WO2023019555A1

WO2023019555A1 - Cell fluorescence image thresholding method and system, terminal, and storage medium

Info

Publication number: WO2023019555A1
Application number: PCT/CN2021/113795
Authority: WO
Inventors: 吴昊; 魏彦杰; 潘毅
Original assignee: 深圳先进技术研究院; 中国科学院深圳理工大学(筹)
Priority date: 2021-08-20
Filing date: 2021-08-20
Publication date: 2023-02-23

Abstract

The present application relates to a cell fluorescence image thresholding method and system, a terminal, and a storage medium. The method comprises: adjusting pixel value distribution of a cell fluorescence image by using a contrast limited adaptive histogram equalization (CLAHE) algorithm; performing bilateral filtering on the cell fluorescence image that has undergone pixel value distribution adjustment; using Otsu's method to perform binarization processing on the cell fluorescence image that has undergone bilateral filtering to obtain a binarized image; and obtaining communicated domains in the binarized image, calculating the area of each communicated domain, and eliminating communicated domains the area of which is smaller than a set threshold, so as to obtain a thresholding cell fluorescence image. According to the embodiments of the present application, the interference of point light source imaging on the fluorescence image is effectively removed, boundary information of a cell is retained while the effect of a point light source is removed, thus increasing the data volume of a cell to be analyzed, and helping to improve the accuracy of subsequently extracting biological parameters.

Description

Cell fluorescence image thresholding method, system, terminal and storage medium

technical field

The application belongs to the technical field of biomedical image processing, and in particular relates to a method, system, terminal and storage medium for thresholding a cell fluorescence image.

Background technique

Taking fluorescence images at the same time as cell experiments can greatly improve the efficiency of cell observation and analysis. As a kind of cell labeling, the detection and tracking of cells can be automated through the processing of fluorescent images, which greatly improves the efficiency of experiments. At the same time, a series of parameters with biological significance can be obtained through the analysis of cell markers in the fluorescence images, such as cell area, nucleus-to-plasma ratio, movement speed, direction, etc. These parameters play an important role in cell behavior analysis and classification identification. Therefore, the shooting of cell fluorescence images has an important role and significance. However, since the light source used by the experimental platform is a point light source when taking fluorescence images, the distribution of pixel values in the center of the fluorescence image is much larger than that in the surrounding areas, and the conventional threshold method cannot preserve the central and surrounding cell markers at the same time. The method adopted by experimentalists is to crop the fluorescence image, and only keep a small part of the image in the center of the field of view for thresholding operation, but the edge information of cell markers is not well preserved during the thresholding process. This makes the finally obtained biological parameters have certain errors. At the same time, the cutting operation greatly reduces the number of available cell markers. In order to ensure a certain amount of data, further experimental operations need to be added, which increases the complexity of the operation.

Contents of the invention

The present application provides a cell fluorescence image thresholding method, system, terminal and storage medium, aiming to solve one of the above-mentioned technical problems in the prior art at least to a certain extent.

In order to solve the above problems, the application provides the following technical solutions:

A cell fluorescence image thresholding method, comprising:

Adjust the pixel value distribution of the cell fluorescence image by using the limited contrast adaptive local histogram equalization algorithm;

performing bilateral filter processing on the cell fluorescence image after the pixel value distribution adjustment;

performing binarization processing on the cell fluorescence image after the bilateral filter processing by using the Otsu method to obtain a binarized image;

The connected domains in the binarized image are obtained, and the area of each connected domain is calculated, and the connected domains whose area is smaller than a set threshold are eliminated to obtain a thresholded cell fluorescence image.

The technical solution adopted in the embodiment of the present application also includes: the adjustment of the pixel value distribution of the cell fluorescence image by using the limited contrast adaptive local histogram equalization algorithm is specifically:

Divide the cell fluorescence image into non-overlapping sub-blocks of equal size, the size of each sub-block is tileGridSize;

Calculate the complete histogram of the cell fluorescence image and the histogram of each sub-block, and determine the clipping threshold clipLimit of the histogram;

Evenly fill the pixel values exceeding the clipping threshold clipLimit in the histogram of each sub-block into the complete histogram, and complete the redistribution of redundant pixels in the histogram of the sub-block:

Perform pixel value equalization processing on the complete histogram after pixel filling through function mapping;

Bilinear interpolation is used to reconstruct the pixel gray value of the complete histogram after pixel value equalization processing, and the processed cell fluorescence image is obtained.

The technical solution adopted in the embodiment of the present application further includes: performing bilateral filtering on the cell fluorescence image after the pixel value distribution adjustment is specifically:

Let G _σ (x) be a two-dimensional Gaussian kernel,

G _σs , G _σr are spatial weight and gray weight respectively, I _p is the pixel value of the position to be calculated, I _q is the pixel value in the neighborhood, p=(p _x , p _y ) is the position of the pixel value to be calculated ,q=(q _x ,q _y ) is the pixel value position in the neighborhood, ||pq|| is the spatial distance between p and q, then:

Among them, BF represents the bilateral filter, BF[I] _p represents the pixel value at position p in the cell fluorescence image calculated using the bilateral filter, and q∈S represents the pixel value at position p calculated using the bilateral filter, and the result is determined by the neighborhood The weighted sum of each pixel value in the range S is determined. The weight of each pixel value in the neighborhood is determined by the spatial weight G _σs and the gray weight G _σr ; W _p is the normalization factor, and the calculation formula of W _p is :

The technical solution adopted in the embodiment of the present application also includes: the binarization processing of the cell fluorescence image processed by the bilateral filter using the Otsu method is specifically:

All pixels of the cell fluorescence image are divided into C1 and C2 by the threshold TH, C1 is a pixel class smaller than TH, and C2 is a pixel class larger than TH, assuming that the pixel averages of C1 and C2 are m1 and m2 respectively, the global pixel of the image The mean is mG, and the probabilities of pixels being classified into C1 and C2 classes are p1 and p2 respectively, then:

p1*m1+p2*m2=mG

p1+p2=1

The expression for the between-class variance is:

σ ² =p1(m1-mG) ² +p2(m2-mG) ²

The threshold that maximizes the value of σ ² is the maximum inter-class variance threshold, and the cell fluorescence image is binarized through the maximum inter-class variance threshold to separate cells and background regions.

The technical solution adopted in the embodiment of the present application also includes: the calculation of the area of each connected domain is specifically:

All the pixel values of each connected domain are accumulated, and the total number of pixel values obtained is regarded as the area of the connected domain.

Another technical solution adopted in the embodiment of the present application is: a cell fluorescence image thresholding system, comprising:

Pixel value adjustment module: used to adjust the pixel value distribution of the cell fluorescence image by using the limited contrast adaptive local histogram equalization algorithm;

Image filtering module: for performing bilateral filtering processing on the cell fluorescence image after the pixel value distribution adjustment;

Binarization module: for performing binarization processing on the cell fluorescence image processed by the bilateral filter by using the Otsu method to obtain a binarized image;

Noise elimination module: used to obtain the connected domains in the binarized image, calculate the area of each connected domain, eliminate the connected domains whose area is smaller than the set threshold, and obtain the thresholded cell fluorescence image.

Another technical solution adopted by the embodiment of the present application is: a terminal, the terminal includes a processor and a memory coupled to the processor, wherein,

The memory stores program instructions for implementing the method for thresholding cell fluorescence images;

The processor is configured to execute the program instructions stored in the memory to control the thresholding of the cell fluorescence image.

Another technical solution adopted in the embodiment of the present application is: a storage medium storing program instructions executable by a processor, the program instructions being used to execute the method for thresholding cell fluorescence images.

Compared with the prior art, the beneficial effect produced by the embodiment of the present application is that the thresholding method, system, terminal and storage medium of the embodiment of the present application use the CLAHE algorithm to adjust the pixel value distribution of the cell fluorescence image, and after CLAHE processing Bilateral filtering was performed on the cytofluorescence image, and the OTSU algorithm was used to binarize the filtered cytofluorescence image, and the residual noise in the image was removed according to the area of the connected region to obtain the final thresholded cytofluorescence image. The embodiment of the present application effectively removes the interference of the point light source imaging on the fluorescence image, and retains the boundary information of the cells while removing the influence of the point light source, which increases the data volume of the cells to be analyzed, and is conducive to improving the accuracy of subsequent extraction of biological parameters.

Description of drawings

Fig. 1 is a flow chart of the cell fluorescence image thresholding method of the embodiment of the present application;

FIG. 2 is a schematic diagram of pixel value equalization processing in an embodiment of the present application;

FIG. 3 is a schematic structural diagram of a cell fluorescence image thresholding system according to an embodiment of the present application;

FIG. 4 is a schematic structural diagram of a terminal according to an embodiment of the present application;

FIG. 5 is a schematic structural diagram of a storage medium according to an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, not to limit the present application.

Please refer to FIG. 1 , which is a flowchart of a method for thresholding a cell fluorescence image according to an embodiment of the present application. The cell fluorescence image thresholding method of the embodiment of the present application includes the following steps:

S10: acquiring cell fluorescence images;

In this step, the cell fluorescence image is captured by a cell microscopic imaging platform.

S20: Use the CLAHE (Contrast Limited Adaptive histgram equalization) algorithm to suppress noise and enhance the contrast of the cell fluorescence image, and adjust the pixel value distribution of the cell fluorescence image;

In this step, the CLAHE algorithm limits the enhancement range of the local contrast by limiting the height of the local histogram, which can enhance the contrast while suppressing the noise, thereby limiting the noise amplification and local contrast enhancement of the cell fluorescence image, and realizing the brightness of the cell fluorescence image Redistribution, to improve the problem of the pixel value distribution of the cell fluorescence image caused by the point light source being large in the middle and small around the four sides, and eliminating the interference of the point light source in the cell fluorescence image.

In the embodiment of this application, the CLAHE algorithm calls the function cv2.createCLAHE() provided by opencv to process the cell fluorescence image. The function parameters are clipLimit and tileGridSize, where clipLimit is the histogram clipping threshold, and tileGridSize is the block size. Using the CLAHE algorithm to suppress the noise and enhance the contrast of the cytofluorescence image is as follows:

S21: divide the cell fluorescence image into non-overlapping sub-blocks of equal size, and the size of each sub-block is tileGridSize;

S22: Calculate the complete histogram of the cell fluorescence image (ie, the pixel value distribution map) and the histogram of each sub-block, and determine the clipping threshold clipLimit of the histogram;

S23: Evenly fill the pixel values exceeding the clipping threshold clipLimit in the histogram of each sub-block into the complete histogram, and complete the redistribution of redundant pixels in the histogram of the sub-block;

S24: Perform pixel value equalization processing on the complete histogram after pixel filling by function mapping;

Among them, the pixel value equalization process is shown in Figure 2, and the pixel values in the range of DA-DA+ΔDA are mapped to DB-DB+ΔDB through the function f, so that the distribution of image pixel values is more uniform. When the range before and after the transformation is known, the function f can be calculated automatically.

S25: Using bilinear interpolation to reconstruct the pixel gray value of the complete histogram after pixel value equalization processing, to obtain a processed cell fluorescence image.

