the image into small blocks with the same size and then perform histogram equalization on each small block respectively. It is also possible to prevent noise from also being enhanced by limiting the contrast.
And step S140, performing Gaussian blur on the tissue enhanced image, and converting the image into a binary image by using a threshold value method.
The image is subjected to noise reduction processing through a Gaussian filter, and besides noise reduction, the outline edge of the tissue can be smoother. Then, the gray value of the image is normalized to be in the interval of [0, 1], a threshold value is set, the image is subjected to binarization processing by using a threshold value method, the area with the gray value higher than the threshold value is taken as a background, and the area with the gray value lower than the threshold value is taken as a foreground.
And S150, carrying out post-processing optimization on the binary image according to the foreground and the background to obtain an optimized image of the tissue region.
In the generated binary image, the foreground is an effective tissue or cell area, and the background is an irrelevant area. In the generated original binary image, fragmentary background holes partially in the effective region and fragmentary foreground regions partially outside the tissue region exist, and the fragmentary error positions need to be removed through post-processing optimization. In the invention, the original binary image is subjected to morphological closing operation to remove holes in the foreground region, and morphological opening operation is used to remove fragmented objects dissociating outside a large foreground region. After removing the errors, the foreground of the binary image is expanded to a certain extent, so that the range of the foreground can completely include the effective tissue or cell area. And (4) solving the edge contour coordinate of the foreground according to the gray value gradient change of the binary image, and multiplying by the scaling of the first step to obtain the contour coordinate of the effective area on the full-field original image.
Step S160, determining the peripheral contour coordinate of the target area according to the gray value gradient change of the tissue area optimization image, and converting the peripheral contour coordinate into the full-field image coordinate according to the preset scaling.
The outer contour coordinates of the target area are determined according to the gray value gradient change, and edge detection can be performed by adopting a Canny operator, a Sobel operator, a LOG operator or a Laplacian operator to obtain the outer contour coordinates of the target area.
In the method for segmenting the effective area of the digital pathology full-field image, a thumbnail image of the digital pathology full-field image is obtained, and a scaling ratio is obtained; removing noises such as characters, shadow, foreign matters and the like on the thumbnail; converting the thumbnail into a grey-scale map; enhancing the contrast of the gray level image by using a histogram equalization algorithm; after the gray level image with the enhanced contrast ratio is subjected to Gaussian blur, converting the image into a binary image by using a threshold value method; carrying out post-processing optimization on the generated binary image; peripheral contour coordinates of the effective area are determined according to the gradient, the coordinates are mapped to the full-field image according to the scaling obtained in the first step, and the effective tissue or cell area can be quickly positioned and segmented, so that the artificial intelligence algorithm only analyzes the effective area, a large amount of analysis time and calculation power are saved, and meanwhile, the robustness and the accuracy of the artificial intelligence analysis are improved.
In one embodiment, the acquiring a thumbnail image of a tissue to be processed of the digital pathology full-field image processed according to a preset scaling further includes: and zooming the digital pathology full-field image into a tissue thumbnail image to be processed according to a preset zooming scale.
In one embodiment, the noise includes text noise, shadow noise, and alien noise. Removing noise on the thumbnail image of the tissue to be processed to obtain a tissue effect image, and further comprising: converting the tissue thumbnail image to be processed from an RGB color space into an HSV color space to obtain an HSV tissue color image; and for the pixel points of which the H value and the S value are lower than the color threshold value in the HSV tissue color image, covering the pixel points with white to obtain a tissue effect image.
In one embodiment, the histogram equalization algorithm is used to enhance the contrast of the gray-scale image, so as to obtain a tissue enhanced image, and the method further includes: dividing the gray level image into a preset number of image blocks; and enhancing the contrast of each image block by using a histogram equalization algorithm, and then combining to obtain a tissue enhanced image.
In one embodiment, the tissue enhanced image is subjected to gaussian blurring and the image is converted into a binary image by using a threshold method, and the method further comprises the following steps: processing the tissue enhancement image through a Gaussian filter to obtain a noise reduction binary image; normalizing the gray value of the noise-reduction binary image to be in the interval of [0, 1], taking the area with the gray value higher than the set gray threshold value as a background area and the area with the gray value lower than or equal to the set gray threshold value as a foreground area according to the set gray threshold value, and obtaining the binary image about the background area and the foreground area.
