Bright-field microdroplet image recognition method and device, computer device and storage medium

Abstract

The application relates to a bright-field microdroplet image recognition method and device of a nucleic acid molecule, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a bright-field microdroplet image, starting from the edge of the bright-field microdroplet image, determining a first pixel point, and determining a second pixel point symmetrical to the first pixel point; determining a circular dot and a radius of the bright-field microdroplet image according to the first pixel point and the second pixel point; taking the circular dot as the center, searching for corresponding target pixel points in a first range of a plurality of preset directions of the circular dot; determining whether the bright-field microdroplet image is a real circle according to the plurality of target pixel points and the radius of the bright-field microdroplet image, obtaining a target microdroplet image of the bright-field microdroplet image being a real circle; performing similarity checking according to the target microdroplet image, and determining concentric circles in the target microdroplet image; and obtaining the diameter of the bright-field microdroplet image according to the radius of the circle with the largest radius in the concentric circles. The method can improve the accuracy of recognizing the microdroplet diameter.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of droplet digital PCR, in particular to a bright field droplet image recognition method and device, computer equipment and storage medium. BACKGROUND

[0002] Droplet digital polymerase chain reaction is the third generation of digital PCR technology, which is a method for absolute quantification of nucleic acid molecules. Through it, the copy number of target genes in the sample can be directly measured. The technology uses a digital PCR chip to dilute the nucleic acid sample and randomly disperse it into tens of thousands of water-in-oil structure droplets. Then the DNA sample in the microdroplet is combined with a specific fluorescent marker for PCR reaction. After the PCR reaction is completed, a microscope is used to take pictures of the microdroplets in bright field and dark field environments using a CCD camera, thereby obtaining bright field microdroplet images and dark field microdroplet images. Finally, image recognition and detection are performed on the two images respectively.

[0003] The main purposes of bright field microdroplet image recognition are microdroplet number statistics and diameter calculation. The bright field microdroplet image does not emit light, and does not need to be taken in complete darkness. Therefore, the exposure time is short, the noise is less, and the brightness and contrast of the image are high. However, the arrangement of the microdroplets is relatively dense, and the edges of the microdroplets are thick, which leads to the problem that the current microdroplet recognition algorithm has low calculation accuracy. SUMMARY

[0004] Therefore, it is necessary to provide a bright field microdroplet image recognition method, device, computer equipment and storage medium capable of improving the accuracy of recognizing the diameter of the microdroplets.

[0005] In a first aspect, the present application provides a bright field microdroplet image recognition method for nucleic acid molecules. The method comprises:

[0006] obtaining the bright field microdroplet image, traversing from the edge of the bright field microdroplet image, determining a first pixel point, and determining a second pixel point symmetrical to the first pixel point;

[0007] determining a circle point and a radius of the microdroplet image according to the first pixel point and the second pixel point;

[0008] finding corresponding target pixel points in a first range of a plurality of preset directions of the circle point;

[0009] determining whether the microdroplet image is a real circle according to the plurality of target pixel points found and the radius of the microdroplet image, and obtaining a target microdroplet image in which the microdroplet image is a real circle;

[0010] According to the target micro-droplet image, similarity checking is performed to determine a concentric circle in the target micro-droplet image;

[0011] According to a radius of a circle with the largest radius in the concentric circle, a diameter of the bright-field micro-droplet image is obtained.

[0012] In one of the embodiments,

[0013] The traversal from the edge of the bright-field micro-droplet image to determine a first pixel point and a second pixel point symmetrical to the first pixel point comprises:

[0014] Determining a first pixel point located at the edge of the bright-field micro-droplet image;

[0015] According to the first pixel point, a second pixel point symmetrical to the first pixel point in the bright-field micro-droplet image is searched;

[0016] If the second pixel point symmetrical to the first pixel point is not found in the bright-field micro-droplet image, returning to the step of determining the first pixel point located at the edge of the bright-field micro-droplet image;

[0017] If the second pixel point symmetrical to the first pixel point is found in the bright-field micro-droplet image, performing the step of determining a circle point and a radius of the micro-droplet image according to the first pixel point and the second pixel point.

[0018] In one of the embodiments, the searching of the second pixel point symmetrical to the first pixel point in the bright-field micro-droplet image according to the first pixel point comprises:

[0019] Searching for other edge pixel points in the same row as the first pixel point;

[0020] Determining whether the first pixel point and the searched edge pixel point are symmetrical;

[0021] If the first pixel point and the searched edge pixel point are symmetrical, the searched edge pixel point is taken as the second pixel point.

[0022] In one of the embodiments, the determination of whether the first pixel point and the searched edge pixel point are symmetrical comprises:

[0023] According to the coordinates of the first pixel point, a first gradient value of the first pixel point is calculated, and according to the coordinates of the searched edge pixel point, a second gradient value of the searched edge pixel point is calculated;

[0024] According to the first gradient value and the second gradient value, it is determined whether the first pixel point and the searched edge pixel point are symmetrical.

[0025] In one embodiment, determining the dot and radius of the droplet image based on the first pixel and the second pixel includes:

[0026] Calculate the line function containing the first pixel and the second pixel based on the coordinates of the first pixel and the second pixel, the first gradient value, and the second gradient value;

[0027] The intersection point of the two lines is determined based on the line function of the first pixel and the second pixel, and the intersection point is taken as the circle point;

[0028] Calculate the distance between the first pixel or the second pixel and the circle to obtain the radius.

[0029] In one embodiment, the step of searching for the corresponding target pixel point within a first range of multiple preset directions centered on the dot includes:

[0030] Centered on the circle, find the third and fourth pixels in the row where the circle is located, and find the fifth and sixth pixels in the column where the circle is located. The distances of the third, fourth, fifth, and sixth pixels from the circle are greater than or equal to a first preset distance and less than or equal to a second preset distance. The absolute value of the difference between the first preset distance and the radius is less than or equal to a preset threshold, and the absolute value of the difference between the second preset distance and the radius is less than or equal to a preset threshold.

[0031] The step of determining whether the droplet image is a true circle based on the found multiple target pixels and the radius of the droplet image includes:

[0032] Calculate the distances between the third, fourth, fifth, and sixth pixels and the center of the circle;

[0033] Determine whether the absolute value of the difference between the distance of the third pixel, the fourth pixel, the fifth pixel, and the sixth pixel and the center of the circle and the radius is less than or equal to a preset difference.

[0034] If the absolute values ​​of the differences between the distances of the third, fourth, fifth, and sixth pixels and the center of the circle and the radius are all less than or equal to a preset difference, then the droplet image is determined to be a real circle.

[0035] In one embodiment, the step of performing similarity verification based on the target droplet image to determine concentric circles in the target droplet image includes:

[0036] Calculate the difference between the centers of the circles in the target droplet image;

[0037] The target droplet image with a difference between the centers of the circles less than a third preset distance is taken as a concentric circle.

[0038] In one embodiment, the diameter of the bright-field droplet image is obtained according to the radius of the circle with the largest radius in the concentric circles.

[0039] The radius of the concentric circle is calculated according to the corresponding target pixel points of the concentric circle.

[0040] The maximum value is selected as the diameter of the bright-field droplet image from the calculation of the radius of the concentric circle from the symmetric target pixel points.

[0041] In a second aspect, the present application further provides a bright-field droplet image recognition device of a nucleic acid molecule. The device comprises:

[0042] The traversal module is configured to obtain the bright-field droplet image, traverse from the edge of the bright-field droplet image, determine the first pixel point, and determine the second pixel point symmetric to the first pixel point.

[0043] The first determination module is configured to determine the circle point and the radius of the droplet image according to the first pixel point and the second pixel point.

[0044] The first difference module is configured to find corresponding target pixel points in a first range in a plurality of directions around the circle point.

[0045] The real circle judgment module is configured to determine whether the droplet image is a real circle according to the plurality of target pixel points and the radius of the droplet image, and obtain a target droplet image with a real circle.

[0046] The concentric circle determination module is configured to perform similarity checking on the target droplet image to determine concentric circles in the target droplet image.

[0047] The second determination module is configured to obtain the diameter of the bright-field droplet image according to the radius of the circle with the largest radius in the concentric circles.

[0048] In a third aspect, the present application further provides a computer device. The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0049] The bright-field droplet image is obtained, the first pixel point is determined by traversing from the edge of the bright-field droplet image, and the second pixel point symmetric to the first pixel point is determined.

[0050] The circle point and the radius of the droplet image are determined according to the first pixel point and the second pixel point.

[0051] finding corresponding target pixel points in a first range of a plurality of directions preset with the center of the dot;

[0052] determining whether the microdroplet image is a real circle according to the found target pixel points and the radius of the microdroplet image, and obtaining a target microdroplet image in which the microdroplet image is a real circle;

[0053] performing similarity checking according to the target microdroplet image to determine concentric circles in the target microdroplet image;

[0054] obtaining the diameter of the bright-field microdroplet image according to the radius of the circle with the largest radius in the concentric circles.

[0055] In a fourth aspect, the present application further provides a computer readable storage medium. The computer readable storage medium has a computer program stored thereon, and the computer program, when executed by a processor, implements the following steps:

[0056] obtaining the bright-field microdroplet image, traversing from the edge of the bright-field microdroplet image, determining a first pixel point, and determining a second pixel point symmetrical to the first pixel point;

[0057] determining a dot and a radius of the microdroplet image according to the first pixel point and the second pixel point;

[0058] finding corresponding target pixel points in a first range of a plurality of directions preset with the center of the dot;

[0059] determining whether the microdroplet image is a real circle according to the found target pixel points and the radius of the microdroplet image, and obtaining a target microdroplet image in which the microdroplet image is a real circle;

[0060] performing similarity checking according to the target microdroplet image to determine concentric circles in the target microdroplet image;

[0061] obtaining the diameter of the bright-field microdroplet image according to the radius of the circle with the largest radius in the concentric circles.

[0062] In a fifth aspect, the present application further provides a computer program product. The computer program product comprises a computer program, and the computer program, when executed by a processor, implements the following steps:

[0063] obtaining the bright-field microdroplet image, traversing from the edge of the bright-field microdroplet image, determining a first pixel point, and determining a second pixel point symmetrical to the first pixel point;

[0064] determining a dot and a radius of the microdroplet image according to the first pixel point and the second pixel point;

[0065] Using the dot as the center, find the corresponding target pixel within a first range of multiple preset directions of the dot;

[0066] Based on the multiple target pixels found and the radius of the droplet image, determine whether the droplet image is a real circle, and obtain a target droplet image in which the droplet image is a real circle;

[0067] Based on the similarity check of the target droplet image, the concentric circles in the target droplet image are determined;

[0068] The diameter of the bright-field droplet image is obtained from the radius of the circle with the largest radius among the concentric circles.

[0069] The aforementioned bright-field droplet image recognition method, apparatus, computer equipment, storage medium, and computer program product for nucleic acid molecules first traverses the image from the edge to determine the first pixel, then finds symmetrical edge pixels as the first and second pixels. The center and radius of the droplet are then calculated using the coordinates of the symmetrical first and second pixels. After determining the center and radius, target pixels are searched in multiple preset directions around the center. The found target pixels and radius are used to verify whether the droplet image is a true circle. By traversing the first pixel, multiple bright-field droplet images that are true circles can be identified. Similarity checks are then performed on the true circle droplet images to determine concentric circles. Finally, the circle with the largest radius among the concentric circles is selected as the circle of the droplet image, and the diameter of the droplet image is obtained based on the largest radius among the concentric circles. Therefore, this application traverses all edge pixels to obtain the corresponding true circle bright-field droplet image, eliminating the influence of droplet edges on the calculation results and thus improving the accuracy of the calculation results. Attached Figure Description

[0070] Figure 1 This is an application environment diagram of a bright-field droplet image recognition method for nucleic acid molecules in one embodiment;

[0071] Figure 2 This is a flowchart illustrating a bright-field droplet image recognition method for nucleic acid molecules in one embodiment.

[0072] Figure 3 This is a schematic diagram of a droplet image for locating the first and second pixels in one embodiment.

[0073] Figure 4 This is a schematic diagram of a droplet image used in one embodiment to search for target pixels in multiple preset directions.

[0074] Figure 5 This is a flowchart illustrating a bright-field droplet image recognition method for nucleic acid molecules in another embodiment;

[0075] Figure 6 A structural block diagram of a bright field microdroplet image recognition device for a nucleic acid molecule in an embodiment;

[0076] Figure 7 An internal structural diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0077] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0078] The bright field microdroplet image recognition method for a nucleic acid molecule provided by the embodiments of the present application can be applied in an application environment as shown in Figure 1 The system includes a camera device 102 and a server 104. The camera device 102 can communicate with the server 104 through wired or wireless means. The data storage system can store image data required to be processed by the server 104. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers, and the camera device 102 and the server 104 can also be integrated into a terminal. The camera device 102 is used to take pictures of the bright field microdroplets of the nucleic acid molecules, and obtain the microdroplet images of each bright field microdroplet of the nucleic acid molecules. The server 104 identifies according to the received microdroplet images, and obtains the diameters of the microdroplet images. The server 104 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers and other devices or platforms with image processing capabilities. In a specific implementation, after the camera device 102 obtains the microdroplet images, the user can also copy or cut the microdroplet images focused by the camera device 102 to the server 104 through a U disk or a card reader.

[0079] In an embodiment, as shown in Figure 2 A bright field microdroplet image recognition method for a nucleic acid molecule is provided, including the following steps:

[0080] In step S210, the bright field microdroplet image is obtained, and a first pixel point is determined by traversing from the edge of the bright field microdroplet image, and a second pixel point symmetrical to the first pixel point is determined.

[0081] Digital droplet PCR is the third generation of digital PCR technology, which is a method for absolute quantification of nucleic acid molecules. Through it, the copy number of target genes in the sample can be directly measured. The technology uses a digital PCR chip to dilute the nucleic acid sample and randomly distribute it into tens of thousands of water-in-oil droplets. Then the DNA sample in the microdroplet is combined with a specific fluorescent marker for PCR reaction. After the PCR reaction is completed, a CCD camera (or other camera) is used to take pictures of the microdroplets under bright field and dark field environments, respectively, to obtain bright field microdroplet images and dark field microdroplet images. Among them, the bright field microdroplet image does not emit light, and it can be directly photographed without complete darkness, so the exposure time is short, the noise is less, the brightness and contrast of the image are high, but the arrangement of the microdroplets is more dense, and the edge of the microdroplet is thicker, which leads to inaccurate identification of the microdroplet diameter in the existing algorithm.

[0082] It can be understood that after obtaining the bright field microdroplet image, the bright field microdroplet image can also be denoised and edge processed to lay a good foundation for subsequent image processing.

[0083] In this embodiment, the method is applied to the server in Figure 1 The server can control the camera to take pictures of the microdroplets to obtain a bright field microdroplet image of any microdroplet. After obtaining the bright field microdroplet image taken by the camera, the server can first determine each droplet in the bright field microdroplet image, and then identify each droplet. In specific implementation, multiple threads can also be used to identify multiple droplets in the bright field microdroplet image. That is, the bright field microdroplet image of the present application can refer to the image of each droplet or the image of multiple droplets. The image of each droplet or the image of multiple droplets does not include a background image, but only includes a droplet body image.

[0084] The edge of the bright field microdroplet image is traversed, specifically starting from any edge of the bright field microdroplet image, a pixel point at the edge of the bright field microdroplet image is determined, defined as a first pixel point, and then a second pixel point symmetrical to the first pixel point is determined according to the coordinates of the first pixel point.

[0085] Specifically, the microdroplet can be considered as a circle, so after determining the first pixel point, the coordinates of the first pixel point can be determined, and the gradient value of the first pixel point is calculated according to the coordinates P1(x1, y1) of the first pixel point:

[0086]

[0087] Then, other edge pixels are found, and the gradient value tanβ of each edge pixel is calculated. According to the geometric properties of a circle, if tanα=-tanβ, it means that the pixel corresponding to the gradient value tanβ is the second pixel that is symmetrical to the first pixel.

[0088] Furthermore, in this embodiment, in order to improve the search efficiency of the second pixel, the process of searching for the second pixel may include:

[0089] Search for other edge pixels in the row containing the first pixel;

[0090] Determine whether the first pixel and the searched edge pixels are symmetrical;

[0091] If the first pixel and the searched edge pixel are symmetrical, the searched edge pixel is taken as the second pixel.

[0092] Specifically, in this embodiment, when searching for the second pixel, the row (or column) where the first pixel is located is determined, such as... Figure 3 As shown, the first pixel is P. Then, in the row containing the first pixel P, other edge pixels M are searched. It is determined whether the found edge pixel M is symmetrical to the first pixel. If the found edge pixel M is symmetrical to the first pixel, then the found edge pixel M is taken as the second pixel. If the found edge pixel M is not symmetrical to the first pixel, the search for other edge pixels in the row containing the first pixel continues. If no other edge pixel symmetrical to the first pixel is found in the row containing the first pixel, the search for the first pixel P is restarted, and the above steps are repeated to find a second pixel M symmetrical to the new first pixel.

[0093] Step S220: Determine the dot and radius of the bright-field droplet image based on the first pixel and the second pixel.

[0094] According to step S210, the first pixel point P(x1, y1) and the second pixel point M(x2, y2), as well as the corresponding gradient values ​​tanα and tanβ, are determined using the following formula:

[0095] y1=tanα*x1+b1

[0096] y2=tanα*x2+b2

[0097] tanα=-tanβ

[0098] b1 and b2 can then be calculated, thus obtaining the line containing point P:

[0099] y = tanα*x + b1,

[0100] and the straight line where M is located: y = tan a * x + b2,

[0101] According to the two straight lines, the intersection point O(x0, y0) of the two straight lines is calculated as the bright-field micro-droplet image circle point, and the radius r of the bright-field micro-droplet image can be calculated according to the first pixel point P or the second pixel point M. For example, the radius is calculated by the following formula:

[0102]

[0103] Step S230, searching for the corresponding target pixel point in the first range of the preset multiple directions of the circle point.

[0104] In order to ensure the accuracy of the recognition result, the embodiment will further verify, specifically, searching for the corresponding target pixel point in the first range of the preset multiple directions of the circle point O as the center, wherein the preset multiple directions can be greater than or equal to two directions, including the left and right directions, the up and down directions, the northeast direction and the like of the circle point O, and the first range can be a range with an absolute value of the difference of the radius less than or equal to a preset threshold, that is, searching in the range of the preset threshold from the edge of the bright-field micro-droplet image. Specifically, in the searching process, the target pixel point is found in each direction in turn, for example, the first target pixel point is found in the first range of the left direction of the circle point, and then the second target pixel point is found in the first range of the right direction of the circle point, and so on, and after the target pixel point is found in the preset multiple directions, a group of data can be obtained. The group of data is used in step S240 to determine whether the bright-field micro-droplet image is a real circle.

[0105] Step S240, determining whether the bright-field micro-droplet image is a real circle according to the multiple target pixel points found and the radius of the bright-field micro-droplet image, and obtaining the target micro-droplet image of the real circle of the bright-field micro-droplet image.

[0106] Based on the radius of the multiple target pixels and the bright-field droplet image obtained in step S230, it is determined whether the bright-field droplet image is a true circle. Specifically, the distance between each target pixel in the group and the center of the circle is calculated and defined as the first distance. Then, it is determined whether the absolute value of the difference between the first distance of each group of target pixels and the corresponding radius (the bright-field droplet image where the target pixel is located) is less than or equal to a preset threshold. If the absolute value of the difference between the first distance of all target pixels in the group and the corresponding radius is less than or equal to the preset threshold, then the circle formed by the group of target pixels is determined to be a true circle. The target pixels corresponding to the bright-field droplet image that is a true circle are saved. Then, the first pixel P is reselected, and the above steps are repeated until all edge pixels on the same bright-field droplet image have been traversed. Each time the P point is selected, it may obtain the data of a corresponding circle. Then, the data of the true circle is selected from these circles to obtain the data of multiple target droplet images.

[0107] Step S250: Perform similarity verification based on the target droplet image to determine the concentric circles in the target droplet image.

[0108] Similarity verification is performed on the multiple target droplet images determined in step S240. Specifically, similar circles are determined based on the data of the multiple target droplet images. In this embodiment, whether a circle is similar is determined based on the center coordinates of the circle formed by the pixels of each target droplet image. Specifically, it is determined whether the absolute value of the difference between the centers of each target droplet image is less than 1. If there are circles whose absolute value of the difference between the centers is less than 1, the circles whose absolute value of the difference between the centers is less than 1 are determined to be concentric circles.

[0109] Step S260: Obtain the diameter of the bright-field droplet image based on the radius of the circle with the largest radius among the concentric circles.

[0110] Based on the concentric circles determined in step S250, the circle with the largest radius is selected from the concentric circles, and the diameter of the bright-field droplet image is obtained based on the radius of the circle with the largest radius, thereby realizing the recognition of the bright-field droplet image.

[0111] The bright field microdroplet image recognition method of the nucleic acid molecule determines a first pixel point by traversing from the edge of the bright field microdroplet image, finds a symmetrical edge pixel point as the first pixel point and a second pixel point, and then calculates the dot and radius of the microdroplet by using the coordinates of the symmetrical first pixel point and second pixel point. After the center and radius of the microdroplet are determined, the target pixel point is found in a plurality of preset directions of the dot with the center as the midpoint, and then it is verified whether the microdroplet image is a real circle according to the target pixel point and the radius. The first pixel point is traversed to determine the data of a plurality of bright field microdroplet images that are real circles. The microdroplet images of the real circles are subjected to similarity checking to determine concentric circles. Finally, the circle with the largest radius is selected from the concentric circles as the circle of the microdroplet image, and the diameter of the microdroplet image is obtained according to the largest radius in the concentric circles. Therefore, the bright field microdroplet images of the corresponding real circles are obtained by traversing all edge pixel points, the influence of the microdroplet edge on the calculation result is excluded, and the accuracy of the calculation result is improved.

[0112] In one embodiment, based on the above embodiment, step S210: traversing from the edge of the bright field microdroplet image to determine a first pixel point, and determining a second pixel point symmetrical to the first pixel point, comprises:

[0113] Step S211: determining a first pixel point located at the edge of the bright field microdroplet image.

[0114] Step S212: finding a second pixel point symmetrical to the first pixel point in the bright field microdroplet image according to the first pixel point.

[0115] If the second pixel point symmetrical to the first pixel point is not found in the bright field microdroplet image, return to step S211: determining a first pixel point located at the edge of the bright field microdroplet image.

[0116] If the second pixel point symmetrical to the first pixel point is found in the bright field microdroplet image, execute step S220: determining the dot and radius of the bright field microdroplet image according to the first pixel point and the second pixel point.

[0117] As an implementation, after obtaining the bright field microdroplet image, the server first determines a first pixel point at the edge of the bright field microdroplet image, and then determines other pixel points in the bright field microdroplet image. The gradient value of the first pixel point and the other pixel points is used to determine whether there is a second pixel point symmetrical to the first pixel point in the other pixel points. The way of calculating the gradient value is the same as in the previous embodiment, which will not be repeated here.

[0118] Further, as another embodiment, in order to improve the calculation efficiency, since the first pixel point is located at the edge, the point symmetrical to the first pixel point in the same circle is also likely to be located at the edge, therefore, in the present embodiment, other edge pixel points in the bright field micro-droplet image can also be determined, then the gradient value of the first pixel point is calculated and the pixel point is searched, and then the relationship between the gradient value of the first pixel point and the gradient value of the other edge pixel points is used to determine whether the second pixel point symmetrical to the first pixel point exists in the other edge pixel points.

[0119] Further, as another embodiment, in order to improve the calculation efficiency, since the first pixel point is located at the edge, the point symmetrical to the first pixel point in the same circle is also likely to be located at the edge, therefore, in the present embodiment, other edge pixel points in the bright field micro-droplet image can also be determined, then the gradient value of the first pixel point is calculated and the pixel point is searched, and then the relationship between the gradient value of the first pixel point and the gradient value of the other edge pixel points is used to determine whether the second pixel point symmetrical to the first pixel point exists in the other edge pixel points.

[0120] Searching other edge pixel points in the row where the first pixel point is located;

[0121] Determining whether the first pixel point and the searched edge pixel point are symmetrical;

[0122] If the first pixel point and the searched edge pixel point are symmetrical, the searched edge pixel point is taken as the second pixel point.

[0123] Specifically, after the first pixel point is obtained, the row (or column) where the first pixel point is located can also be determined first, then other edge pixel points are searched in the row (or column) where the first pixel point is located, after the other edge pixel points in the row where the first pixel point is located are searched, it is determined whether the pixel point is symmetrical to the first pixel point, which specifically includes:

[0124] The first gradient value of the first pixel point is calculated according to the coordinates of the first pixel point, and the second gradient value of the searched edge pixel point is calculated according to the coordinates of the searched edge pixel point;

[0125] It is determined whether the first pixel point and the searched edge pixel point are symmetrical according to the first gradient value and the second gradient value.

[0126] Specifically, the first gradient value tanα of the first pixel point and the second gradient value tanβ of the searched edge pixel point are calculated, and then it is determined whether the first pixel point and the second pixel point are symmetrical by using whether the relationship between the first gradient value of the first pixel point and the second gradient value of the other edge pixel point satisfies tanα=-tanβ.

[0127] If the second pixel point symmetrical to the first pixel point is not searched, the next first pixel point is searched in the bright field micro-droplet image, and the above steps are repeated to search the second pixel point symmetrical to the first pixel point; if the second pixel point symmetrical to the first pixel point is searched, the step of determining the circle point and the radius of the bright field micro-droplet image according to the first pixel point and the second pixel point is performed.

[0128] As an embodiment, based on the above embodiment, the step S230: finding corresponding target pixel points in a first range of a plurality of directions of the circle point as the center, comprises:

[0129] Finding the third pixel point and the fourth pixel point in the row where the circle point is located, and finding the fifth pixel point and the sixth pixel point in the column where the circle point is located, the distances between the third pixel point, the fourth pixel point, the fifth pixel point, and the sixth pixel point and the circle point are greater than or equal to a first preset distance and less than or equal to a second preset distance, wherein the difference between the radius and the first preset distance is less than or equal to a preset threshold, and the difference between the second preset distance and the radius is less than or equal to the preset threshold.

[0130] Specifically, referring to Figure 4 In this embodiment, the third pixel point C and the fourth pixel point D are found in the row where the center O is located, and the fifth pixel point A and the sixth pixel point B are found in the column where the circle point is located, the distances between the points A, B, C, and D and the center O are greater than or equal to a first preset distance M and less than or equal to a second preset distance N, wherein M-r=△ and N-r=△. For example, the first C point can be found from the position with a distance of r-△=M from the left side of the center, the first D point can be found from the position with a distance of r-△=M from the right side of the center, the first A point can be found from the position with a distance of r-△=M from the upper side of the center, and the first B point can be found from the position with a distance of r-△=M from the lower side of the center.

[0131] It should be noted that in image processing, each pixel point is actually a small square block with integer coordinates, and then the radius r in the result may not be an integer, so it is necessary to find the ABCD pixel points within a certain range of the distance radius r to ensure that the pixel points can be found (if the ABCD pixel points are found with the distance radius r, it may not be possible to find the pixel points due to the decimal radius r). The pixel points are found in multiple directions of the center to ensure that the microdroplet is a complete circle.

[0132] After finding a plurality of target pixel points, determining whether the microdroplet image is a real circle according to the plurality of target pixel points and the radius of the microdroplet image can comprise:

[0133] The distances between the third pixel point, the fourth pixel point, the fifth pixel point, and the sixth pixel point and the center are calculated.

[0134] determine whether the absolute value of the difference between the distance of the third pixel point, the fourth pixel point, the fifth pixel point, the sixth pixel point and the center of the circle and the radius is less than or equal to a preset difference value;

[0135] If the absolute value of the difference between the distance of the third pixel point, the fourth pixel point, the fifth pixel point, the sixth pixel point and the center of the circle and the radius is less than or equal to a preset difference value, it is determined that the microdroplet image is a real circle.

[0136] Specifically, according to the third pixel point, the fourth pixel point, the fifth pixel point, the sixth pixel point, and the radius, the following formula is used for judgment:

[0137] |r1-r|≤Q,|r2-r|≤Q,|r3-r|≤Q,|r4-r|≤Q,

[0138] wherein r1 is the distance between the third pixel point and the center O of the circle, r2 is the distance between the fourth pixel point and the center O of the circle, r3 is the distance between the fifth pixel point and the center O of the circle, r4 is the distance between the sixth pixel point and the center O of the circle, and Q is a preset difference value.

[0139] If the above conditions are all met, that is, the absolute value of the difference between the distance of the third pixel point C, the fourth pixel point D, the fifth pixel point A, the sixth pixel point B and the center O of the circle and the radius r is less than or equal to a preset difference value Q, it is determined that the bright field microdroplet image corresponding to ABCD is a real circle, and the microdroplet image corresponding to the real circle is the target microdroplet image.

[0140] As an embodiment, based on the above embodiment, the similarity check according to the target microdroplet image to determine the concentric circles in the target microdroplet image comprises:

[0141] calculating the difference between the centers of the circles of the target microdroplet image;

[0142] target microdroplet images with a difference between the centers of the circles less than a third preset distance are regarded as concentric circles.

[0143] Specifically, since the radius r may not be an integer, the coordinates of the pixel points where the center of the circle is located are integers, that is, although the radii are different, the centers of the circles may be the same. Therefore, in this embodiment, the process of determining the concentric circles comprises: first calculating the difference between the centers of the circles of a plurality of target microdroplet images, and if the difference between the centers of the circles is less than a third preset distance, the corresponding target microdroplet image is a concentric circle. In this embodiment, the third preset distance is set to 1, and in specific implementation, the third preset distance can be another value, for example, 0.5. It should be noted that the third preset distance is less than or equal to 1.

[0144] As an embodiment, based on the above embodiment, the diameter of the bright field microdroplet image is obtained according to the radius of the circle with the largest radius in the concentric circles, which comprises:

[0145] According to the corresponding target pixel points on the concentric circles, the radii of the concentric circles are calculated based on the symmetrical target pixel points;

[0146] The maximum value is selected as the diameter of the bright-field droplet image from the radii of the concentric circles calculated based on the symmetrical target pixel points.

[0147] Specifically, since the edge of the droplet is thick, after edge detection, discontinuous double-edge characteristics appear. However, according to biological requirements, the size of the outermost diameter of the droplet needs to be measured. Therefore, the algorithm needs to select the circle with the maximum diameter as the final detection result when obtaining the final result. The process of determining the diameter of the bright-field droplet image can include: selecting the maximum radius of the concentric circles as the radius of the bright-field droplet image, thereby obtaining the diameter of the bright-field droplet image.

[0148] As another embodiment, the process of determining the diameter of the bright-field droplet image can further include: according to all the symmetrical target pixel points on the concentric circles, for example, A and B are two symmetrical pixel points on a certain circle, which are the two outermost points, and C and D are two symmetrical pixel points, which are the two innermost points. Although they can all be detected by the algorithm to obtain a circle, the diameter of the circle detected by A and B will be larger. In order to make the diameter of the finally detected droplet closer to the biological requirements, the algorithm needs to exclude the smaller diameter circle detected by C and D, and then select the maximum value of the circle detected by A and B from all the concentric circles as the final detection result of the algorithm. It can also be considered that after the concentric circles are determined, the radii (or diameters) of the concentric circles are calculated based on all the target pixel points on the concentric circles (or diameters), and the maximum value of the radius (or diameter) is selected to obtain the diameter of the bright-field droplet image.

[0149] As an embodiment, refer to Figure 5After the server obtains the bright-field droplet image, the image can be preprocessed, including denoising, edge processing, etc. The edge pixels in the bright-field droplet image are determined, and each pixel point can be put into an edge pixel set. Then, the pixel points in the edge pixel set are iterated in turn. After a pixel point P (a first pixel point) is selected from the edge pixel set each time, another pixel point M (a second pixel point) that is symmetrical to the first pixel point P can be searched for in the row where the pixel point P is located from the bright-field droplet image. After the pixel point M that is symmetrical to the pixel point P is found, the corresponding center and radius of the circle can be calculated according to the pixel points P and M. The calculation process can refer to the above embodiments, and details are not described herein. Then, according to the calculated center and radius, the ABCD four points can be searched for in the range of [r-△, r+△] from the center in the left, right, top, and bottom directions, respectively. If the ABCD four points are found, a circle is determined (at this time, it can also be determined whether the circle is a real circle according to the four points), and the data of the circle (including the center coordinates and the radius of the circle, and the coordinates of the ABCD four points) is recorded. In a specific implementation, the data of the circle can be put into a circle list. After all the points in the edge pixel set are iterated, the data of multiple circles can be obtained. Then, the multiple circles are screened to obtain concentric circles (the circles whose center coordinates have a difference less than a third preset distance are determined as concentric circles), and finally, the largest radius in the concentric circles is selected as the detection result, and other concentric circles are excluded, so that the radius of the bright-field droplet image is determined, and finally, the diameter of the bright-field droplet image is obtained.

[0150] It should be understood that, although each step in the flowchart involved in each of the above embodiments is displayed in turn according to the indication of the arrow, these steps are not necessarily executed in turn according to the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each of the above embodiments can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or stages or steps or stages in other steps.

[0151] Based on the same inventive concept, the present application also provides a bright-field droplet image recognition device for implementing the nucleic acid molecule involved in the above method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, and therefore the specific limitations in one or more bright-field droplet image recognition device embodiments provided below can refer to the limitations of the bright-field droplet image recognition method of the nucleic acid molecule described above, and details are not described herein.

[0152] In one embodiment, asFigure 6 As shown, a bright field micro-droplet image recognition device of a nucleic acid molecule is provided, comprising:

[0153] The traversal module 610 is configured to obtain the bright field micro-droplet image, traverse from the edge of the bright field micro-droplet image, determine the first pixel point, and determine the second pixel point symmetrical to the first pixel point.

[0154] The first determination module 620 is configured to determine the circle point and the radius of the micro-droplet image according to the first pixel point and the second pixel point.

[0155] The first difference module 630 is configured to search for the corresponding target pixel point in a first range of a plurality of directions centered on the circle point.

[0156] The real circle judgment module 640 is configured to determine whether the micro-droplet image is a real circle according to the plurality of target pixel points and the radius of the micro-droplet image, and obtain a target micro-droplet image in which the micro-droplet image is a real circle.

[0157] The concentric circle determination module 650 is configured to perform similarity checking according to the target micro-droplet image, and determine the concentric circles in the target micro-droplet image.

[0158] The second determination module 660 is configured to obtain the diameter of the bright field micro-droplet image according to the radius of the circle with the largest radius in the concentric circles.

[0159] In one embodiment, the traversal module 610 is further configured to:

[0160] determine the first pixel point located at the edge of the bright field micro-droplet image;

[0161] search for the second pixel point symmetrical to the first pixel point in the bright field micro-droplet image according to the first pixel point;

[0162] if the second pixel point symmetrical to the first pixel point is not found in the bright field micro-droplet image, return to the step of determining the first pixel point located at the edge of the bright field micro-droplet image;

[0163] if the second pixel point symmetrical to the first pixel point is found in the bright field micro-droplet image, call the first determination module 620 to perform the step of determining the circle point and the radius of the bright field micro-droplet image according to the first pixel point and the second pixel point.

[0164] In one embodiment, the traversal module 610 is further configured to:

[0165] search for other edge pixel points in the row where the first pixel point is located;

[0166] determine whether the first pixel point and the searched edge pixel point are symmetrical;

[0167] if the first pixel point and the searched edge pixel point are symmetrical, take the searched edge pixel point as a second pixel point.

[0168] In one embodiment, the traversal module 610 is further configured to:

[0169] calculate a first gradient value of the first pixel point according to the coordinates of the first pixel point, and calculate a second gradient value of the searched edge pixel point according to the coordinates of the searched edge pixel point;

[0170] determine whether the first pixel point and the searched edge pixel point are symmetrical according to the first gradient value and the second gradient value.

[0171] In one embodiment, the first determination module 620 is further configured to:

[0172] calculate a linear function in which the first pixel point and the second pixel point are located according to the coordinates of the first pixel point and the second pixel point, the first gradient value and the second gradient value;

[0173] determine an intersection point of the two lines according to the linear function in which the first pixel point and the second pixel point are located, and take the intersection point as the circle point.

[0174] calculate the distance between the first pixel point or the second pixel point and the circle point to obtain the radius.

[0175] In one embodiment, the first difference module 630 is further configured to:

[0176] find a third pixel point and a fourth pixel point in the row in which the circle point is located, and find a fifth pixel point and a sixth pixel point in the column in which the circle point is located, the distances between the third pixel point, the fourth pixel point, the fifth pixel point and the sixth pixel point and the circle point are greater than or equal to a first preset distance and less than or equal to a second preset distance, wherein the difference between the radius and the first preset distance is less than or equal to a preset threshold, and the difference between the second preset distance and the radius is less than or equal to the preset threshold.

[0177] In one embodiment, the real circle determination module 640 is further configured to:

[0178] calculate the distances between the third pixel point, the fourth pixel point, the fifth pixel point and the sixth pixel point and the circle center;

[0179] determine whether the absolute values of the differences between the distances between the third pixel point, the fourth pixel point, the fifth pixel point and the sixth pixel point and the circle center and the radius are less than or equal to a preset difference value;

[0180] If the absolute values of the differences between the distances from the third pixel point, the fourth pixel point, the fifth pixel point, the sixth pixel point and the center of the circle and the radius are less than or equal to a preset difference value, it is determined that the microdroplet image is a real circle.

[0181] In one embodiment, the concentric circle determination module 650 is further configured to:

[0182] calculate the difference between the centers of the circles of the target microdroplet images;

[0183] take the target microdroplet image with the difference between the centers of the circles less than a third preset distance as a concentric circle.

[0184] In one embodiment, the second determination module 660 is further configured to:

[0185] calculate the radius of the concentric circle according to the corresponding target pixel point of the concentric circle;

[0186] select the maximum value from the calculation of the radius of the concentric circle as the diameter of the bright-field microdroplet image.

[0187] The above-mentioned modules in the nucleic acid molecule bright-field microdroplet image recognition device can be realized by software, hardware and combinations thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to call and execute the operations corresponding to the above-mentioned modules by the processor.

[0188] In one embodiment, a computer device is provided, which can be a server, and the internal structure diagram thereof can be as shown in Figure 7 The computer device includes a processor, a memory and a network interface connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device can be used to store bright-field microdroplet images, and can also store the coordinates of the first pixel point P, the coordinates of the second pixel point M, the coordinates of the center of the circle, the radius and other data in the identification process. The network interface of the computer device is configured to communicate with the external terminal through the network connection. The computer program is executed by the processor to implement a nucleic acid molecule bright-field microdroplet image recognition method.

[0189] Those skilled in the art can understand that, Figure 7The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0190] In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the bright-field microdroplet image recognition method of nucleic acid molecules of any of the above embodiments when executing the computer program.

[0191] In one embodiment, a computer readable storage medium is provided, having stored thereon a computer program, the computer program implementing the steps of the bright-field microdroplet image recognition method of nucleic acid molecules of any of the above embodiments when executed by a processor.

[0192] In one embodiment, a computer program product is provided, comprising a computer program, the computer program implementing the steps of the bright-field microdroplet image recognition method of nucleic acid molecules of any of the above embodiments when executed by a processor.

[0193] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0194] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0195] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for bright-field droplet image recognition of nucleic acid molecules, characterized in that, The method includes: Acquire the bright-field droplet image and determine the first pixel located at the edge of the bright-field droplet image; Search for other edge pixels in the row containing the first pixel; The first gradient value of the first pixel is calculated based on its coordinates, and the second gradient value of the edge pixel is calculated based on its coordinates; wherein the gradient value is the tangent of the direction angle corresponding to the pixel. Determine whether the first pixel and its searched edge pixels are symmetrical based on the first gradient value and the second gradient value; If the first pixel and the searched edge pixel are symmetrical, the searched edge pixel is taken as the second pixel; Calculate the line function containing the first pixel and the second pixel based on the coordinates of the first pixel and the second pixel, the first gradient value, and the second gradient value; The intersection point of the two lines is determined based on the line function of the first pixel and the second pixel, and the intersection point is used as the center of the circle. Calculate the distance between the first pixel or the second pixel and the center of the circle to obtain the radius; Using the center of the circle as the center, the corresponding target pixel is searched within a first range of multiple preset directions around the center of the circle; the first range is the range where the absolute value of the difference from the radius is less than or equal to a preset threshold. Based on the multiple target pixels found and the radius of the bright-field droplet image, determine whether the bright-field droplet image is a true circle, and obtain the target droplet image where the bright-field droplet image is a true circle; Based on the similarity check of the target droplet image, the concentric circles in the target droplet image are determined; The diameter of the bright-field droplet image is obtained from the radius of the circle with the largest radius among the concentric circles.

2. The method according to claim 1, characterized in that, The method further includes: If no second pixel symmetrical to the first pixel is found in the bright-field droplet image, return to step: determine the first pixel located at the edge of the bright-field droplet image.

3. The method according to claim 1, characterized in that, The step of searching for the corresponding target pixel point within a first range of multiple preset directions centered on the circle includes: Centered on the circle, find the third and fourth pixel points in the row where the circle is located, and find the fifth and sixth pixel points in the column where the circle is located. The distances of the third, fourth, fifth, and sixth pixel points from the circle center are greater than or equal to a first preset distance and less than or equal to a second preset distance. The difference between the radius and the first preset distance is less than or equal to a preset threshold, and the difference between the second preset distance and the radius is less than or equal to the preset threshold.

4. The method according to claim 3, characterized in that, The step of determining whether the droplet image is a true circle based on the found multiple target pixels and the radius of the droplet image includes: Calculate the distances from the center of the circle to the third, fourth, fifth, and sixth pixels, respectively; Determine whether the absolute value of the difference between the distance of the third pixel, the fourth pixel, the fifth pixel, and the sixth pixel and the center of the circle and the radius is less than or equal to a preset difference. If the absolute values ​​of the differences between the distances of the third, fourth, fifth, and sixth pixels and the center of the circle and the radius are all less than or equal to a preset difference, then the droplet image is determined to be a real circle.

5. A bright-field droplet image recognition device for nucleic acid molecules, characterized in that, The device includes: The traversal module is used to determine a first pixel located at the edge of the bright-field droplet image; search for other edge pixels in the row where the first pixel is located; calculate a first gradient value of the first pixel based on its coordinates, and calculate a second gradient value of the searched edge pixels based on their coordinates; wherein the gradient value is the tangent of the direction angle corresponding to the pixel; determine whether the first pixel and its searched edge pixels are symmetrical based on the first gradient value and the second gradient value; if the first pixel and its searched edge pixels are symmetrical, the searched edge pixels are taken as the second pixel. The first determining module is used to calculate the line function of the first pixel and the second pixel based on the coordinates of the first pixel and the second pixel, the first gradient value and the second gradient value; determine the intersection point of the two lines based on the line function of the first pixel and the second pixel, and take the intersection point as the center of a circle; calculate the distance between the first pixel or the second pixel and the center of the circle to obtain the radius; The first difference module is used to find the corresponding target pixel point within a first range of multiple preset directions centered on the center of the circle; the first range is the range where the absolute value of the difference from the radius is less than or equal to a preset threshold. The real circle determination module is used to determine whether the microdroplet image is a real circle based on the multiple target pixels found and the radius of the microdroplet image, and to obtain the target microdroplet image that is a real circle. The concentric circle determination module is used to perform similarity verification based on the target droplet image and determine the concentric circles in the target droplet image. The second determining module is used to obtain the diameter of the bright-field droplet image based on the radius of the circle with the largest radius among the concentric circles.

6. The apparatus according to claim 5, characterized in that, The traversal module is also used for: If no second pixel symmetrical to the first pixel is found in the bright-field droplet image, the first determining module is invoked to perform the step of determining the first pixel located at the edge of the bright-field droplet image.

7. The apparatus according to claim 5, characterized in that, The first difference module is also used for: Centered on the circle, find the third and fourth pixel points in the row where the circle is located, and find the fifth and sixth pixel points in the column where the circle is located. The distances of the third, fourth, fifth, and sixth pixel points from the circle center are greater than or equal to a first preset distance and less than or equal to a second preset distance. The difference between the radius and the first preset distance is less than or equal to a preset threshold, and the difference between the second preset distance and the radius is less than or equal to the preset threshold.

8. The apparatus according to claim 7, characterized in that, The real circle determination module is also used for: Calculate the distances from the center of the circle to the third, fourth, fifth, and sixth pixels, respectively; Determine whether the absolute value of the difference between the distance of the third pixel, the fourth pixel, the fifth pixel, and the sixth pixel and the center of the circle and the radius is less than or equal to a preset difference. If the absolute values ​​of the differences between the distances of the third, fourth, fifth, and sixth pixels and the center of the circle and the radius are all less than or equal to a preset difference, then the droplet image is determined to be a real circle.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images