CN113570628B - White blood cell segmentation method based on movable contour model - Google Patents
White blood cell segmentation method based on movable contour model Download PDFInfo
- Publication number
- CN113570628B CN113570628B CN202110871648.6A CN202110871648A CN113570628B CN 113570628 B CN113570628 B CN 113570628B CN 202110871648 A CN202110871648 A CN 202110871648A CN 113570628 B CN113570628 B CN 113570628B
- Authority
- CN
- China
- Prior art keywords
- image
- white blood
- white
- local
- pixel
- 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.)
- Active
Links
- 210000000265 leukocyte Anatomy 0.000 title claims abstract description 78
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000011218 segmentation Effects 0.000 title claims abstract description 36
- 210000003855 cell nucleus Anatomy 0.000 claims abstract description 45
- 210000004940 nucleus Anatomy 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 20
- 230000000877 morphologic effect Effects 0.000 claims abstract description 15
- 238000003672 processing method Methods 0.000 claims abstract description 8
- 238000009499 grossing Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 7
- 238000005260 corrosion Methods 0.000 claims description 4
- 230000007797 corrosion Effects 0.000 claims description 4
- 210000004027 cell Anatomy 0.000 abstract description 11
- 230000001086 cytosolic effect Effects 0.000 abstract description 2
- 210000000805 cytoplasm Anatomy 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 3
- 238000004043 dyeing Methods 0.000 description 3
- 210000003743 erythrocyte Anatomy 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 210000005259 peripheral blood Anatomy 0.000 description 2
- 239000011886 peripheral blood Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000001772 blood platelet Anatomy 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004976 peripheral blood cell Anatomy 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000010186 staining Methods 0.000 description 1
- 238000007447 staining method Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30024—Cell structures in vitro; Tissue sections in vitro
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a white blood cell segmentation method based on an active contour model, which comprises the following steps: step 1: obtaining the number of all pixel values in each small cube, assigning all pixel values in each small cube to predetermined pixel values, and obtaining a preprocessed image, wherein the step 2 is as follows: performing density clustering on the preprocessed image by using a density peak clustering segmentation method to obtain an initial white blood cell image, and then performing smooth processing on the white blood cell nucleus edge outline in the initial white blood cell image by using a morphological processing method to obtain a final white blood cell nucleus image, wherein the step 3 is as follows: processing the final white cell nucleus image by using a level set method to obtain a white cell nucleus contour curve, and taking the white cell nucleus contour curve as an initial contour curve of a movable contour model to obtain a final white cell image; according to the local gray level characteristic of the white blood cells, the invention evolves to the cytoplasmic edge to divide the white blood cells by using the cell nucleus contour line, so as to obtain a more accurate result.
Description
Technical Field
The invention belongs to the field of medical image processing, and particularly relates to a white blood cell segmentation method based on a movable contour model.
Background
Medical image segmentation is one of the main research directions in the current image processing field due to its great contribution to the fields of auxiliary medical treatment, clinical medicine and the like. For a specific medical image, such as a white blood cell image in a peripheral blood cell image. The change in the number of white blood cells can be used as a judgment as to whether or not the human body is infected. However, due to the characteristics of white blood cells or the influence of devices, the nuclei and cytoplasm of white blood cells are inevitably influenced by uneven gray levels and noise in the dyeing and image acquisition stages, and although segmentation algorithms are continuously proposed, the precision and the speed of a single traditional segmentation algorithm are still not ideal.
The peripheral blood image contains a large number of red blood cells, platelets and other substances in addition to various white blood cells. White blood cells are usually marked well after a standard staining method is adopted, but the nuclei of the white blood cells are deeply stained due to the characteristics of the white blood cells, and the cytoplasm part is in dark purple and has overlarge light purple difference. Meanwhile, due to the problems of interference of red blood cells, dyeing leakage and the like in the background, most segmentation algorithms have excellent segmentation performance on white blood cell nuclei, but have unsatisfactory segmentation effect on white blood cell cytoplasm.