In one embodiment, the post-processing optimization of the generated binary image comprises: removing holes in the foreground area of the binary image by adopting a morphological closing operation, and removing fragmentary targets outside the foreground area by adopting a morphological opening operation; the foreground area is an effective tissue or cell area, and the fragmentary target comprises a fragmentary foreground area and a fragmentary error area.
As shown in fig. 2, in the generated binary image, the foreground region is an effective tissue or cell region, and as shown in fig. 2, the background region is an irrelevant region; in the generated binary image, fragmentary background holes partially in the foreground region exist, such as the holes in the effective region in fig. 2; and a fragmentary foreground region partially outside the large tissue region, the fragmentary foreground region being some fragmentary small tissue regions separated from the large tissue region, such regions being tissue fragments accidentally incidental during flaking, such as dye liquor impurities in fig. 2; there are also fragmentary error regions, which are areas of impurities in the image, such as impurities including glass fragments, marker marks, etc., such as the marked text noise and foreign object noise in fig. 2, which need to be removed by post-processing optimization. In the invention, the generated binary image is subjected to morphological closing operation to remove holes in the foreground region, and morphological opening operation is used to remove fragmented objects which are free outside a large foreground region. After the errors are removed, performing certain expansion processing on the foreground of the binary image to enable the range of the foreground to completely comprise an effective tissue or cell area, solving edge contour coordinates of the foreground according to the gray value gradient change of the binary image, and multiplying the edge contour coordinates by the scaling of the first step to obtain contour coordinates of the effective area on the full-field original image.
In the embodiment, the filtering of the null-breaking targets can be beneficial to the analysis of the artificial intelligence algorithm on the effective area in the later period, and the processing efficiency is improved.
It should be understood that, although the steps in the flowchart of fig. 1 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 1 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.
In one embodiment, as shown in fig. 3, there is provided a digital pathology full-field image effective region segmentation apparatus, the apparatus comprising: a thumbnail acquisition module 310, a noise processing module 320, a contrast enhancement module 330, a binary image conversion module 340, a post-processing module 350, and a contour coordinate calculation module 360, wherein: a thumbnail obtaining module 310, configured to obtain a thumbnail image of a tissue to be processed, where the digital pathology full-field image is processed according to a preset scaling; the noise processing module 320 is configured to remove noise on the tissue thumbnail image to be processed to obtain a tissue effect image; a contrast enhancement module 330, configured to convert the tissue effect image into a grayscale image, and enhance the contrast of the grayscale image by using a histogram equalization algorithm to obtain a tissue enhanced image; a binary image conversion module 340, configured to perform gaussian blur on the tissue enhanced image, and convert the image into a binary image by using a threshold method; the post-processing module 350 is configured to perform post-processing optimization on the binary image according to the foreground and the background to obtain an organization region optimized image; and the contour coordinate calculation module 360 is configured to determine a peripheral contour coordinate of the target region according to the gray value gradient change of the tissue region optimization image, and convert the peripheral contour coordinate into a full-field image coordinate according to a preset scaling.
In one embodiment, the thumbnail acquiring module 310 is further configured to scale the digital pathology full-field image into a thumbnail image of the tissue to be processed according to a preset scaling.
In one embodiment, the noise processing module 320 includes: the HSV tissue color image conversion unit is used for converting the to-be-processed tissue thumbnail image from an RGB color space to an HSV color space to obtain an HSV tissue color image; and the tissue effect image acquisition unit is used for covering the pixel points of which the H value and the S value are lower than the color threshold value in the HSV tissue color image with white to obtain the tissue effect image.
In one embodiment, the contrast enhancement module 330 includes: a dividing unit for dividing the gray image into a preset number of image blocks; and the contrast enhancement unit is used for enhancing the contrast of each image block by using a histogram equalization algorithm and then combining the image blocks to obtain a tissue enhanced image.
In one embodiment, the binary image conversion module 340 includes: the noise reduction unit is used for processing the tissue enhancement image through a Gaussian filter to obtain a noise reduction binary image; and the binary image conversion unit is used for normalizing the gray value of the noise-reduced binary image to be in the interval of [0, 1], taking the area with the gray value higher than the set gray threshold value as a background area and taking the area with the gray value lower than or equal to the set gray threshold value as a foreground area according to the set gray threshold value, and obtaining the binary image about the background area and the foreground area.
In one embodiment, the post-processing module 350 is further configured to remove the holes in the foreground region of the binary image by using a morphological close operation, and remove the fragmentary objects outside the foreground region by using a morphological open operation; the foreground area is an effective tissue or cell area, and the fragmentary target comprises a fragmentary foreground area and a fragmentary error area.
For specific definition of the device for rapidly segmenting the effective region in the digital pathology full-field image, reference may be made to the above definition of the method for segmenting the effective region in the image, and details are not described here. The modules in the device for rapidly segmenting the effective region of the digital pathology full-field image can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules. In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 4. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program to run on the non-volatile storage medium. The database of the computer device is used to store the thumbnail data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a fast and efficient area focusing method.
It will be appreciated by those skilled in the art that the configuration shown in fig. 4 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In an embodiment, a computer device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.
All possible combinations of the technical features in the above embodiments may not be described for the sake of brevity, but should be considered as being within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is specific and detailed, but not to be understood as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.
Claims (9)
1. A method for segmenting effective regions of a digital pathology full-field image is characterized by comprising the following steps:
acquiring a to-be-processed tissue thumbnail image of a digital pathology full-field image processed according to a preset scaling;
removing noise on the to-be-processed tissue thumbnail image to obtain a tissue effect image;
converting the tissue effect image into a gray-scale image, and enhancing the contrast of the gray-scale image by using a histogram equalization algorithm to obtain a tissue enhanced image;
performing Gaussian blur on the tissue enhanced image, and converting the image into a binary image by using a threshold value method;
carrying out post-processing optimization on the binary image according to the foreground and the background to obtain an organization area optimization image;
and determining the peripheral contour coordinate of the target area according to the gray value gradient change of the tissue area optimization image, and converting the peripheral contour coordinate into the full-field image coordinate according to a preset scaling.
2. The method of claim 1, wherein obtaining the thumbnail image of the tissue to be processed with the digital pathology full field image processed according to the preset scaling further comprises: and zooming the digital pathology full-field image into a tissue thumbnail image to be processed according to a preset zooming scale.
3. The method of claim 1, wherein the noise comprises text noise, shadow noise, and foreign object noise;
removing noise on the thumbnail image of the tissue to be processed to obtain a tissue effect image, and further comprising:
converting the tissue thumbnail image to be processed from an RGB color space into an HSV color space to obtain an HSV tissue color image;
and for pixel points of which the H value and the S value in the HSV tissue color image are lower than the color threshold, covering the pixel points in white to obtain a tissue effect image.
4. The method of claim 1, wherein enhancing contrast of the grayscale image using a histogram equalization algorithm results in a tissue enhanced image, further comprising:
dividing the gray level image into a preset number of image blocks;
and enhancing the contrast of each image block by using a histogram equalization algorithm, and combining to obtain a tissue enhanced image.
5. The method of claim 1, wherein the tissue enhanced image is gaussian blurred and converted to a binary image using a thresholding method, further comprising:
processing the tissue enhancement image through a Gaussian filter to obtain a noise reduction binary image;
normalizing the gray value of the noise-reduction binary image to be in the interval of [0, 1], taking the area with the gray value higher than the set gray threshold value as a background area and the area with the gray value lower than or equal to the set gray threshold value as a foreground area according to the set gray threshold value, and obtaining the binary image about the background area and the foreground area.
6. The method of claim 1, wherein post-processing optimization of the generated binary image comprises:
removing holes in the foreground area of the binary image by adopting a morphological closing operation, and removing fragmentary targets outside the foreground area by adopting a morphological opening operation; wherein, the foreground region is an effective tissue or cell region, and the fragmentary target comprises a fragmentary foreground region and a fragmentary error region.
7. An apparatus for segmenting an effective region of a digital pathology full-field image, the apparatus comprising:
the thumbnail acquisition module is used for acquiring a thumbnail image of the tissue to be processed, which is processed by the digital pathology full-field image according to a preset scaling;
the noise processing module is used for removing noise on the to-be-processed tissue thumbnail image to obtain a tissue effect image;
the contrast enhancement module is used for converting the tissue effect image into a gray image and enhancing the contrast of the gray image by using a histogram equalization algorithm to obtain a tissue enhancement image;
the binary image conversion module is used for carrying out Gaussian blur on the tissue enhanced image and converting the image into a binary image by using a threshold value method;
the post-processing module is used for carrying out post-processing optimization on the binary image according to the foreground and the background to obtain an organization area optimized image;
and the contour coordinate calculation module is used for determining the peripheral contour coordinate of the target region according to the gray value gradient change of the tissue region optimization image, and converting the peripheral contour coordinate into the full-field image coordinate according to a preset scaling.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210710971.XA CN115272362A (en) | 2022-06-22 | 2022-06-22 | Method and device for segmenting effective area of digital pathology full-field image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210710971.XA CN115272362A (en) | 2022-06-22 | 2022-06-22 | Method and device for segmenting effective area of digital pathology full-field image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115272362A true CN115272362A (en) | 2022-11-01 |
Family
ID=83761557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210710971.XA Pending CN115272362A (en) | 2022-06-22 | 2022-06-22 | Method and device for segmenting effective area of digital pathology full-field image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115272362A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117333373A (en) * | 2023-12-01 | 2024-01-02 | 武汉宇微光学软件有限公司 | Curve polygon image scaling method, system and electronic equipment |
-
2022
- 2022-06-22 CN CN202210710971.XA patent/CN115272362A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117333373A (en) * | 2023-12-01 | 2024-01-02 | 武汉宇微光学软件有限公司 | Curve polygon image scaling method, system and electronic equipment |
| CN117333373B (en) * | 2023-12-01 | 2024-02-23 | 武汉宇微光学软件有限公司 | Curve polygon image scaling method, system and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107578035B (en) | Human body contour extraction method based on super-pixel-multi-color space | |
| KR101795823B1 (en) | Text enhancement of a textual image undergoing optical character recognition | |
| Gatos et al. | ICDAR 2009 document image binarization contest (DIBCO 2009) | |
| CN108629343B (en) | License plate positioning method and system based on edge detection and improved Harris corner detection | |
| CN111260616A (en) | Insulator crack detection method based on Canny operator two-dimensional threshold segmentation optimization | |
| JP2010525486A (en) | Image segmentation and image enhancement | |
| CN109241973B (en) | Full-automatic soft segmentation method for characters under texture background | |
| CN110309806B (en) | Gesture recognition system and method based on video image processing | |
| CN105701491A (en) | Method for making fixed-format document image template and application thereof | |
| CN110930321A (en) | Blue/green screen digital image matting method capable of automatically selecting target area | |
| CN112529853A (en) | Method and device for detecting damage of netting of underwater aquaculture net cage | |
| EP3510526B1 (en) | Particle boundary identification | |
| CN115272362A (en) | Method and device for segmenting effective area of digital pathology full-field image | |
| CN112381084B (en) | Automatic contour recognition method for tomographic image | |
| CN115995078A (en) | Image preprocessing method and system for plankton in-situ observation | |
| CN111445402B (en) | Image denoising method and device | |
| CN110766614B (en) | Image preprocessing method and system of wireless scanning pen | |
| Tabatabaei et al. | A novel method for binarization of badly illuminated document images | |
| CN108205678A (en) | A kind of nameplate Text region processing method containing speck interference | |
| CN110930358A (en) | Solar panel image processing method based on self-adaptive algorithm | |
| CN113643290B (en) | Straw counting method and device based on image processing and storage medium | |
| CN116363097A (en) | Defect detection method and system for photovoltaic panel | |
| CN112070771B (en) | Adaptive threshold segmentation method and device based on HS channel and storage medium | |
| CN114463352A (en) | Slide scanning image target segmentation and extraction method and system | |
| CN110298816B (en) | Bridge crack detection method based on image regeneration |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |