CN114821578A - Microsphere counting method and system based on microscopic imaging and electronic equipment - Google Patents

Abstract

The invention relates to a microsphere counting method, system and electronic equipment based on microscopic imaging. The method includes: acquiring a first to-be-measured image of a microsphere solution to be measured in each to-be-counted area based on the microscopic imaging technology; Preprocess each first image to be measured to obtain a second image to be measured of each area to be counted; perform statistics on each circular connected area in each second image to be measured to obtain a solution of microspheres to be measured The number of each type of microspheres in . The invention obtains the number of microspheres with different diameters by adopting the microscopic imaging technology, and by combining the microscopic imaging technology with the computer vision technology, while overcoming the defects of the existing microsphere counting software such as low resolution and narrow application range, The automatic identification, classification and statistics of microsphere images are realized, and the universality of the microscopic imaging microsphere counting platform is further broadened, and the speed and accuracy of analysis and detection are also well improved.

Description

A method, system and electronic device for counting microspheres based on microscopic imaging

technical field

The invention relates to the technical field of image recognition, in particular to a microsphere counting method, system and electronic device based on microscopic imaging.

Background technique

At present, most of the existing microsphere counting systems use biosensors for analysis and detection, such as electrochemical biosensors, mechanical biosensors, magnetic relaxation biosensors, etc. Although they have the advantages of high sensitivity and fast analysis speed, their signal readout Both the method and the equipment have many shortcomings to be solved. Electrochemical biosensors usually need to modify the working electrode, and their signal processing methods are complex, requiring special signal acquisition cards; mechanical biosensors have high requirements for device processing and are difficult to mass-produce; magnetic relaxation biosensors need to use A special low-field NMR instrument, which has strict temperature requirements and a large equipment volume, which is difficult to adapt to rapid on-site detection. Optical biosensors mainly use the light signal generated by the reaction between the substance to be tested and the detection reagent as the detection basis, and are usually composed of three functional elements: a light sensing layer, an optical signal conversion element and a signal amplification processing element. However, the economic cost of the existing optical sensors is relatively high, the structure is precise, the signal processing requirements are high, and there is a certain gap between their stability and other biosensors.

Therefore, it is urgent to develop new optical sensing platforms, improve biological functional substances, and explore new optical labeling materials or signal probes, which are of great significance for the further development and application of optical biosensors.

SUMMARY OF THE INVENTION

In order to solve the above technical problems, the present invention provides a microsphere counting method, system and electronic device based on microscopic imaging.

The technical scheme of a microsphere counting method based on microscopic imaging of the present invention is as follows:

S1. Based on the microscopic imaging technology, obtain the first to-be-measured image of the to-be-measured microsphere solution in each to-be-counted area;

S2. Preprocess each first image to be measured to obtain a second image to be measured of each area to be counted, and each second image to be measured includes at least one circular connected domain, one circular The connected domain corresponds to a microsphere with a diameter;

S3. Perform statistics on each type of circular connected region in each second image to be measured, respectively, to obtain the quantity of each type of microspheres in the microsphere solution to be measured.

The beneficial effects of the microsphere counting method based on microscopic imaging of the present invention are as follows:

The method of the invention obtains the number of microspheres with different diameters by adopting the microscopic imaging technology, and by combining the microscopic imaging technology with the computer vision technology, it overcomes the defects of the existing microsphere counting software such as low resolution and narrow application range. At the same time, the automatic identification, classification and statistics of microsphere images are realized, which further broadens the universality of the microscopic imaging microsphere counting platform, and improves the speed and accuracy of analysis and detection.

On the basis of the above solution, the microsphere counting method based on microscopic imaging of the present invention can also be improved as follows.

Further, the S2 specifically includes:

S21, performing grayscale transformation on each of the first images to be tested to obtain a first preprocessed image of each of the first images to be tested;

S22, performing binarization processing on each first preprocessed image to obtain a second preprocessed image of each first image to be measured;

S23. Perform interference information filtering on each second preprocessed image to obtain each second image to be tested.

Further, the S23 specifically includes: setting a first threshold, and filtering the interference information of each second preprocessed image according to the first threshold to obtain each second image to be tested; wherein, ResNet50 is used. The neural network model predicts the first threshold.

Further, the S3 specifically includes:

S31, marking each circular connected domain in each of the second images to be tested, to obtain a marked image corresponding to each of the second images to be tested;

S32, performing identification statistics on the quantity of each marker in each marker image to obtain the quantity of each circular connected domain;

S33. Obtain the quantity of each type of microspheres in the microsphere solution to be tested according to the quantity of each type of circular connected domains.

Further, the diameter of each microsphere is not less than 2 microns.

The technical scheme of a microsphere technology system based on microscopic imaging of the present invention is as follows:

It includes: a first processing module, a second processing module and a counting module;

The first processing module is used for: acquiring the first image to be measured of the microsphere solution to be measured in each area to be counted based on the microscopic imaging technology;

The second processing module is used to: preprocess each first image to be measured to obtain a second image to be measured of each area to be counted, and each second image to be measured includes at least one kind of circle respectively. Shape connected domain, a circular connected domain corresponds to a microsphere of a diameter;

The counting module is used for: separately performing statistics on each type of circular connected domain in each second image to be tested, to obtain the quantity of each type of microspheres in the microsphere solution to be tested.

The beneficial effects of a microscopic imaging-based microsphere technology system of the present invention are as follows:

The system of the invention obtains the number of microspheres with different diameters by adopting the microscopic imaging technology, and by combining the microscopic imaging technology with the computer vision technology, it can overcome the defects of the existing microsphere counting software such as low resolution and narrow application range. At the same time, the automatic identification, classification and statistics of microsphere images are realized, which further broadens the universality of the microscopic imaging microsphere counting platform, and improves the speed and accuracy of analysis and detection.

On the basis of the above solution, the microsphere technology system based on microscopic imaging of the present invention can also be improved as follows.

Further, the second processing module specifically includes: a first preprocessing module, a second preprocessing module and a third preprocessing module;

The first preprocessing module is used for: performing grayscale transformation on each of the first images to be tested to obtain a first preprocessed image of each of the first images to be tested;

The second preprocessing module is used for: performing binarization processing on each first preprocessing image to obtain a second preprocessing image of each first image to be tested;

The third preprocessing module is configured to: filter interference information on each second preprocessed image to obtain each second image to be tested.

Further, the counting module specifically includes: a first counting module, a second counting module and a third counting module;

The first counting module is used to: mark each circular connected domain in each of the second images to be tested to obtain a marked image corresponding to each of the second images to be tested;

The second counting module is used to: identify and count the quantity of each type of mark in each marked image, and obtain the quantity of each type of circular connected domain;

The third counting module is used for: obtaining the quantity of each type of microspheres in the microsphere solution to be measured according to the quantity of each type of circular connected domains.

A technical solution of a storage medium of the present invention is as follows:

Instructions are stored in the storage medium, and when the computer reads the instructions, the computer is made to execute the steps of a method for counting microspheres based on microscopic imaging of the present invention.

The technical scheme of a kind of electronic equipment of the present invention is as follows:

Including a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that, when the processor executes the computer program, the computer is made to execute a microscopic-based microcomputer according to the present invention. Imaging the steps of the microsphere counting method.

Description of drawings

1 is a schematic flowchart of a microsphere counting method based on microscopic imaging according to an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a microsphere counting system based on microscopic imaging according to an embodiment of the present invention.

Detailed ways

As shown in FIG. 1 , the first embodiment of a method for counting microspheres based on microscopic imaging of the present invention includes the following steps:

S1. Based on the microscopic imaging technology, obtain a first image to be measured of the microsphere solution to be measured in each area to be counted.

Among them, the microscopic imaging technology adopts the magnification imaging function of a bright-field optical microscope, and the model of the bright-field optical microscope is: LK-50, which was purchased from Tianjin Leike Co., Ltd. It should be noted that, in addition to a bright-field optical microscope, the imaging acquisition device may also be a dark-field optical microscope, a smartphone with a photographing function, a CMOS camera, and the like, which is not limited herein.

Wherein, the microsphere solution in this embodiment is a polystyrene microsphere solution as an example, and can also be any other polymer material microspheres, such as latex microspheres, silica microspheres, etc., which are not limited herein.

The microspheres in the polystyrene microsphere solution were carboxyl-modified microspheres with diameters of 2 μm PS ₂₀₀₀ , 4 μm PS ₄₀₀₀ , and 6 μm PS ₆₀₀₀ , which were all purchased from Bangs Laboratories.

It should be noted that the microspheres in this embodiment adopt carboxyl-modified microspheres with diameters of 2 microns, 4 microns, and 6 microns, and microspheres with a diameter of not less than 2 microns can also be used according to actual needs, and there is no limitation here. .

Among them, for the convenience of illustration, the PS ₂₀₀₀ was serially diluted to 1 μg/mL, 0.5 μg/mL, 0.25 μg/mL, 0.1 μg/mL, 0.05 μg/mL, 0.01 μg/mL with pure water; PS 2000 was diluted with pure water. ₄₀₀₀ was serially diluted to 10μg/mL, 5μg/mL, 2μg/mL, 1μg/mL, 0.5μg/mL, 0.25μg/mL; PS ₆₀₀₀ was serially diluted to 30μg/mL, 20μg/mL, 10μg/mL with pure water , 5μg/mL, 1μg/mL, 0.5μg/mL; the microsphere solution to be tested is made of microspheres with any concentration of at least one diameter of PS ₂₀₀₀ , PS ₄₀₀₀ , and PS ₆₀₀₀ . For example, take 10ml of PS ₂₀₀₀ with a concentration of 0.5μg/mL, 10ml of PS ₄₀₀₀ with a concentration of 5μg/mL, and 10ml of PS ₆₀₀₀ with a concentration of 20μg/mL to prepare the microsphere solution to be tested.

Specifically, the microsphere solution to be tested was placed in an ultrasonic cleaner to sonicate for 15 minutes. To ensure a good observation effect, the coverslips were soaked and washed with ethanol-water solution (75%, v/v), and dried with lens tissue. Drop 10 μL of the microsphere solution to be tested on the counting area of a clean hemocytometer, and gently cover the wiped coverslip, taking care not to generate air bubbles during this process. Under the action of capillary phenomenon, the microsphere solution to be tested will fill the entire counting area. After standing for 3 minutes, the particles in the area to be counted naturally settled and stabilized to avoid counting errors caused by delamination. Place the assembled glass slide under an optical microscope, select a 10x eyepiece and a 20x objective, and pass Computer software (eg, ImageView) captures images in the counting area on a computer screen.

Among them, the blood cell counter was purchased from Shanghai Qiujing Biochemical Reagent Instrument Co., Ltd.; the model of ultrasonic cleaner was BX5200HP, which was purchased from Shanghai Xinmiao Medical Equipment Manufacturing Co., Ltd.; ethanol was purchased from Sinopharm Chemical Reagent Co., Ltd.

Among them, a blood cell counter includes two sample adding areas, each sample adding area corresponds to a to-be-counted area. In order to increase the accuracy of the method and the reliability of the results, this embodiment selects two counting plates (four to-be-counted The image of the microspheres in the region) was captured and identified.

It should be noted that the number of counting areas can be set according to requirements. The more counting areas that are captured, the higher the accuracy, but the time cost will increase accordingly.

The first to-be-measured image is an image of the microspheres in the to-be-counted area captured by computer software (ImageView).

S2. Preprocess each first image to be measured to obtain a second image to be measured of each area to be counted, and each second image to be measured includes at least one circular connected domain, one circular A connected domain corresponds to a microsphere of one diameter.

The preprocessing process generally includes steps such as grayscale image processing, binarization processing, and image interference information filtering.

The second to-be-measured image includes at least one circular connected domain, each circular connected domain has a different diameter, and one circular connected domain corresponds to a microsphere with one diameter.

S3. Perform statistics on each type of circular connected region in each second image to be measured, respectively, to obtain the quantity of each type of microspheres in the microsphere solution to be measured.

The microsphere itself is spherical, but in the first image to be measured captured by the optical microscope, the microsphere initially appears to be circular, and the second image to be measured obtained by preprocessing the first image to be measured can be more It is clearly distinguished that the microspheres are circular, so they are circular connected domains in the second image to be measured.

Among them, each circular connected domain adopts a different mark, which is convenient for the software to identify the quantity of each object to be recognized.

Among them, after identifying each circular connected domain, the number of each circular connected domain is obtained, and classification statistics are performed according to the number of each circular connected domain to obtain the number of each type of microspheres of each diameter; finally The data information in the image is classified and counted into a table for easy reference.

Preferably, the S2 specifically includes:

S21. Perform grayscale transformation on each of the first images to be tested to obtain a first preprocessed image of each of the first images to be tested.

The function of performing grayscale image conversion on the first image to be tested is to avoid the interference of natural light on the color information in the image.

Wherein, the first preprocessed image is an image obtained after grayscale transformation of the first image to be tested.

S22. Perform binarization processing on each of the first pre-processed images to obtain a second pre-processed image of each of the first to-be-measured images.

Among them, the function of binarizing the first preprocessed image is to set the gray value of the pixel on the image to 0 or 255, that is, the process of presenting an obvious black and white effect on the entire image, which is convenient for the first The information in the preprocessed image can be processed by computer language.

Wherein, the second preprocessed image is an image obtained after the first image to be tested is binarized.

S23. Perform interference information filtering on each second preprocessed image to obtain each second image to be tested.

Wherein, each second to-be-measured image includes at least one circular connected domain.

Preferably, the S23 specifically includes: setting a first threshold, and filtering the interference information of each second preprocessed image according to the first threshold to obtain each second image to be tested; wherein, The first threshold is predicted using a ResNet50 neural network model.

Specifically, because the background of each counter board (counter) is relatively simple, the object to be detected (circular connected domain) and the background have obvious differences in color depth (the darker the color, the smaller the gray value of the pixel, the lighter the color , the gray value of the pixel is larger), using this feature, set a first threshold, the pixels greater than the set threshold are set to the maximum value of 255, and the pixels less than the threshold are set to 0, so it can achieve filtering noise information and Focus on the location information of the target to be detected.

It should be noted that the set first threshold is generally set manually. In this embodiment, the neural network model ResNet50 is used for prediction, which replaces the global threshold and improves the detection accuracy.

Preferably, the S3 specifically includes:

S31. Mark each circular connected region in each of the second images to be tested to obtain a marked image corresponding to each of the second images to be tested.

S32. Perform identification statistics on the quantity of each type of mark in each mark image, and obtain the quantity of each type of circular connected domain.

S33. Obtain the quantity of each type of microspheres in the microsphere solution to be tested according to the quantity of each type of circular connected domains.

Preferably, the diameter of each microsphere is not less than 2 microns.

Among them, the microspheres are used as signal probes because they can be observed when their size is greater than or equal to 2 microns.

Among them, in this embodiment, four areas to be counted are used by default for counting, classification and identification.

The microsphere counting method (system) of this embodiment has a sensitive and stable signal readout function, and can be applied to fields such as rapid detection of food safety, in vitro diagnosis, and environmental monitoring.

The technical solution of the present embodiment obtains the number of microspheres with different diameters by using the microscopic imaging technology, and combines the microscopic imaging technology with the computer vision technology to overcome the low resolution and narrow application range of the existing microsphere counting software. At the same time, it realizes the automatic identification, classification and statistics of microsphere images, which further broadens the universality of the microscopic imaging microsphere counting platform, and also improves the speed and accuracy of analysis and detection.

In this example, by preparing a series of solutions of microspheres with concentrations of 2 μm, 4 μm, 6 μm and three kinds of diameters, and counting the number of particles in the counting area through an optical microscope, it is found that when the concentration of the solution increases, the number of particles also increases. will increase, and there is a good correlation between the two. Through linear fitting, the linear equation between the concentration of 2μm, 4μm, and 6μm microspheres and the corresponding number in the counting area is: Y=450.86X-1.79 (R2=0.997), Y=39.822X+0.21 (R2= 0.992), Y=23.94X-5.38 (R2=0.993).

At the same time, manual counting and QSCounting counting are used to count the number of microspheres in the counting area. The results show that the manual counting has good consistency with the microscopic imaging-based microsphere counting results of this embodiment. Among them, for the microspheres with diameters of 4 μm and 6 μm, the manual counting result is taken as the abscissa, and the counting result of this embodiment is the ordinate, and the two are linearly fitted, and the linear equations are: Y=0.97X+1.11, Y =0.97X+1.04, and the correlation coefficient (R2) reached 0.999. For a microsphere with a diameter of 2 μm, the linear equation is: Y=0.97X+0.16, and R2 is 0.996.

To sum up, when the concentration of microspheres increases gradually, the number of microspheres in the counting area increases accordingly. Through parameter optimization, the computer vision microsphere counting software also obtained good accuracy, which greatly improved the analysis efficiency. This proves that the microscopic imaging microsphere counting system made by the method of this embodiment is a sensitive and stable signal readout device, and has a large application in food safety rapid detection, in vitro diagnosis and environmental monitoring, etc. potential in the field.

The advantages of the technical solution of this embodiment are: (1) the bright-field optical microscope has a mature manufacturing process, low manufacturing cost, and is easy to miniaturize; (2) different from fluorescent microspheres, which require complex modification processes, and require strict Preservation conditions, as a kind of polymer material, microspheres have stable chemical properties and simple preservation conditions; (3) the combination of microspheres with different diameters provides a new direction for simultaneous detection of multiple targets; (4) the use of microspheres The combination of imaging platform and computer vision technology can realize automatic identification, classification and statistics of batch images, which can better improve the efficiency of analysis and detection.

In the second embodiment of the microscopic imaging-based microsphere counting method of the present invention, on the basis of the above-mentioned first embodiment, the technical solution of this embodiment further includes:

S100, based on the microscopic imaging technology, obtain a first image to be measured of the solution of microspheres to be counted in each area to be counted.

S200. Preprocess each first image to be measured to obtain a second image to be measured of each area to be counted, and each second image to be measured includes at least one circular connected domain, one circular The shape-connected domain corresponds to a microsphere with a diameter; wherein, the second image to be measured includes: the microsphere to be measured and the free microsphere.

Among them, the microspheres to be tested are: the microspheres that need to be counted in this embodiment; the free microspheres are: the microspheres that do not need to be counted in this embodiment.

S300: Counting each circular connected area of the microspheres to be measured in each second image to be measured, to obtain the quantity of each type of microspheres to be measured in the solution of the microspheres to be measured.

Specifically, for example, there is a microsphere with a diameter of 6 microns in the second image to be tested, and several microspheres with a diameter of 2 microns are connected around the microsphere; at this time, the microsphere to be tested is: attached to the 6 micron microsphere The free microspheres are: 6-micron microspheres and 2-micron microspheres in the free state; step S300 is used to count only the number of 2-micron microspheres attached to the 6-micron microspheres in the image.

It should be noted that, in this embodiment, parameters such as the material and diameter of the microspheres are not specifically required, and may be in any range.

The technical solution of this embodiment, on the basis of the first embodiment, can also effectively reduce the steps of biochemical reactions and improve the speed of analysis and detection.

As shown in FIG. 2 , a microsphere counting system 200 based on microscopic imaging according to an embodiment of the present invention includes: a first processing module 210 , a second processing module 220 and a counting module 230 ;

The first processing module 210 is used for: based on the microscopic imaging technology, to obtain the first to-be-measured image of the to-be-measured microsphere solution in each to-be-counted area;

The second processing module 220 is configured to: preprocess each first image to be measured to obtain a second image to be measured of each area to be counted, and each second image to be measured includes at least one Circular connected domain, a circular connected domain corresponds to a microsphere of a diameter;

The counting module 230 is used to: respectively perform statistics on each circular connected region in each second image to be measured, to obtain the number of each type of microspheres in the microsphere solution to be measured, wherein one circular A connected domain corresponds to a microsphere of one diameter.

Preferably, the second processing module 220 specifically includes: a first preprocessing module, a second preprocessing module and a third preprocessing module;

The first preprocessing module is used for: performing grayscale transformation on each of the first images to be tested to obtain a first preprocessed image of each of the first images to be tested;

The second preprocessing module is used for: performing binarization processing on each first preprocessing image to obtain a second preprocessing image of each first image to be tested;

The third preprocessing module is configured to: filter interference information on each second preprocessed image to obtain each second image to be tested.

Preferably, the counting module 230 specifically includes: a first counting module, a second counting module and a third counting module;

The first counting module is used to: mark each circular connected domain in each of the second images to be tested to obtain a marked image corresponding to each of the second images to be tested;

The second counting module is used to: identify and count the quantity of each type of mark in each marked image, and obtain the quantity of each type of circular connected domain;

The third counting module is used for: obtaining the quantity of each type of microspheres in the microsphere solution to be measured according to the quantity of each type of circular connected domains.

The technical solution of the present embodiment obtains the number of microspheres with different diameters by using the microscopic imaging technology, and combines the microscopic imaging technology with the computer vision technology to overcome the low resolution and narrow application range of the existing microsphere counting software. At the same time, it realizes the automatic identification, classification and statistics of microsphere images, which further broadens the universality of the microscopic imaging microsphere counting platform, and also improves the speed and accuracy of analysis and detection.

For the steps of implementing corresponding functions of each parameter and each module in the microscopic imaging-based microsphere counting system 200 of the present embodiment, reference may be made to the first section on the microscopic imaging-based microsphere counting method above. The parameters and steps in the embodiment will not be repeated here.

A storage medium provided by an embodiment of the present invention includes: an instruction is stored in the storage medium, and when a computer reads the instruction, the computer is made to execute steps such as a method for counting microspheres based on microscopic imaging. Refer to the parameters and steps in the first embodiment of the microscopic imaging-based microsphere counting method above, which will not be repeated here.

Computer storage media such as: USB flash drive, mobile hard disk, etc.

An electronic device provided by an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, when the processor executes the computer program, the processor causes the The computer executes steps such as a method for counting microspheres based on microscopic imaging. For details, reference may be made to the parameters and steps in the first embodiment of the method for counting microspheres based on microscopic imaging, which will not be repeated here.

Those skilled in the art know that the present invention can be implemented as a method, a system, a storage medium and an electronic device.

Therefore, the present invention can be embodied in the following forms, that is, it can be complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software. Called a "circuit," "module," or "system." Furthermore, in some embodiments, the present invention may also be implemented in the form of a computer program product on one or more computer-readable media having computer-readable program code embodied thereon. Any combination of one or more computer-readable media may be employed. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples (non-exhaustive list) of computer readable storage media include: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), Erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. Although the embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Embodiments are subject to variations, modifications, substitutions and variations.

Claims (10)

1. a microsphere counting method based on microscopic imaging, is characterized in that, comprises: S1. Based on the microscopic imaging technology, obtain the first to-be-measured image of the to-be-counted microsphere solution in each to-be-counted area; S2. Preprocess each first image to be measured to obtain a second image to be measured of each area to be counted, and each second image to be measured includes at least one circular connected domain, one circular The connected domain corresponds to a microsphere with a diameter; S3. Perform statistics on each type of circular connected domain in each second image to be tested to obtain the quantity of each type of microspheres in the microsphere solution to be tested. 2. The method for counting microspheres based on microscopic imaging according to claim 1, wherein the S2 specifically comprises: S21, performing grayscale transformation on each of the first images to be tested to obtain a first preprocessed image of each of the first images to be tested; S22, performing binarization processing on each first preprocessed image to obtain a second preprocessed image of each first image to be measured; S23. Perform interference information filtering on each second preprocessed image to obtain each second image to be tested. 3 . The method for counting microspheres based on microscopic imaging according to claim 2 , wherein the S23 specifically comprises: setting a first threshold, and pre-processing each second according to the first threshold. 4 . The interference information is filtered on the image to obtain each second image to be tested; wherein, the first threshold is predicted by using the ResNet50 neural network model. 4. The method for counting microspheres based on microscopic imaging according to claim 1, wherein the S3 specifically comprises: S31, marking each circular connected domain in each of the second images to be tested, to obtain a marked image corresponding to each of the second images to be tested; S32, performing identification statistics on the quantity of each marker in each marker image to obtain the quantity of each circular connected domain; S33. Obtain the quantity of each type of microspheres in the microsphere solution to be tested according to the quantity of each type of circular connected domains. 5. The method for counting microspheres based on microscopic imaging according to claim 4, wherein the diameter of each microsphere is not less than 2 microns. 6. A microsphere counting system based on microscopic imaging, characterized in that, comprising: a first processing module, a second processing module and a counting module; The first processing module is used for: acquiring the first image to be measured of the microsphere solution to be measured in each area to be counted based on the microscopic imaging technology; The second processing module is used to: preprocess each first image to be measured to obtain a second image to be measured of each area to be counted, and each second image to be measured includes at least one kind of circle respectively. Shape connected domain, a circular connected domain corresponds to a microsphere of a diameter; The counting module is used for: performing statistics on each type of circular connected domain in each second image to be measured, to obtain the quantity of each type of microspheres in the microsphere solution to be measured. 7. The microsphere counting system based on microscopic imaging according to claim 6, wherein the second processing module specifically comprises: a first preprocessing module, a second preprocessing module and a third preprocessing module; The first preprocessing module is used for: performing grayscale transformation on each of the first images to be tested to obtain a first preprocessed image of each of the first images to be tested; The second preprocessing module is used for: performing binarization processing on each first preprocessing image to obtain a second preprocessing image of each first image to be tested; The third preprocessing module is configured to: filter interference information on each second preprocessed image to obtain each second image to be tested. 8. The microsphere counting system based on microscopic imaging according to claim 6, wherein the counting module specifically comprises: a first counting module, a second counting module and a third counting module; The first counting module is used to: mark each circular connected domain in each of the second images to be tested to obtain a marked image corresponding to each of the second images to be tested; The second counting module is used to: identify and count the quantity of each type of mark in each marked image, and obtain the quantity of each type of circular connected domain; The third counting module is used for: obtaining the quantity of each type of microspheres in the microsphere solution to be measured according to the quantity of each type of circular connected domains. 9. A storage medium, characterized in that, the storage medium stores instructions, and when a computer reads the instructions, the computer is made to execute the microscopic-based method according to any one of claims 1 to 5. Microsphere counting method for imaging. 10. An electronic device, comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein when the processor executes the computer program, the computer executes the program as claimed in the right. The microscopic imaging-based microsphere counting method of any one of claims 1 to 5.

Images