In general, the recognition and classification system of white blood cells includes the following aspects: preprocessing an image, segmenting white blood cells, extracting features and classifying the features. Segmentation is used as a link before classification tasks and is important to classification accuracy. The main segmentation methods at present have two aspects: the first is a segmentation method based on a graph theory, mainly comprising a threshold method, a clustering method and the like; the second is a segmentation method based on variation theory, which mainly comprises a movable contour method. The segmentation method based on the graph theory does not exist at present, and the problem of uneven gray level of the white blood cell image under various conditions can be effectively solved, but the segmentation method based on the variation theory only causes that the evolution curve of the movable contour is difficult to fit the edge of the cytoplasm or can be converged at the edge of the nucleus due to the huge gray level difference of the nucleus and the cytoplasm of the white blood cell, so that the white blood cell is difficult to segment.
Disclosure of Invention
The invention aims to provide a leukocyte segmentation method based on an active contour model, which can accurately segment leukocyte nuclei and cytoplasm.
The invention adopts the following technical scheme: a white blood cell segmentation method based on an active contour model comprises the following steps:
step 1: dividing RGB pixel space of an original stained leukocyte image into small cubes with equal length, calculating pixel values at central points of all the small cubes, namely preset pixel values, obtaining the number of all the pixel values in each small cube, assigning all the pixel values in each small cube to be preset pixel values, obtaining a preprocessed image,
step 2: performing density clustering on the preprocessed image by using a density peak clustering segmentation method to obtain an initial white blood cell image, then performing smooth processing on the white blood cell nucleus edge outline in the initial white blood cell image by using a morphological processing method to obtain a final white blood cell nucleus image,
step 3: and processing the final white cell nucleus image by using a level set method to obtain a white cell nucleus contour curve, taking the white cell nucleus contour curve as an initial contour curve of the movable contour model, and constructing a local gray information expression by using local JS divergence as an edge stopping function, so that the initial contour curve is gradually calculated to obtain the white cell edge, thereby obtaining the final white cell image.
Further, the edge stopping function in step 3 is:
wherein I (x) represents a local image region, C 1 And C 2 The average gray values of the inner part and the outer part divided by the contour curve in the local area are respectively represented by p and q, the gray distribution of the inner part and the outer part divided by the contour curve is respectively represented by JS (p, q), the α is a local threshold value, 0.02 is taken to measure the uniformity of the local gray, the α -JS (p, q) is more than 0, the gray is uniform, the gray is non-uniform, and the local window is 5*5.
Further, step 2 consists of the steps of:
step 21: firstly, calculating local density rho and relative distance delta of the preprocessed image pixel by pixel, and determining a cut-off distance d c ；
Wherein, the local density ρ of the density cluster is:
in the above, d i，j Representing the distance between pixel i and pixel j, d using Euclidean distance computation cutoff Representing the cutoff distance, i.e. the effective density radius, which is the only parameter in the density peak cluster, taking 0.5% of the maximum relative distance, ρ i The distribution of the pixel points i around the pixel points, namely the local density, can be represented;
wherein, the relative distance delta of the density clusters is:
step 22: the products of the local densities rho of all the pixel points and the relative distances delta are arranged in a descending order, the first N points are taken as clustering center points, the N value of the clustering center is taken as 2, the clustering structure is divided into two types, one type is a background part, and the other type is a foreground white cell nucleus;
step 23: classifying and dividing the rest points, namely the pixel points except N clustering center points, according to the clustering centers to obtain clustered images, namely white cell nucleus images, binarizing the clustered white cell nucleus images, taking white cell nuclei as a foreground and the rest as a background, and eliminating a tiny area by using a morphological processing method and morphological expansion and corrosion operation; and (3) processing the foreground part to obtain the edge profile of the white cell nucleus, and then carrying out smoothing treatment on the edge profile of the white cell nucleus, namely smoothing the edge of the image by using Gaussian filtering, and taking 1.8 by Sigma to obtain a final white cell nucleus image.
The beneficial effects of the invention are as follows: according to the invention, a white blood cell image is segmented by a clustering method combining a graph theory and a movable contour method combining a difference theory, a rough contour of white blood cell nuclei is obtained by density clustering, and meanwhile, according to the local gray level characteristics of white blood cells, white blood cells are segmented by evolving a contour line of the white blood cells to the cytoplasmic edges, so that a more accurate result is obtained.
Drawings
FIG. 1 shows the results before and after the processing of step 1 of the stained leukocyte image in example 1 of the present invention; FIG. 1a is a pre-processed image, FIG. 1b is a post-processed image, and FIG. 1c is a post-processed image;
FIG. 2 shows the results before and after the density peak clustering of the leukocyte image in example 1, FIG. 2a shows the leukocyte image obtained after density clustering, and FIG. 2b shows the leukocyte image obtained after binarization, morphological treatment and smoothing;
fig. 3 is a result of fine segmentation of a leukocyte image by a moving contour method in example 1 of the present invention, fig. 3a is an initial contour line image of the initialization of a nuclear contour, fig. 3b is a curve evolution result, and fig. 3c is a final segmentation result.
Detailed Description
The invention will be described in detail below with reference to the drawings and the detailed description.
The invention discloses a white blood cell segmentation method based on an active contour model, which comprises the following steps:
step 1: dividing the RGB pixel space of the original stained leucocyte image into small cubes with equal length, calculating the pixel value at the center point of all the small cubes, namely a preset pixel value, obtaining the number of all the pixel values in each small cube, assigning all the pixel values in each small cube to be the preset pixel value, and obtaining the preprocessed image.
Because the density clustering processing is carried out on the original white cell image, the huge data processing problem is faced, so that the white cell image is preprocessed in a mini-RGBCUBE mode, and the number of pixels is reduced.
Step 2: and performing density clustering on the preprocessed image by using a density peak clustering segmentation method to obtain an initial white blood cell image, and then performing smooth processing on the white blood cell nucleus edge outline in the initial white blood cell image by using a morphological processing method to obtain a final white blood cell nucleus image.
Besides the density peak clustering segmentation method in the step 2, other segmentation methods, such as K-Means, C-Means and the like, can be adopted, but the color consistency of the segmented small areas and the time complexity of the algorithm are comprehensively considered, and the performance of the density peak clustering algorithm is relatively optimal.
Performing density clustering on the preprocessed color images in the step 1, binarizing a clustering result by adopting Euclidean distance measurement, wherein white cell nuclei are used as a foreground, and the rest are used as a background; and then, smoothing the edge outline of the white cell nucleus in the initial white cell image by using a morphological treatment method to obtain a final white cell nucleus image.
Wherein, the step 2 consists of the following steps:
step 21: firstly, calculating local density rho and relative distance delta of the preprocessed image pixel by pixel, and determining a cut-off distance d c 。
Wherein, the local density ρ of the density cluster is:
in the above, d i,j The distance between the pixel point i and the pixel point j is represented, and euclidean distance calculation is used. d, d cutoff The cutoff distance, i.e., the effective density radius, is expressed as the only parameter in the density peak cluster, taking 0.5% of the maximum relative distance. Then ρ is i The distribution of pixel points i around the pixel point, i.e. the local density, can be represented.
Wherein, the relative distance delta of the density clusters is:
step 22: the products of the local densities ρ of all the pixel points and the relative distances δ are arranged in a descending order, the first N points are taken as clustering center points, the N value of the clustering center is taken as 2, the clustering structure is divided into two types, one type is a background part, and the other type is a foreground white cell nucleus.
Step 23: classifying and dividing the rest points, namely the pixel points except N clustering center points, according to the clustering centers to obtain clustered images, namely white cell nucleus images, binarizing the clustered white cell nucleus images, taking white cell nuclei as a foreground and the rest as a background, and eliminating a tiny area by using a morphological processing method and morphological expansion and corrosion operation; and (3) processing the foreground part to obtain the edge profile of the white cell nucleus, and then carrying out smoothing treatment on the edge profile of the white cell nucleus, namely smoothing the edge of the image by using Gaussian filtering, and taking 1.8 by Sigma to obtain a final white cell nucleus image.
Step 3: and processing the final white cell nucleus image by using a level set method to obtain a cell nucleus contour curve, taking the cell nucleus contour curve as an initial contour curve of the movable contour model, and constructing a local gray information expression by using local JS divergence as an edge stopping function, so that the initial contour curve is gradually calculated to obtain the white cell edge, thereby obtaining the final white cell image.
Wherein, the step 3 consists of the following steps:
step 31: and (3) processing the final white cell nucleus image by using a level set method to obtain a cell nucleus contour curve, taking the cell nucleus contour curve as an initial contour line, processing the white cell nucleus edge contour obtained in the step (24) by using the level set method, setting the white cell nucleus part as 1, and setting the rest as 0.
Step 32: constructing an edge stopping function driven by the local JS divergence;
the edge stop function is:
wherein I (x) represents a partial image region, C1 and C 2 The average gray values of the partial region divided into an inner part and an outer part by the contour curve are respectively obtained. JS (p||q) represents JS divergence information between gray level distribution of the inner part and the outer part, alpha is a local threshold value, and 0.02 is taken to measure uniformity of local gray level.
Step 33: and (3) utilizing an edge stopping function to evolve a movable contour curve according to the local gray level change condition of the white blood cells so as to obtain a final white blood cell image.
According to the invention, the movable contour curve evolution is utilized to carry out white blood cell fine segmentation, the cell nucleus contour curve is used as an initial contour curve of the movable contour model, and a local gray information expression is constructed by using local JS divergence as an edge stop function to guide the initial contour curve to evolve to the white blood cell edge, so that a white blood cell accurate segmentation image is obtained.
The invention firstly carries out pretreatment on an original white blood cell image, then carries out rough segmentation on cell nuclei, and carries out accurate segmentation of a movable contour model of curve evolution by utilizing rough segmentation contours. On the one hand, dividing the white blood cells according to the problem of uneven cytoplasm and nucleus staining of the white blood cell image by a two-step method; on the other hand, a variable active contour model segmentation method with a solid mathematical theory basis is adopted, the mathematical theory is effectively combined with an actual environment, an edge stop function of the active contour model is obtained, more accurate segmentation is carried out on white blood cells with uneven gray scale, the dyeing characteristics of white blood cell nuclei in a peripheral blood image are considered, the white blood cell nuclei are obtained in a clustering mode, and the contour curve is evolved by adopting the edge stop function according to the characteristics of uneven gray scale of the white blood cell cytoplasm, so that the more accurate white blood cell segmentation method is realized in two aspects.
Example 1
Step 1: the input original stained white blood cell color image is preprocessed in an RGB-CUBE mode as shown in fig. 1a, the number of local pixels is reduced, the RGB pixel space (256 x 256) is divided into small CUBEs of each side length 16, side length lenmini tube=16 is defined.
Calculating pixel values at the center point of all the microcubes, namely, preset pixel values, obtaining the number npixel of all the pixel values in each microcubes, assigning all the pixel values in each microcubes to the preset pixel values, and obtaining a preprocessed image, so that the number of pixels of the original image is reduced.
Selecting the length of the small cube to be 2 n In =16 (n=4), the actual n can be (0-9), but as the n value is not increased, after the pixel points in the small cubes are reduced to the center of each small cube, the whole pixel of the image is reduced sharply, so that the original image is indistinguishable, when n=0, the length of the small cube is 1, and at the moment, the central pixel is the pixel of the small cube, which is equivalent to unchanged image; when n=9, the small cube length is 256, where the center pixel is (127, 127, 127), which corresponds to changing the entire image to a gray value, as shown in fig. 1b. Through experiments, when n=4 is selected, efficiency and accuracy can be considered, and meanwhile, the calculation method of each small cube center point pixel is as follows:
wherein,representing a down-rounded symbol, and pixels representing the pixel values of the original image.
The rgbacube method is used to reduce the number of original pixels, facilitating the subsequent clustering process, whereby a preprocessed image is obtained, as in fig. 1c.
Step 2: and (3) performing density clustering on the preprocessed images in the step (1), binarizing a clustering result by adopting Euclidean distance measurement, wherein white nuclei are used as a foreground (white), and the rest is used as a background (black). And then, a morphological processing method is used for processing the foreground part, and a connected domain marking mode is adopted for carrying out smooth processing on the edge outline of the white cell nucleus in the initial white cell image so as to obtain a final white cell nucleus image.
The pixel number npixel in each small cube and the pixel value centroids of the center points of the small cubes are obtained for the preprocessed image in the step 1, and then the local density ρ of each central pixel value centroids is calculated, wherein the local density ρ of the density clusters is as follows:
in the above, d i，j The distance between the pixel point i and the pixel point j is represented, and euclidean distance calculation is used. d, d cutoff The cutoff distance, i.e., the effective density radius, is expressed as the only parameter in the density peak cluster, taking 0.5% of the maximum relative distance. Proved by experimental tests: d, d c Better clustering results can be obtained by taking 0.5% of the maximum relative distance; then ρ is i The distribution of pixel points i around the pixel point, i.e. the local density, can be represented.
Wherein, the relative distance delta of the density clusters is:
after the local density and the relative distance of each pixel point centroids are calculated, products of the local density and the relative distance are arranged in a descending order, the point with the front N of 2 is taken as a clustering center, and the number N of the clustering centers is selected to obtain a white cell nucleus part, so that the number of the selected clusters is 2.
After finding the clustering center, the rest pixel points are subjected to expanded clustering, and are distributed to the clusters of the data points which are closest to the clustering center and have local densities larger than the clustering center, and finally, a density peak clustering result is obtained, as shown in fig. 2a. And eliminating the tiny areas of the clustered results by using morphological expansion and corrosion operation, wherein the nuclear size is 5X5. The morphological processed image is subjected to Gaussian smoothing filtering, edge smoothing and smoothing operation can also adopt median filtering, mean filtering and the like, but the overall gray level distribution characteristics of the image can be more reserved when the image is smoothed by adopting Gaussian filtering, and a final white cell nuclear image is obtained, as shown in fig. 2b.
Step 3: accurate segmentation of white blood cells using active contour algorithm
Taking the profile of the white blood cell nucleus as an initial level set, as shown in fig. 3a, the final white blood cell nucleus image obtained in step 2 is set to 1 in the cell nucleus part and 0 in the rest part, and then the final white blood cell nucleus image is used as a function of the initial level set.
Calculating a gray average value C of a local area divided into an inner part and an outer part by a contour curve 1 And C 2 ：
Wherein I is a partial image, H (phi) is a smooth form of a unit step function, and is used for calculating the inner and outer parts of a partial area, specifically:
where phi is a contour curve, eps is a very small constant, and eps=1e-3 is taken.
Constructing a local Jensen-Shannon divergence driven edge stop function by using the local gray image mean value:
wherein I (x) represents a local image region, C 1 And C 2 The average gray values of the inner part and the outer part divided by the contour curve in the local area are respectively represented by p and q, the gray distribution of the inner part and the outer part divided by the contour curve is respectively represented by JS (p, q), the α is a local threshold value, 0.02 is taken to measure the uniformity of the local gray, the α -JS (p, q) is more than 0, the gray is uniform, the gray is non-uniform, and the local window is 5*5.
Guiding the contour curve evolution by using the constructed edge stop function, assuming that the gray density of the target area in the image is greater than that of the background area (and vice versa), when JS (p, q) > alpha, namely alpha-JS (p, q) < 0, g jspf < 0, meanThe contour curves move towards the areas with large gray level changes and are finally positioned at the edges of the images; when JS (p, q) < alpha, i.e. alpha-JS (p, q) > 0, g jspf By > 0 is meant that the evolution curve is in a gray level plateau in the image, with the gray values of the individual pixels not being significantly different. Continuous detection g as profile curve evolves jspf The contour curve is continuously approaching the target edge, eventually stopping at the target boundary.
By g jspf The continuous change of the sign, the contour curve finally converges at the cell edge, at this time, the sign inside the contour curve is negative, and the sign outside the contour curve is positive, thus achieving the purpose of precisely dividing the white blood cells, as shown in fig. 3b and 3c.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (2)
1. A white blood cell segmentation method based on an active contour model is characterized by comprising the following steps:
step 1: dividing RGB pixel space of an original stained leukocyte image into small cubes with equal length, calculating pixel values at central points of all the small cubes, namely preset pixel values, obtaining the number of all the pixel values in each small cube, assigning all the pixel values in each small cube to be preset pixel values, obtaining a preprocessed image,
step 2: performing density clustering on the preprocessed image by using a density peak clustering segmentation method to obtain an initial white blood cell image, then performing smooth processing on the white blood cell nucleus edge outline in the initial white blood cell image by using a morphological processing method to obtain a final white blood cell nucleus image,
step 3: processing the final leukocyte nucleus image by using a level set method to obtain a leukocyte nucleus contour curve, taking the leukocyte nucleus contour curve as an initial contour curve of the movable contour model, and constructing a local gray information expression by using local JS divergence as an edge stopping function, so that the initial contour curve is gradually calculated to obtain the leukocyte edge, thereby obtaining the final leukocyte image;
wherein, the step 2 consists of the following steps:
step 21: firstly, calculating local density rho and relative distance delta of the preprocessed image pixel by pixel, and determining a cut-off distance d c ；
Wherein, the local density ρ of the density cluster is:
in the above, d i,j Representing the distance between pixel i and pixel j, d using Euclidean distance computation cutoff Representing the cutoff distance, i.e. the effective density radius, which is the only parameter in the density peak cluster, taking 0.5% of the maximum relative distance, ρ i The distribution of the pixel points i around the pixel points, namely the local density, can be represented;
wherein, the relative distance delta of the density clusters is:
step 22: the products of the local densities rho of all the pixel points and the relative distances delta are arranged in a descending order, the first N points are taken as clustering center points, the N value of the clustering center is taken as 2, the clustering structure is divided into two types, one type is a background part, and the other type is a foreground white cell nucleus;
step 23: classifying and dividing the rest points, namely the pixel points except N clustering center points, according to the clustering centers to obtain clustered images, namely white cell nucleus images, binarizing the clustered white cell nucleus images, taking white cell nuclei as a foreground and the rest as a background, and eliminating a tiny area by using a morphological processing method and morphological expansion and corrosion operation; and (3) processing the foreground part to obtain the edge profile of the white cell nucleus, and then carrying out smoothing treatment on the edge profile of the white cell nucleus, namely smoothing the edge of the image by using Gaussian filtering, and taking 1.8 by Sigma to obtain a final white cell nucleus image.
2. The method of claim 1, wherein the edge stop function in step 3 is:
wherein I (x) represents a local image area, C1 and C2 are average gray values of an inner part and an outer part divided into an inner part and an outer part by a contour curve in the local area respectively, p and q represent gray distribution of the inner part and the outer part divided into the inner part and the outer part by the contour curve respectively, JS (p, q) represents JS divergence information between the gray distribution of the inner part and the outer part, α is a local threshold value, 0.02 is taken to measure uniformity of local gray, α -JS (p, q) is greater than 0, the uniformity of gray is represented, otherwise, the non-uniformity of gray is represented, and the local window is selected as 5*5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110871648.6A CN113570628B (en) | 2021-07-30 | 2021-07-30 | White blood cell segmentation method based on movable contour model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110871648.6A CN113570628B (en) | 2021-07-30 | 2021-07-30 | White blood cell segmentation method based on movable contour model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113570628A CN113570628A (en) | 2021-10-29 |
| CN113570628B true CN113570628B (en) | 2024-04-02 |