S30: performing bilateral filter (Bilateral Filter) processing on the cytofluorescence image processed by CLAHE, removing the noise in the cytofluorescence image while retaining the edge information of the cytofluorescence image;

In this step, bilateral filtering is a variant of Gaussian filtering, and a distance mechanism is introduced on the basis of Gaussian filtering to better protect the edge information of objects in the image while removing image noise. The basic idea of bilateral filtering is that the pixel value of each position in the image is jointly determined by the pixel values of the neighborhood, which is implemented by calling the function cv2.bilateralFilter() provided by opencv. The algorithm principle of bilateral filtering for cell fluorescence images is:

Let G _σ (x) be a two-dimensional Gaussian kernel,

G _σs and G _σr are space weight and gray scale weight (rangeweight) respectively, the space weight plays the role of protecting edge information, and the gray scale weight plays the role of removing noise. I _p is the pixel value of the position to be calculated, I _q is the pixel value in the neighborhood, p=(p _x , p _y ) is the position of the pixel value to be calculated, q=(q _x ,q _y ) is the pixel value in the neighborhood The pixel value position, ||pq|| is the spatial distance between p and q, then:

Among them, BF represents the bilateral filter, BF[I] _p represents the pixel value at position p in the image calculated by the bilateral filter, s in G _σs represents space, G _σs is the space weight calculated according to the Gaussian kernel formula, q∈S The S represents the neighborhood range, that is, when the bilateral filter is used to calculate the pixel value at position p, the result is determined by the weighted sum of each pixel value in the neighborhood range, and the weight of each pixel value in the neighborhood range is determined by the spatial weight G _σs and gray weight G _σr are jointly determined. W _p is the normalization factor, and the calculation formula of W _p is:

In the embodiment of the present application, bilateral filtering is used to process the cell fluorescence image, which can ensure that the edges of the cells collected in the image are clear and well-defined while eliminating noise.

S40: Use the OTSU (Otsu method-maximum class variance method) algorithm to perform binarization processing on the filtered cell fluorescence image to obtain a binarized image;

In this step, the OTSU algorithm is an algorithm for determining the image binarization segmentation threshold. The algorithm divides the image into two parts, the background and the foreground (i.e., cells) according to the grayscale characteristics of the image. After binarization processing, the inter-class variance between the foreground and background images of the image is maximized. The greater the inter-class variance between the background and the foreground, the greater the difference between the two parts of the image. When part of the foreground is misclassified as the background or part of the background is misclassified as the foreground, the difference between the background and the foreground will change. Small. In this embodiment of the present application, the binarized segmentation using the OTSU algorithm can minimize the misclassification probability of the background and the foreground.

Specifically, the function called by the OTSU algorithm is the function cv2.THRESH_OTSU provided by opencv. The algorithm process is as follows: Assume that there is a threshold TH to divide all pixels of the cell fluorescence image into two categories: C1 (less than TH) and C2 (greater than TH). The respective pixel mean values of class pixels are m1 and m2 respectively, the global pixel mean value of the image is mG, and the probabilities of pixels being divided into C1 and C2 classes are p1 and p2 respectively, then:

p1*m1+p2*m2＝mG (3)

p1+p2=1 (4)

The expression for the between-class variance is:

σ ² =p1(m1-mG) ² +p2(m2-mG) ² (5)

The threshold that maximizes the value of formula (5) is the OTSU threshold, and the OTSU threshold can be obtained by traversing 0-255 gray levels, and the cell fluorescence image is binarized to separate the cells and the background area (the cell pixel value is used as 1 means that the pixel value of the background area is represented by 0).

S50: Obtain the connected domains in the binarized image, calculate the area of each connected domain, and eliminate the connected domains whose area is smaller than a set threshold from the cell fluorescence image, to obtain a final thresholded cell fluorescence image;

In this step, since the edge of the cell is easily damaged when the image is filtered, the shape of the cell is incomplete, resulting in some residual noise in the binarized image. Since each cell has an area of a certain size, and the size of the cell area far exceeds the area of the noise, the embodiment of this application uses the function cv2.connectedComponents() provided by opencv to obtain the connected components in the binarized image, and calculate each When the connected domain area is less than the set threshold, the connected domain is considered to be noise, and its pixel value is changed from 1 to 0, which is the same as the background area, so as to remove the residual noise in the image.

Specifically, since the pixel values in the connected domain are all 1, all pixel values in the connected domain are accumulated, and the total number of pixel values obtained is the area of the connected domain.

Based on the above, the cell fluorescence image thresholding method of the embodiment of the present application uses the CLAHE algorithm to adjust the pixel value distribution of the cell fluorescence image, performs bilateral filtering on the cell fluorescence image after CLAHE processing, and uses the OTSU algorithm to filter the cell fluorescence image Binarization was performed to minimize the probability of background and foreground misclassification, and the residual noise in the image was removed according to the area of the connected region to obtain the final thresholded cytofluorescence image. The embodiment of the present application effectively removes the interference of the point light source imaging on the fluorescent image, and retains the boundary information of the cells while removing the influence of the point light source, which increases the data volume of the cells to be analyzed and is beneficial to improve the accuracy of subsequent extraction of biological parameters. This application is suitable for the imaging scene of cell fluorescence images with relatively large point light source light, and is applicable to different types of cell microscopic imaging platforms (for the same set of cell microscopic imaging platforms, no secondary parameter adjustment is required, and for different cell microscopic imaging platforms, only It can be applied with simple parameter adjustment, and the number of parameters is small).

Please refer to FIG. 3 , which is a schematic structural diagram of the cell fluorescence image thresholding system of the embodiment of the present application. The cell fluorescence image thresholding system 40 of the embodiment of the present application includes:

Image acquisition module 41: for acquiring cell fluorescence images;

Pixel value adjustment module 42: used to suppress noise and enhance contrast processing on the cell fluorescence image by using the CLAHE algorithm, and adjust the pixel value distribution of the cell fluorescence image; wherein, the CLAHE algorithm limits the enhancement range of the local contrast by limiting the height of the local histogram , can enhance the contrast while suppressing the noise, thereby limiting the noise amplification and local contrast enhancement of the cytofluorescence image, realizing the brightness redistribution of the cytofluorescence image, and improving the pixel value distribution of the cytofluorescence image caused by the point light source. , to eliminate the interference of point light sources in the cell fluorescence image.

Image filtering module 43: used to perform bilateral filter (Bilateral Filter) processing on the cell fluorescence image after CLAHE processing, to remove the noise in the cell fluorescence image while retaining the edge information of the cell fluorescence image; wherein, the bilateral filter is a variant of the Gaussian filter , the distance mechanism is introduced on the basis of Gaussian filtering, so that the edge information of objects in the image can be better protected while removing image noise. In the embodiment of the present application, bilateral filtering is used to process the cell fluorescence image, which can ensure that the edges of the cells collected in the image are clear and well-defined while eliminating noise.

Binarization module 44: used to use OTSU algorithm to binarize the filtered cell fluorescence image to separate cells and background areas in the cell fluorescence image; wherein, the OTSU algorithm is a method to determine image binarization segmentation Threshold algorithm, which divides the image into two parts, the background and the foreground, according to the grayscale characteristics of the image. Variance is a measure of the uniformity of the gray distribution. The larger the inter-class variance between the background and the foreground, the greater the difference between the two parts that make up the image. When part of the foreground is misclassified as the background or part of the background is misclassified Foreground will cause the difference between background and foreground to be smaller. In this embodiment of the present application, the binarized segmentation using the OTSU algorithm can minimize the misclassification probability of the background and the foreground.

Noise elimination module 45: used to calculate the area of each connected region in the cell fluorescence image after binarization processing, and eliminate the connected region whose area is smaller than the set threshold from the cell fluorescence image to obtain the final thresholded cell fluorescence image; Among them, the shape of the cell is incomplete because it is easy to destroy the cell edge when filtering the image. Therefore, there is still some residual noise in the binarized cell fluorescence image. Since each cell has a certain area, and the size of the cell area far exceeds the noise area, the embodiment of the present application calculates the area of each connected region in the binarized cell fluorescence image, if the area of the connected region is less than The set threshold, that is, the area is considered to be noise, and it will be erased from the image, thereby removing the residual noise in the image.

Please refer to FIG. 4 , which is a schematic diagram of a terminal structure in an embodiment of the present application. The terminal 50 includes a processor 51 and a memory 52 coupled to the processor 51 .

The memory 52 stores program instructions for realizing the above-mentioned method for thresholding a cell fluorescence image.

The processor 51 is used to execute the program instructions stored in the memory 52 to control the thresholding of the cell fluorescence image.

Wherein, the processor 51 may also be referred to as a CPU (Central Processing Unit, central processing unit). The processor 51 may be an integrated circuit chip with signal processing capabilities. The processor 51 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components . A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.

Please refer to FIG. 5 , which is a schematic structural diagram of a storage medium according to an embodiment of the present application. The storage medium of the embodiment of the present application stores a program file 61 capable of realizing all the above-mentioned methods, wherein the program file 61 can be stored in the above-mentioned storage medium in the form of a software product, and includes several instructions to make a computer device (which can It is a personal computer, a server, or a network device, etc.) or a processor (processor) that executes all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc., which can store program codes. , or terminal devices such as computers, servers, mobile phones, and tablets.

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined in this invention may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention will not be limited to these embodiments shown in the present invention, but will conform to the widest scope consistent with the principles and novel features disclosed in the present invention.

Claims

A cell fluorescence image thresholding method, characterized in that, comprising:

Adjust the pixel value distribution of the cell fluorescence image by using the limited contrast adaptive local histogram equalization algorithm;

performing bilateral filter processing on the cell fluorescence image after the pixel value distribution adjustment;

performing binarization processing on the cell fluorescence image after the bilateral filter processing by using the Otsu method to obtain a binarized image;

The connected domains in the binarized image are obtained, and the area of each connected domain is calculated, and the connected domains whose area is smaller than a set threshold are eliminated to obtain a thresholded cell fluorescence image.
The cell fluorescence image thresholding method according to claim 1, wherein the adjustment of the pixel value distribution of the cell fluorescence image by using a limited contrast adaptive local histogram equalization algorithm is specifically:

Divide the cell fluorescence image into non-overlapping sub-blocks of equal size, the size of each sub-block is tileGridSize;

Calculate the complete histogram of the cell fluorescence image and the histogram of each sub-block, and determine the clipping threshold clipLimit of the histogram;

Evenly fill the pixel values exceeding the clipping threshold clipLimit in the histogram of each sub-block into the complete histogram, and complete the redistribution of redundant pixels in the histogram of the sub-block:

Perform pixel value equalization processing on the complete histogram after pixel filling through function mapping;

Bilinear interpolation is used to reconstruct the pixel gray value of the complete histogram after pixel value equalization processing, and the processed cell fluorescence image is obtained.
The cell fluorescence image thresholding method according to claim 2, wherein the bilateral filter processing of the cell fluorescence image after the pixel value distribution adjustment is specifically:

Let G σ (x) be a two-dimensional Gaussian kernel,
G σs , G σr are spatial weight and gray weight respectively, I p is the pixel value of the position to be calculated, I q is the pixel value in the neighborhood, p=(p x , p y ) is the position of the pixel value to be calculated ,q=(q x ,q y ) is the pixel value position in the neighborhood, ||pq|| is the spatial distance between p and q, then:

Among them, BF represents the bilateral filter, BF[I] p represents the pixel value at position p in the cell fluorescence image calculated using the bilateral filter, and q∈S represents the pixel value at position p calculated using the bilateral filter, and the result is determined by the neighborhood The weighted sum of each pixel value in the range S is determined. The weight of each pixel value in the neighborhood is determined by the spatial weight G σs and the gray weight G σr ; W p is the normalization factor, and the calculation formula of W p is :
The method for thresholding cell fluorescence images according to claim 3, wherein the binarization processing of the cell fluorescence images processed by the bilateral filter using the Otsu method is specifically:

All pixels of the cell fluorescence image are divided into C1 and C2 by the threshold TH, C1 is a pixel class smaller than TH, and C2 is a pixel class larger than TH, assuming that the pixel averages of C1 and C2 are m1 and m2 respectively, the global pixel of the image The mean is mG, and the probabilities of pixels being classified into C1 and C2 classes are p1 and p2 respectively, then:

p1*m1+p2*m2=mG

p1+p2=1

The expression for the between-class variance is:

σ 2 =p1(m1-mG) 2 +p2(m2-mG) 2

The threshold that maximizes the value of σ 2 is the maximum inter-class variance threshold, and the cell fluorescence image is binarized through the maximum inter-class variance threshold to separate cells and background regions.
The cell fluorescence image thresholding method according to claim 4, wherein the calculation of the area of each connected domain is specifically:

All the pixel values of each connected domain are accumulated, and the total number of pixel values obtained is regarded as the area of the connected domain.
A cell fluorescence image thresholding system, characterized in that, comprising:

Pixel value adjustment module: used to adjust the pixel value distribution of the cell fluorescence image by using the limited contrast adaptive local histogram equalization algorithm;

Image filtering module: for performing bilateral filtering processing on the cell fluorescence image after the pixel value distribution adjustment;

Binarization module: for performing binarization processing on the cell fluorescence image processed by the bilateral filter by using the Otsu method to obtain a binarized image;

Noise elimination module: used to obtain the connected domains in the binarized image, calculate the area of each connected domain, eliminate the connected domains whose area is smaller than the set threshold, and obtain the thresholded cell fluorescence image.
The cell fluorescence image thresholding system according to claim 6, wherein the pixel value adjustment module uses a limited contrast adaptive local histogram equalization algorithm to adjust the pixel value distribution of the cell fluorescence image specifically as follows: the cell fluorescence The image is divided into non-overlapping sub-blocks of equal size, and the size of each sub-block is tileGridSize; calculate the sub-block histogram, and determine the clipping threshold clipLimit of the histogram; uniform the pixel values exceeding the clipping threshold clipLimit in each sub-block Fill the entire histogram to complete the redistribution of redundant pixels: perform pixel value equalization processing on the histogram through function mapping; use bilinear interpolation to reconstruct the pixel gray value of the histogram to obtain the processed cells Fluorescence image.
The cellular fluorescence image thresholding system according to claim 6 or 7, wherein the binarization module utilizes the Otsu method to perform binarization processing on the cellular fluorescence image after the bilateral filtering process is specifically:

All pixels of the cell fluorescence image are divided into C1 and C2 by the threshold TH, C1 is a pixel class smaller than TH, and C2 is a pixel class larger than TH, assuming that the pixel averages of C1 and C2 are m1 and m2 respectively, the global pixel of the image The mean is mG, and the probabilities of pixels being classified into C1 and C2 classes are p1 and p2 respectively, then:

p1*m1+p2*m2=mG

p1+p2=1

The expression for the between-class variance is:

σ 2 =p1(m1-mG) 2 +p2(m2-mG) 2

The threshold that maximizes the value of σ 2 is the maximum inter-class variance threshold, and the cell fluorescence image is binarized through the maximum inter-class variance threshold to separate cells and background regions.
A terminal, characterized in that the terminal includes a processor and a memory coupled to the processor, wherein,

The memory stores program instructions for realizing the method for thresholding cell fluorescence images according to any one of claims 1-5;

The processor is configured to execute the program instructions stored in the memory to control the thresholding of the cell fluorescence image.
A storage medium, which is characterized by storing program instructions executable by a processor, the program instructions being used to execute the cell fluorescence image thresholding method according to any one of claims 1 to 5.