Family
ID=78169445
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110871648.6A Active CN113570628B (en) | 2021-07-30 | 2021-07-30 | White blood cell segmentation method based on movable contour model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113570628B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115620075B (en) * | 2022-12-16 | 2023-03-14 | 南昌大学 | Method, system and equipment for generating data set for leukocyte classification model |
| CN116433701B (en) * | 2023-06-15 | 2023-10-10 | 武汉中观自动化科技有限公司 | Workpiece hole profile extraction method, device, equipment and storage medium |
| CN117237646B (en) * | 2023-11-15 | 2024-01-30 | 深圳市润海电子有限公司 | PET high-temperature flame-retardant adhesive tape flaw extraction method and system based on image segmentation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473739A (en) * | 2013-08-15 | 2013-12-25 | 华中科技大学 | White blood cell image accurate segmentation method and system based on support vector machine |
| WO2016091016A1 (en) * | 2014-12-12 | 2016-06-16 | 山东大学 | Nucleus marker watershed transformation-based method for splitting adhered white blood cells |
| CN110288582A (en) * | 2019-06-26 | 2019-09-27 | 电子科技大学 | A kind of accurate dividing method of nucleus |
| WO2019223706A1 (en) * | 2018-05-22 | 2019-11-28 | 杭州智微信息科技有限公司 | Saturation clustering-based method for positioning bone marrow white blood cells |
-
2021
- 2021-07-30 CN CN202110871648.6A patent/CN113570628B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473739A (en) * | 2013-08-15 | 2013-12-25 | 华中科技大学 | White blood cell image accurate segmentation method and system based on support vector machine |
| WO2016091016A1 (en) * | 2014-12-12 | 2016-06-16 | 山东大学 | Nucleus marker watershed transformation-based method for splitting adhered white blood cells |
| WO2019223706A1 (en) * | 2018-05-22 | 2019-11-28 | 杭州智微信息科技有限公司 | Saturation clustering-based method for positioning bone marrow white blood cells |
| CN110288582A (en) * | 2019-06-26 | 2019-09-27 | 电子科技大学 | A kind of accurate dividing method of nucleus |
Non-Patent Citations (2)
| Title |
|---|
| 利用单类支持向量机分割血细胞图像;潘晨, 闫相国, 郑崇勋, 梁成文;西安交通大学学报(02);全文 * |
| 基于活动轮廓模型的彩色白细胞图像自动分割方法研究;蔡隽;鲍旭东;吴磊;罗立民;;生物医学工程研究(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113570628A (en) | 2021-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113570628B (en) | White blood cell segmentation method based on movable contour model | |
| CN106803247B (en) | Microangioma image identification method based on multistage screening convolutional neural network | |
| CN109522908B (en) | Image significance detection method based on region label fusion | |
| CN109166095B (en) | Fundus image cup and disc segmentation method based on generation countermeasure mechanism | |
| CN111340824B (en) | Image feature segmentation method based on data mining | |
| Miao et al. | Simultaneous segmentation of leukocyte and erythrocyte in microscopic images using a marker-controlled watershed algorithm | |
| CN114067114B (en) | Area nuclear segmentation counting method based on algae cell morphology | |
| Wang et al. | Nucleus segmentation of cervical cytology images based on depth information | |
| Alasadi et al. | Early detection and classification of melanoma skin cancer | |
| Pandit et al. | Survey on automatic rbc detection and counting | |
| CN111126162A (en) | Method, device and storage medium for identifying inflammatory cells in image | |
| El-Khamy et al. | An efficient brain mass detection with adaptive clustered based fuzzy C-mean and thresholding | |
| Singh et al. | An efficient combined approach for medical brain tumour segmentation | |
| CN114494318B (en) | Cornea contour extraction method based on cornea dynamic deformation video of Ojin algorithm | |
| CN110473224B (en) | Automatic RSF level set image segmentation method based on KL entropy | |
| Somasundaram et al. | Automatic segmentation of nuclei from pap smear cell images: A step toward cervical cancer screening | |
| CN110136139B (en) | Dental nerve segmentation method in facial CT image based on shape feature | |
| CN107977976A (en) | A kind of image partition method | |
| Gunawan et al. | Fuzzy Region Merging Using Fuzzy Similarity Measurement on Image Segmentation | |
| Saha et al. | Segmentation of cervical nuclei using SLIC and pairwise regional contrast | |
| CN111429461B (en) | Novel segmentation method for overlapped and exfoliated epithelial cells | |
| Zhang et al. | A Morphological Image Segmentation Algorithm for Circular Overlapping Cells. | |
| Zhao et al. | An automatic counting method of maize ear grain based on image processing | |
| CN113850792A (en) | Cell classification counting method and system based on computer vision | |
| Taouil et al. | Automatic segmentation and classification of skin lesion images |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |