Sperm Tracking Method, Device, Electronic Device and Storage Medium

Abstract

An embodiment of the present invention discloses a sperm tracking method, device, electronic device, and storage medium. The method includes: acquiring multi-sperm images of multiple consecutive frames, and respectively determining the key point data recognition results of the multi-sperm images of the consecutive frames; for any frame of multi-sperm image, based on the position region data and region pixel data of the current frame multi-sperm image, determining the part key points of the sperm belonging to the same sperm in the current frame multi-sperm image; determining the head angles of each sperm in each consecutive frame multi-sperm image, and based on the position region data and head angle of the current sperm in the current frame multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame multi-sperm image, determining the sperm position of the current sperm in the adjacent frame multi-sperm image, so as to realize the trajectory tracking of multiple sperm within a preset time; improving the accuracy of determining sperm trajectories and sperm activity analysis.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of computer technology, and in particular, to a sperm tracking method, device, electronic device, and storage medium. Background Art

[0002] With the influence of factors such as environmental pollution and life pressure, the incidence of male infertility has been increasing year by year. Semen quality analysis is an important examination to understand male fertility, and sperm motility is the main indicator of sperm movement ability. Existing medical clinical tests use different sample preparation and microscopic detection methods to detect sperm concentration, motility, and morphology respectively.

[0003] The motility of sperm can be completed manually or with computer assistance. Using computer-aided sperm analysis technology can provide quantitative parameters of sperm motility with high accuracy, which is beneficial to the standardization and standardization of semen examination. Computer-aided sperm analysis technology uses a low-power microscopic imaging system, image acquisition, processing, and analysis system to continuously record sperm positions, and quantitatively statistics and analyze sperm motility and concentration, etc. Therefore, it is necessary to propose a more reliable and accurate sperm tracking method. Summary of the Invention

[0004] The present invention provides a sperm tracking method, device, electronic device, and storage medium to improve the accuracy of determining the continuous trajectory of sperm, thereby improving the accuracy of sperm activity analysis.

[0005] In a first aspect, embodiments of the present invention provide a sperm tracking method, which includes:

[0006] Obtain multi-sperm images of multiple consecutive frames, and respectively determine the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition results include the position region data of the region where the part key points of at least one part are located and the region pixel data of the region where the part key points are located;

[0007] For any frame of multi-sperm image, based on the position region data and region pixel data of the current frame multi-sperm image, determine the part key points belonging to the same sperm in the current frame multi-sperm image;

[0008] Determine the head angles of each sperm in each consecutive frame multi-sperm image, and based on the position region data and head angle of the current sperm in the current frame multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame multi-sperm image, determine the sperm position of the current sperm in the adjacent frame multi-sperm image.

[0009] Optionally, the step of respectively determining the key point data recognition results of the multi-sperm images of the consecutive frames includes:

[0010] For each frame of multi-sperm image in the continuous-frame multi-sperm images, perform image preprocessing on the current frame of multi-sperm image, and input the preprocessed multi-sperm image into a pre-trained sperm key-point data recognition model to obtain the key-point data recognition result of the current frame of multi-sperm image output by the sperm key-point data recognition model.

[0011] Optionally, after respectively determining the key-point data recognition results of the continuous-frame multi-sperm images, it further includes:

[0012] Generate a position region image corresponding to the position region data in the key-point recognition result, and perform image scaling processing on the position region image so that the resolution of the scaled position region image is the same as that of the input image of the sperm key-point data recognition model.

[0013] Optionally, the position region data where the part key-points are located includes the position coordinates within the position region and the position probability data of the current part key-points;

[0014] Correspondingly, after respectively determining the key-point data recognition results of the continuous-frame multi-sperm images, it further includes:

[0015] Obtain a preset key-point position threshold, and determine the key-point positions of each of the part key-points in the multi-sperm image based on the position probability data in the position region data and the key-point position threshold.

[0016] Optionally, the parts of the key-points include the head vertex and the head tail point of the sperm;

[0017] Correspondingly, determining the part key-points belonging to the same sperm in the current frame of multi-sperm image based on the position region data and the region pixel data of the current frame of multi-sperm image includes:

[0018] Based on the position region data, determine the first data distance between the vertex position data of the head vertex and the tail point position data of the head tail point, and based on the region pixel data, determine the second data distance between the vertex pixel data of the head vertex and the tail point pixel data of the head tail point;

[0019] Based on the first data distance, the second data distance, a preset first distance weight, and a preset second distance weight, determine the head vertex and the head tail point belonging to the same sperm in the current frame of multi-sperm image.

[0020] Optionally, the determining the sperm position of the current sperm in the adjacent frame of multi-sperm image based on the position region data, the head angle of the current sperm in the current frame of multi-sperm image, and the position region data and the head angle of each sperm in the adjacent frame of multi-sperm image includes:

[0021] Determine the third data distance between the vertex position data of the current sperm and the vertex position data of each sperm in the adjacent frame multi-sperm image, the fourth data distance between the tail point position data of the current sperm and the tail point position data of each sperm in the adjacent frame multi-sperm image, and the fifth data distance between the head angle of the current sperm and the head angle of each sperm in the adjacent frame multi-sperm image;

[0022] Based on the third data distance, the fourth data distance, the fifth data distance, a preset third distance weight, a preset fourth distance weight, and a preset fifth distance weight, determine the sperm position of the current sperm in the adjacent frame multi-sperm image.

[0023] Optionally, after determining the sperm position of the current sperm in the adjacent frame multi-sperm image, it further includes:

[0024] Based on the sperm positions of each sperm in each consecutive frame multi-sperm image, determine the sperm movement trajectories of each sperm in each consecutive frame multi-sperm image.

[0025] In a second aspect, an embodiment of the present invention further provides a sperm tracking device, and the device includes:

[0026] A key point data recognition result determination module, configured to obtain multi-sperm images of multiple consecutive frames, and respectively determine the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition result includes position region data of the region where the part key points of at least one part are located and region pixel data of the region where the part key points are located;

[0027] The same sperm key point determination module, configured to, for any frame of multi-sperm image, determine the part key points belonging to the same sperm in the current frame multi-sperm image based on the position region data and region pixel data of the current frame multi-sperm image;

[0028] A sperm tracking module, configured to determine the head angles of each sperm in each consecutive frame multi-sperm image, and based on the position region data and head angle of the current sperm in the current frame multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame multi-sperm image, determine the sperm position of the current sperm in the adjacent frame multi-sperm image.

[0029] In a third aspect, an embodiment of the present invention further provides an electronic device, and the electronic device includes:

[0030] One or more processors;

[0031] A storage device, configured to store one or more programs,

[0032] When the one or more programs are executed by the one or more processors, the one or more processors implement the sperm tracking method provided in any embodiment of the present invention.

[0033] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the sperm tracking method provided in any embodiment of the present invention.

[0034] The technical solution provided in this embodiment specifically obtains multi-sperm images of multiple consecutive frames, and respectively determines the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition results include the position region data of the region where the part key points of at least one part are located and the region pixel data of the region where the part key points are located; for any frame of multi-sperm image, based on the position region data and region pixel data of the current frame multi-sperm image, determine the part key points of the same sperm in the current frame multi-sperm image; determine the head angles of each sperm in each consecutive frame multi-sperm image, and based on the position region data and head angle of the current sperm in the current frame multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame multi-sperm image, determine the sperm position of the current sperm in the adjacent frame multi-sperm image, so as to realize multi-sperm tracking, so as to improve the accuracy of determining the continuous trajectory of sperm, and thus improve the accuracy of sperm activity analysis. Description of the Drawings

[0035] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the introduced drawings are only the drawings of a part of the embodiments to be described by the present invention, rather than all the drawings. For those of ordinary skill in the art, other drawings can be obtained according to these drawings without creative efforts.

[0036] Figure 1 is a schematic flowchart of the sperm tracking method provided in Embodiment 1 of the present invention;

[0037] Figure 2 is a schematic structural diagram of the sperm key point data recognition model provided in Embodiment 1 of the present invention;

[0038] Figure 3 is a schematic flowchart of the sperm tracking method provided in Embodiment 2 of the present invention;

[0039] Figure 4 is a schematic structural diagram of the sperm tracking device provided in Embodiment 3 of the present invention;

[0040] Figure 5 is a schematic structural diagram of the electronic device provided in Embodiment 4 of the present invention. Detailed implementation manners

[0041] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present invention, rather than limiting the present invention. In addition, it should be noted that for the sake of description, only the parts related to the present invention rather than all the structures are shown in the accompanying drawings.

[0042] Embodiment 1

[0043] Figure 1 FIG. is a flowchart of a sperm tracking method provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of analyzing sperm motility based on the trajectory of sperm. This method can be executed by a sperm tracking device, and the device can be implemented in a software and / or hardware manner.

[0044] Before introducing the technical solution of the embodiment of the present invention, an exemplary introduction to the application scenario of implementing the technical solution of this embodiment will be given first. Of course, the following application scenarios are only optional application scenarios, and this embodiment can also be implemented in other application scenarios. This embodiment does not limit the application scenarios of the implemented technical methods. Specifically, the application scenarios include: due to the influence of factors such as environmental pollution and life pressure, the incidence of male infertility has been increasing year by year. Semen quality analysis is an important examination to understand male fertility, and sperm motility is the main indicator of sperm motility. Existing medical clinical tests use different sample preparation and microscopic detection methods to detect the concentration, motility and morphology of sperm respectively.

[0045] The motility of sperm can be completed manually or with computer assistance. Using computer-aided sperm analysis technology can provide quantitative parameters of sperm motility with high accuracy, which is beneficial to the standardization and standardization of semen examination. Computer-aided sperm analysis technology uses a low-power microscopic imaging system, an image acquisition, processing and analysis system, and quantifies and analyzes the motility and concentration of sperm by continuously recording the position of sperm. Therefore, it is necessary to propose a more reliable and accurate sperm tracking method.

[0046] In current actual clinical work, a Computer Assisted Sperm Analysis (CASA) system is generally used to analyze sperm concentration and motility. However, due to commercial barriers, CASA basically only provides a human-computer interaction interface, and does not disclose much about the specific algorithms. The "black box" system has also raised doubts about its accuracy during use. Sperm real-time tracking is a basic technology for sperm motility analysis, and accurate motility analysis results depend on accurate sperm tracking. Therefore, there is an urgent need for a method to present the monitoring results to doctors in real time, enabling doctors to more clearly and intuitively observe the actual movement state of sperm and compare it with the analysis results, making the analysis results more credible.

[0047] Regarding the above technical problems, this embodiment provides a sperm real-time tracking method, which can accurately track each sperm based on the key parts and head angles of sperm, clearly display the movement trajectories of each sperm, and enable doctors to more clearly and intuitively observe the actual movement state of sperm.

[0048] Based on the above technical ideas,

[0049] The technical solution provided in this embodiment specifically obtains multi-sperm images of multiple consecutive frames, and respectively determines the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition results include the position region data of the region where the part key points of at least one part are located and the region pixel data of the region where the part key points are located; for any frame of multi-sperm image, based on the position region data and region pixel data of the current frame multi-sperm image, determine the part key points of the same sperm in the current frame multi-sperm image; determine the head angles of each sperm in each consecutive frame multi-sperm image, and based on the position region data and head angle of the current sperm in the current frame multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame multi-sperm image, determine the sperm position of the current sperm in the adjacent frame multi-sperm image, so as to achieve multi-sperm tracking, improve the accuracy of determining the continuous trajectory of sperm, and thus improve the accuracy of sperm activity analysis.

[0050] As Figure 1 shown, the method specifically includes the following steps:

[0051] S110. Obtain multi-sperm images of multiple consecutive frames, and respectively determine the key point data recognition results of the multi-sperm images of the consecutive frames.

[0052] In an embodiment of the present invention, a multi-sperm image may be an image containing multiple sperm. Obtaining a multi-sperm image may be achieved by collecting fresh semen to obtain an initial multi-sperm image. Optionally, the initial multi-sperm image may be a real-time captured image. Then, the method for obtaining the initial multi-sperm image may include: obtaining in real time the image captured by an image capturing device. The multi-sperm image may also be an image pre-collected and stored in a local database or a server database. In this case, the method for obtaining the initial multi-sperm image may further include: obtaining the image from the local database or the server database. Specifically, the method for performing image preprocessing on the initial multi-sperm image may include: dividing the pixel value of each pixel point in the multi-sperm image by 255 and performing normalization to obtain the preprocessed multi-sperm image, so as to make the recognition result of sperm key point data obtained based on this multi-sperm image more accurate.

[0053] In this embodiment, multiple consecutive frames of multi-sperm images may be obtained by continuously taking pictures of a preset area of a semen smear under an objective lens with a preset magnification using a microscope equipped with a camera device, or by recording a video using the above microscope and then extracting consecutive frames from the sperm video to obtain multiple consecutive frames of multi-sperm images.

[0054] Further, after obtaining multiple consecutive frames of multi-sperm images, key point data recognition is performed on each of the multiple consecutive frames of multi-sperm images to obtain the recognition results of the key point data of the multiple multi-sperm images.

[0055] Optionally, the method for determining the recognition result of the key point data of the consecutive frame multi-sperm images may include: for each frame of the consecutive frame multi-sperm images, performing image preprocessing on the current frame of the multi-sperm image and inputting the preprocessed multi-sperm image into a pre-trained sperm key point data recognition model to obtain the recognition result of the key point data of the current frame of the multi-sperm image output by the sperm key point data recognition model.

[0056] Specifically, after obtaining multiple consecutive frames of multi-sperm images, the pixel value of each pixel point in the multi-sperm image is divided by 255 and normalized to obtain the preprocessed multi-sperm image, and the preprocessed multi-sperm image is input into a pre-trained sperm key point data recognition model.

[0057] In this embodiment, the sperm key point data recognition model includes a pixel processing module, a feature extraction module, a feature splicing module, and an identification module. Among them, the pixel processing module is used to perform image scaling processing on the multi-sperm image. The feature extraction module includes at least one level of feature extraction sub-module, and the feature extraction sub-module of the first level is connected to the pixel processing module. Any level of feature extraction sub-module is used to take the output data of the pixel processing module or at least one upper level of feature extraction sub-module, and after performing downsampling processing with a preset stride, as the input data of the current level of feature extraction sub-module, perform feature extraction on the input data and output it. Among them, the output data of the current level of feature extraction sub-module is used to be used as the input data of the next level of feature extraction sub-module and / or the output data of the current level of feature extraction sub-module after performing downsampling with a preset stride. The feature splicing module, each level of feature extraction sub-module is respectively connected to the feature splicing module, and is used to perform upsampling with a preset stride on the output data of each level of feature extraction sub-module, and then perform feature splicing to obtain the spliced features output by the feature splicing module. The identification module is connected to the feature splicing module, and is used to determine the key point data recognition result of the current frame multi-sperm image based on the spliced features obtained by the feature splicing process.

[0058] Specifically, as Figure 2 shown, the multi-sperm image is input into the model. First, it passes through a basic module, and the basic module is a pixel processing module. Specifically, this pixel processing module can be composed of two convolutional layers with a stride of 2 and a convolution kernel of 3*3. The convolutional layer is defined as the combination of convolution operation, BN operation, and ReLU activation operation. Specifically, after passing through this pixel processing module, the multi-sperm image obtains a pixel processing image with a resolution of 1 / 4 of the original multi-sperm image.

[0059] Specifically, the pixel processing image is input into the feature extraction module. The feature extraction module in this embodiment is composed of four levels of feature extraction sub-modules, and the corresponding resolutions of each feature extraction sub-module are 1 / 4, 1 / 8, 1 / 16, and 1 / 32 of the resolution of the original multi-sperm image.

[0060] In this embodiment, the pixel processing sample image is input into the feature extraction sub-module of the first level in the feature extraction module. The output result of the first convolutional layer in the first level serves as both the input image of the second convolutional layer and, after performing downsampling with a stride of 2, as the input image of the first convolutional layer in the feature extraction sub-module of the second level. Further, the feature image output by the last convolutional layer in the first level serves as the output result of the feature extraction sub-module of the first level.

[0061] Specifically, the output result of the second convolutional layer in the feature extraction sub-module of the second level serves as the input image of the third convolutional layer. After performing downsampling with a stride of 2, it is also used as the input image of the first convolutional layer in the feature extraction sub-module of the third level after performing downsampling with a stride of 4 on the feature image output by the second convolutional layer in the first level. Further, the feature image output by the last convolutional layer in the second level serves as the output result of the feature extraction sub-module of the second level.

[0062] Specifically, the output result of the second convolutional layer in the feature extraction sub-module of the third level serves as the input image of the third convolutional layer. After performing downsampling with a stride of 2, it is also used as the input image of the first convolutional layer in the fourth level after performing downsampling with a stride of 4 on the feature image output by the third convolutional layer in the second level and downsampling with a stride of 8 on the feature sample image output by the third convolutional layer in the first level. Further, the feature image output by the last convolutional layer in the third level serves as the output result of the feature extraction sub-module of the third level.

[0063] Further, the feature image output by the last convolutional layer in the fourth level serves as the output result of the feature extraction sub-module of the fourth level.

[0064] The feature images output by the last convolutional layers in the feature extraction sub-modules of the four levels are upsampled to the corresponding resolutions and then input into the splicing model, so that the resolutions of the upsampled feature images are the same as those of the pixel processing images input into the feature extraction module. Specifically, the output result of the second level is upsampled with a stride of 2, the output result of the third level is upsampled with a stride of 4, and the output result of the fourth level is upsampled with a stride of 8. Then, the upsampled feature images are input into the splicing model for image splicing. Further, each feature sample image is subjected to a convolutional layer operation to obtain the spliced image output by the splicing model.

[0065] Specifically, the spliced image is input into the recognition model and subjected to a convolutional operation with 4 channels to obtain the recognition result of the key point data output by the recognition module.

[0066] It should be noted that in this embodiment, the number of parts corresponding to the key points to be recognized is 2. Therefore, a convolutional operation with 4 channels is required in the recognition model. If the number of parts corresponding to the key points to be recognized is other values, a convolutional operation with twice the number of channels of the number of parts is required. The number of channels is not limited in this embodiment.

[0067] The resolution of the result image in the key point data recognition result in this embodiment is (H / 4, W / 4, 4). The first two channels are respectively used to obtain the regional position data of the area where the key points of each part are located, and this position information is used to determine the position information of the key points of each part; the last two channels are used to obtain the pixel data of the key points of the part, and this pixel data is used to calculate the correlation information between the key points of each part.

[0068] It should be noted that in the process of obtaining the result image based on the sperm key point data recognition model in this embodiment, different degrees of downsampling processing are performed on the multi-sperm image, resulting in a decrease in the resolution of the result image output by the model. In order to obtain the multi-sperm key point results corresponding to the input image one by one, it is necessary to ensure the consistency of the resolution of the result image and the input image. After obtaining the position area data of the area where the key points of each part are located output by the model in the technical solution of this embodiment, a position area image corresponding to the position area data in the key point recognition result is also generated, and image scaling processing is performed on the position area image so that the resolution of the scaled position area image is the same as that of the input image of the sperm key point data recognition model.

[0069] In the embodiment of the present invention, the position area data where the key points of the part are located includes that each position coordinate in the position area is the position probability data of the current key point of the part; correspondingly, after obtaining the key point data recognition result output by the sperm key point data recognition model, it further includes: obtaining a preset key point position threshold, and determining the key point positions of the key points of each part in the multi-sperm image based on the position probability data in the position area data and the key point position threshold.

[0070] Specifically, taking the determination of the key point positions of the first part in the multi-sperm image as an example, the position region data of the region where the key points of the first part are located is obtained, that is, the position region data of the key points of the first channel in the key point data recognition result is obtained. Specifically, the position region data where the key points of the first part are located includes: the position coordinates within the position region are the position probability data of the key points of the first part. The position coordinates corresponding to the position probability data greater than the threshold t in the position probability data of the key points of the first part in this position region are determined as the key point coordinates of the first part. It is also possible to determine the key point coordinates of the second part as the position coordinates corresponding to the position probability data greater than the threshold t in the position probability data of the key points of the second part in the position region where the key points of the second part are located. It should be noted that since the number of sperm in one frame of the multi-sperm image is multiple, the number of key point coordinates of the same part in the multi-sperm image is also multiple. In this embodiment, the value of the threshold t is not limited and can be set according to the magnitude of the position probability data of each position coordinate in the region where the key points are located; exemplarily, in this embodiment, according to the historical experimental data results, t is set to 0.01.

[0071] S120. For any frame of the multi-sperm image, based on the position region data and the region pixel data of the current frame of the multi-sperm image, determine the key points of the parts belonging to the same sperm in the current frame of the multi-sperm image.

[0072] In the embodiment of the present invention, in order to track the same sperm in different frames of the multi-sperm image and thus determine the sperm movement trajectory of the sperm in different frames of the multi-sperm image, it is necessary to first determine the key points of each part belonging to the same sperm in the multi-sperm image; in other words, if the parts of the key points include the head vertex and the head tail point of the sperm head; then it is necessary to first determine the sperm head vertex and the head tail point belonging to the same sperm in the multi-sperm image, and it can also be understood as determining the association relationship between the sperm head vertex and the head tail point in each sperm in the multi-sperm image.

[0073] Optionally, the method for determining the key points of the parts belonging to the same sperm in the current frame of the multi-sperm image may include: determining the first data distance between the vertex position data of the head vertex and the tail point position data of the head tail point based on the position region data, and determining the second data distance between the vertex pixel data of the head vertex and the tail pixel data of the head tail point based on the region pixel data; determining the head vertex and the head tail point belonging to the same sperm in the current frame of the multi-sperm image based on the first data distance, the second data distance, the preset first distance weight, and the preset second distance weight.

[0074] Specifically, in this embodiment, the following solution is introduced by taking the example of determining the head and tail points of the same sperm as the current head vertex in the multi-sperm image of the current frame. Specifically, obtain the vertex position data and vertex pixel data corresponding to the current head vertex in the key point data recognition result of the multi-sperm image of the current frame, and obtain the tail point position data and tail point pixel data corresponding to each head and tail point in the multi-sperm image of the current frame. Respectively determine the first data distance Dist1 between the vertex position data corresponding to the current head vertex and the tail point position data corresponding to each head and tail point, and respectively determine the second data distance Dist2 between the vertex pixel data corresponding to the current head vertex and the tail point pixel data corresponding to each head and tail point. Further, obtain the first distance weight w1 corresponding to the first data distance Dist1, and the second distance weight w2 corresponding to the second data distance Dist2.

[0075] Further, based on the expression w1·Dist1 + w2·Dist2, determine the distance function between the current head vertex and each head and tail point, and determine the head and tail point with the minimum distance function value from the current head vertex, so as to determine the head and tail point of the same sperm as the current head vertex in the multi-sperm image of the current frame, and further realize the association between the head vertex and the head and tail point of each sperm based on the above technical solution.

[0076] S130. Determine the head angles of each sperm in each consecutive frame of multi-sperm images, and based on the position region data and head angle of the current sperm in the multi-sperm image of the current frame, and the position region data and head angle of each sperm in the multi-sperm image of the adjacent frame, determine the sperm position of the current sperm in the multi-sperm image of the adjacent frame.

[0077] In the embodiment of the present invention, the head angle of the sperm may be the angle between the sperm head of the current sperm and the horizontal line. Specifically, the method for obtaining the sperm head angle may include: measuring the angle formed between the connection line between the head vertex and the head and tail point of the sperm and the horizontal line to determine the sperm head angle.

[0078] Optionally, the method for determining the sperm position of the current sperm in the adjacent-frame multi-sperm image includes: determining a third data distance between the vertex position data of the current sperm and the vertex position data of each sperm in the adjacent-frame multi-sperm image, a fourth data distance between the tail point position data of the current sperm and the tail point position data of each sperm in the adjacent-frame multi-sperm image, and a fifth data distance between the head angle of the current sperm and the head angles of each sperm in the adjacent-frame multi-sperm image; determining the sperm position of the current sperm in the adjacent-frame multi-sperm image based on the third data distance, the fourth data distance, the fifth data distance, a preset third distance weight, a preset fourth distance weight, and a preset fifth distance weight.

[0079] Specifically, obtain the vertex position data corresponding to the head vertex of the current sperm in the current-frame multi-sperm image, the tail point position data corresponding to the head tail of the current sperm, and the sperm head angle of the current sperm, and obtain the vertex position data corresponding to the head vertices of each sperm in the adjacent-frame multi-sperm image, the tail point position data corresponding to the head tails respectively, and the head angles of each sperm.

[0080] Determine a third data distance Dist3 between the vertex position data of the current sperm and the vertex position data of each sperm in the adjacent-frame multi-sperm image, a fourth data distance Dist4 between the tail point position data of the current sperm and the tail point position data of each sperm in the adjacent-frame multi-sperm image, and calculate the head angle distance AngleDiff between the head angle of the current sperm in the current frame and the head angles of each sperm in the adjacent frame; wherein, the head angle distance between two sperm in multi-sperm images of different frames can be represented by the angle difference obtained by subtracting the head angles of the two sperm. Further, obtain the third distance weight w3 corresponding to the third data distance Dist3, the fourth distance weight w4 corresponding to the fourth data distance Dist4, and the angle weight w5 corresponding to the head angle distance AngleDiff.

[0081] Further, determine the distance function between the current sperm in the current frame and each sperm in the adjacent frame based on the expression w3·Dist3 + w4·Dist4 + w5·AngleDiff, and determine the sperm in the adjacent frame with the minimum distance function value from the current sperm in the current frame, so as to determine the position of the current sperm in the adjacent frame, thereby realizing the tracking of each sperm in multiple consecutive frames.

[0082] The technical solution provided in this embodiment specifically obtains multi-sperm images of multiple consecutive frames, and respectively determines the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition results include the position area data of the area where the part key points of at least one part are located and the area pixel data of the area where the part key points are located; for any frame of multi-sperm image, based on the position area data and area pixel data of the current frame of multi-sperm image, determine the part key points of the same sperm in the current frame of multi-sperm image; determine the head angles of each sperm in each consecutive frame of multi-sperm image, and based on the position area data and head angle of the current sperm in the current frame of multi-sperm image, and the position area data and head angle of each sperm in the adjacent frame of multi-sperm image, determine the sperm position of the current sperm in the adjacent frame of multi-sperm image, so as to realize multi-sperm tracking, so as to improve the accuracy of determining the continuous trajectory of sperm, and thus improve the accuracy of sperm activity analysis.

[0083] Embodiment 2

[0084] Figure 3 It is a flowchart of a sperm tracking method provided in Embodiment 2 of the present invention. On the basis of the above embodiments, after the step of "determining the sperm position of the current sperm in the adjacent frame of multi-sperm image", the step of "determining the sperm movement trajectory of each sperm in each consecutive frame of multi-sperm image based on the sperm positions of each sperm in each consecutive frame of multi-sperm image" is added. The explanations of the same or corresponding terms as those in the above embodiments will not be repeated here. See Figure 3 , the sperm tracking method provided in this embodiment includes:

[0085] S210. Obtain multi-sperm images of multiple consecutive frames, and respectively determine the key point data recognition results of the multi-sperm images of the consecutive frames.

[0086] S220. For any frame of multi-sperm image, based on the position area data and area pixel data of the current frame of multi-sperm image, determine the part key points of the same sperm in the current frame of multi-sperm image.

[0087] S230. Determine the head angles of each sperm in each consecutive frame of multi-sperm image, and based on the position area data and head angle of the current sperm in the current frame of multi-sperm image, and the position area data and head angle of each sperm in the adjacent frame of multi-sperm image, determine the sperm position of the current sperm in the adjacent frame of multi-sperm image.

[0088] S240. Determine the sperm movement trajectory of each sperm in each consecutive frame of multi-sperm image based on the sperm positions of each sperm in each consecutive frame of multi-sperm image.

[0089] In an embodiment of the present invention, based on the technical solutions of the above embodiments, the sperm positions of each sperm in each consecutive frame of multi-sperm images are determined, and the sperm positions of each sperm in each consecutive frame of multi-sperm images are respectively extracted to generate the sperm movement trajectories of each sperm in each consecutive frame of multi-sperm images, thereby realizing sperm tracking of each sperm in each consecutive frame of multi-sperm images.

[0090] Furthermore, the activity level of sperm can be analyzed according to the sperm movement speed and movement position obtained during sperm tracking.

[0091] The technical solution provided in this embodiment specifically obtains multiple consecutive frames of multi-sperm images, and respectively determines the key point data recognition results of the consecutive frame multi-sperm images; wherein, the key point data recognition results include the position area data of the area where the part key points of at least one part are located and the area pixel data of the area where the part key points are located; for any frame of multi-sperm image, based on the position area data and area pixel data of the current frame multi-sperm image, the part key points belonging to the same sperm in the current frame multi-sperm image are determined; the head angles of each sperm in each consecutive frame of multi-sperm images are determined, and based on the position area data and head angle of the current sperm in the current frame multi-sperm image, and the position area data and head angle of each sperm in the adjacent frame multi-sperm image, the sperm position of the current sperm in the adjacent frame multi-sperm image is determined, thereby realizing multi-sperm tracking, so as to improve the accuracy of determining the continuous trajectory of sperm, and thus improve the accuracy of sperm activity analysis.

[0092] The following is an embodiment of the sperm tracking device provided in the embodiment of the present invention. This device and the sperm tracking methods of the above embodiments belong to the same inventive concept. For the details not described in detail in the embodiment of the sperm tracking device, reference can be made to the embodiments of the above sperm tracking methods.

[0093] Embodiment III

[0094] Figure 4 It is a schematic structural diagram of the sperm tracking device provided in Embodiment III of the present invention. This embodiment is applicable to the situation of analyzing sperm activity based on the trajectory of sperm. See Figure 4 The specific structure of this sperm tracking device includes: a key point data recognition result determination module 310, a same sperm key point determination module 320, and a sperm tracking module 330; wherein,

[0095] The key point data recognition result determination module 310 is used to obtain multiple consecutive frames of multi-sperm images and respectively determine the key point data recognition results of the consecutive frame multi-sperm images; wherein, the key point data recognition results include the position area data of the area where the part key points of at least one part are located and the area pixel data of the area where the part key points are located;

[0096] The same sperm key point determination module 320 is configured to, for any frame of multi-sperm image, determine the part key points belonging to the same sperm in the current frame of multi-sperm image based on the position region data and region pixel data of the current frame of multi-sperm image.

[0097] The sperm tracking module 330 is configured to determine the head angles of the sperm in each consecutive frame of multi-sperm image, and determine the sperm position of the current sperm in the adjacent frame of multi-sperm image based on the position region data and head angle of the current sperm in the current frame of multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame of multi-sperm image.

[0098] The technical solution provided in this embodiment specifically obtains multiple consecutive frames of multi-sperm images, and respectively determines the key point data recognition results of the consecutive frames of multi-sperm images; wherein, the key point data recognition results include the position region data of the region where the part key points of at least one part are located and the region pixel data of the region where the part key points are located; for any frame of multi-sperm image, based on the position region data and region pixel data of the current frame of multi-sperm image, determine the part key points belonging to the same sperm in the current frame of multi-sperm image; determine the head angles of the sperm in each consecutive frame of multi-sperm image, and determine the sperm position of the current sperm in the adjacent frame of multi-sperm image based on the position region data and head angle of the current sperm in the current frame of multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame of multi-sperm image, so as to realize multi-sperm tracking, so as to improve the accuracy of determining the continuous trajectory of sperm, and thus improve the accuracy of sperm activity analysis.

[0099] On the basis of the above technical solution, the key point data recognition result determination module 310 includes:

[0100] The key point data recognition result determination unit is configured to, for each frame of multi-sperm image in the consecutive frames of multi-sperm images, perform image preprocessing on the current frame of multi-sperm image, and input the preprocessed multi-sperm image into a pre-trained sperm key point data recognition model to obtain the key point data recognition result of the current frame of multi-sperm image output by the sperm key point data recognition model.

[0101] On the basis of the above technical solution, the device further includes:

[0102] The image scaling processing unit is configured to, after recognizing the key point data recognition results of the sperm in the consecutive frames of multi-sperm images,

[0103] Generate a position region image corresponding to the position region data in the key point recognition result, and perform image scaling processing on the position region image so that the resolution of the scaled position region image is the same as that of the input image of the sperm key point data recognition model.

[0104] Based on the above technical solution, the position region data where the part key points are located includes the position coordinates within the position region, which are the position probability data of the current part key points;

[0105] Correspondingly, the device further includes:

[0106] A key point position determination unit, configured to, after respectively determining the key point data recognition results of the consecutive frame multi-sperm images, obtain a preset key point position threshold, and determine the key point positions of each of the part key points in the multi-sperm image based on the position probability data in the position region data and the key point position threshold.

[0107] Based on the above technical solution, the parts of the key points include the head vertex and the head tail point of the sperm;

[0108] Correspondingly, the same sperm key point determination module 320 includes:

[0109] A data distance determination unit, configured to determine a first data distance between the vertex position data of the head vertex and the tail point position data of the head tail point based on the position region data, and determine a second data distance between the vertex pixel data of the head vertex and the tail point pixel data of the head tail point based on the region pixel data;

[0110] A same sperm key point determination unit, configured to determine the head vertex and the head tail point of the same sperm in the current frame multi-sperm image based on the first data distance, the second data distance, a preset first distance weight, and a preset second distance weight.

[0111] Based on the above technical solution, the sperm tracking module 330 includes:

[0112] A data distance determination unit, configured to determine a third data distance between the vertex position data of the current sperm and the vertex position data of each sperm in the adjacent frame multi-sperm image respectively, a fourth data distance between the tail point position data of the current sperm and the tail point position data of each sperm in the adjacent frame multi-sperm image respectively, and a fifth data distance between the head angle of the current sperm and the head angles of each sperm in the adjacent frame multi-sperm image respectively;

[0113] A sperm position determination unit, configured to determine the sperm position of the current sperm in the multi-sperm image of adjacent frames based on the third data distance, the fourth data distance, the fifth data distance, a preset third distance weight, a preset fourth distance weight, and a preset fifth distance weight.

[0114] Based on the above technical solution, the device further includes:

[0115] A sperm movement trajectory determination unit, configured to determine the sperm movement trajectories of the sperms in the multi-sperm images of each continuous frame based on the sperm positions of the sperms in the multi-sperm images of each continuous frame after determining the position of the current sperm in the multi-sperm image of adjacent frames.

[0116] The sperm tracking device provided by the embodiments of the present invention can execute the sperm tracking method provided by any embodiment of the present invention, and has corresponding function modules and beneficial effects for executing the method.

[0117] It should be noted that in the embodiments of the above sperm tracking device, the included units and modules are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0118] Embodiment 4

[0119] Figure 5 It is a schematic structural diagram of an electronic device provided by Embodiment 4 of the present invention. Figure 5 It shows a block diagram of an exemplary electronic device 12 suitable for implementing the embodiments of the present invention. Figure 5 The shown electronic device 12 is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present invention.

[0120] As Figure 5 shown, the electronic device 12 is presented in the form of a general-purpose computing electronic device. The components of the electronic device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).

[0121] Bus 18 represents one or more of several types of bus architectures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of the various bus architectures. By way of example, these architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

[0122] Electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 12, including both volatile and nonvolatile media, removable and non-removable media.

[0123] System memory 28 can include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Electronic device 12 can further include other removable / non-removable, volatile / nonvolatile computer system storage media. By way of example only, storage system 34 can be used for reading and writing on non-removable, nonvolatile magnetic media ( Figure 5 not shown, typically called a "hard disk drive"). Although Figure 5 not shown in, a disk drive for reading and writing on removable nonvolatile disks (such as a "floppy disk"), and an optical disk drive for reading and writing on removable nonvolatile optical disks (such as a CD-ROM, DVD-ROM, or other optical media) can be provided. In these instances, each drive can be connected to bus 18 by one or more data media interfaces. System memory 28 can include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the present invention.

[0124] A program / utility 40 having a set (at least one) of program modules 42 can be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of these examples or some combination thereof may include an implementation of a network environment. Program modules 42 generally carry out the functions and / or methods of the embodiments described herein.

[0125] The electronic device 12 can also communicate with one or more external devices 14 (such as a keyboard, a pointing device, a display 24, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 12, and / or communicate with any device that enables the electronic device 12 to communicate with one or more other computing devices (such as a network card, a modem, etc.). Such communication can be carried out through an input / output (I / O) interface 22. Moreover, the electronic device 12 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 20. As Figure 5 shown, the network adapter 20 communicates with other modules of the electronic device 12 through a bus 18. It should be understood that although Figure 5 not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0126] The processing unit 16 executes various functional applications and sample data acquisition by running programs stored in the system memory 28. For example, it implements the steps of a sperm tracking method provided by an embodiment of the present invention. The sperm tracking method includes:

[0127] Obtaining multi-sperm images of multiple consecutive frames, and respectively determining the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition results include the position region data of the region where the part key points of at least one part are located and the region pixel data of the region where the part key points are located;

[0128] For any frame of multi-sperm image, based on the position region data and region pixel data of the current frame of multi-sperm image, determining the part key points of the same sperm in the current frame of multi-sperm image;

[0129] Determining the head angles of each sperm in each consecutive frame of multi-sperm images, and based on the position region data and head angle of the current sperm in the current frame of multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame of multi-sperm image, determining the sperm position of the current sperm in the adjacent frame of multi-sperm image.

[0130] Of course, those skilled in the art can understand that the processor can also implement the technical solutions of the sample data acquisition method provided by any embodiment of the present invention.

[0131] Embodiment Five

[0132] Embodiment 5 of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements, for example, the steps of a sperm tracking method provided by the present invention. The sperm tracking method includes:

[0133] Obtain multi-sperm images of multiple consecutive frames, and respectively determine the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition results include the position region data of the region where the part key points of at least one part are located and the region pixel data of the region where the part key points are located;

[0134] For any frame of multi-sperm image, based on the position region data and region pixel data of the current frame multi-sperm image, determine the part key points belonging to the same sperm in the current frame multi-sperm image;

[0135] Determine the head angles of each sperm in each consecutive frame of multi-sperm images, and based on the position region data and head angle of the current sperm in the current frame multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame multi-sperm image, determine the sperm position of the current sperm in the adjacent frame multi-sperm image.

[0136] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can 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 any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device.

[0137] The computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0138] The program code contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber cable, RF, etc., or any suitable combination of the above.

[0139] The computer program code for performing the operations of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0140] Those of ordinary skill in the art should understand that the various modules or steps of the present invention described above can be implemented using a general-purpose computing device. They can be centralized on a single computing device or distributed over a network composed of multiple computing devices. Optionally, they can be implemented using program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0141] Note that the above is only the preferred embodiment of the present invention and the technical principles applied. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein. Various obvious changes, re-adjustments, and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A sperm tracking method, characterized in that, it includes: Obtain multi-sperm images of multiple consecutive frames, and respectively determine the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition results include the position region data of the region where the part key points of at least one part are located and the region pixel data of the region where the part key points are located; For any frame of multi-sperm image, based on the position region data and region pixel data of the current frame multi-sperm image, determine the part key points belonging to the same sperm in the current frame multi-sperm image; Determine the head angles of each sperm in each consecutive frame multi-sperm image, and based on the position region data and head angle of the current sperm in the current frame multi-sperm image, and the position region data and head angle of each sperm in the adjacent frame multi-sperm image, determine the sperm position of the current sperm in the adjacent frame multi-sperm image; Among them, the position region data where the part key points are located includes that the position coordinates within the position region are the position probability data of the current part key points; After respectively determining the key point data recognition results of the multi-sperm images of the consecutive frames, it further includes: Obtain a preset key point position threshold, and determine the key point positions of each of the part key points in the multi-sperm image based on the position probability data in the position region data and the key point position threshold; The parts of the key points include the head vertex and the head tail point of the sperm; The determining the part key points belonging to the same sperm in the current frame multi-sperm image based on the position region data and region pixel data of the current frame multi-sperm image includes: Based on the position region data, determine the first data distance between the vertex position data of the head vertex and the tail point position data of the head tail point, and based on the region pixel data, determine the second data distance between the vertex pixel data of the head vertex and the tail point pixel data of the head tail point; Based on the first data distance, the second data distance, a preset first distance weight, and a preset second distance weight, determine the head vertex and the head tail point belonging to the same sperm in the current frame multi-sperm image.

2. The method according to claim 1, characterized in that, The respectively determining the key point data recognition results of the multi-sperm images of the consecutive frames includes: For each frame of multi-sperm image in the multi-sperm images of the consecutive frames, perform image preprocessing on the current frame multi-sperm image, and input the preprocessed multi-sperm image into a pre-trained sperm key point data recognition model to obtain the key point data recognition result of the current frame multi-sperm image output by the sperm key point data recognition model.

3. The method according to claim 1, characterized in that, After respectively determining the key point data recognition results of the multi-sperm images of the consecutive frames, it further includes: Generate a position region image corresponding to the position region data in the key point recognition result, and perform image scaling processing on the position region image so that the resolution of the scaled position region image is the same as that of the input image of the sperm key point data recognition model.

4. The method according to claim 1, characterized in that, Determining the sperm position of the current sperm in the multi-sperm image of the adjacent frame based on the position area data and head angle of the current sperm in the multi-sperm image of the current frame, and the position area data and head angle of each sperm in the multi-sperm image of the adjacent frame, includes: Determining a third data distance between the vertex position data of the current sperm and the vertex position data of each sperm in the multi-sperm image of the adjacent frame, a fourth data distance between the tail point position data of the current sperm and the tail point position data of each sperm in the multi-sperm image of the adjacent frame, and a fifth data distance between the head angle of the current sperm and the head angle of each sperm in the multi-sperm image of the adjacent frame; Determining the sperm position of the current sperm in the multi-sperm image of the adjacent frame based on the third data distance, the fourth data distance, the fifth data distance, a preset third distance weight, a preset fourth distance weight, and a preset fifth distance weight.

5. The method according to claim 1, wherein, after determining the sperm position of the current sperm in the multi-sperm image of the adjacent frame, further includes: Determining the sperm movement trajectories of each sperm in each consecutive multi-sperm image based on the sperm positions of each sperm in each consecutive multi-sperm image.

6. A sperm tracking device, wherein, includes: A key point data recognition result determination module, configured to obtain multi-sperm images of multiple consecutive frames, and respectively determine the key point data recognition results of the multi-sperm images of the consecutive frames; wherein, the key point data recognition result includes position area data of the position area where the part key points of at least one part are located and the area pixel data of the position area where the part key points are located; The same sperm key point determination module, configured to, for any frame of the multi-sperm image, determine the part key points of the same sperm in the multi-sperm image of the current frame based on the position area data and area pixel data of the multi-sperm image of the current frame; A sperm tracking module, configured to determine the head angles of each sperm in each consecutive multi-sperm image, and determine the sperm position of the current sperm in the multi-sperm image of the adjacent frame based on the position area data and head angle of the current sperm in the multi-sperm image of the current frame, and the position area data and head angle of each sperm in the multi-sperm image of the adjacent frame; wherein, the position area data where the part key points are located includes the position probability data that each position coordinate in the position area is the position of the current part key point; The device further includes: A key point position determination unit, configured to, after respectively determining the key point data recognition results of the multi-sperm images of the consecutive frames, obtain a preset key point position threshold, and determine the key point positions of each of the part key points in the multi-sperm image based on the position probability data in the position area data and the key point position threshold; The parts of the key points include the head vertex and head tail point of the sperm; The same sperm key point determination module includes: A data distance determination unit, configured to determine a first data distance between the vertex position data of the head vertex and the tail position data of the head tail point based on the position area data, and determine a second data distance between the vertex pixel data of the head vertex and the tail pixel data of the head tail point based on the area pixel data; A same sperm key point determination unit, configured to determine the head vertex and the head tail point belonging to the same sperm in the current frame multi-sperm image based on the first data distance, the second data distance, a preset first distance weight, and a preset second distance weight.

7. An electronic device, characterized in that, it includes: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the sperm tracking method according to any one of claims 1-5.

8. A computer-readable storage medium, having a computer program stored thereon, characterized in that, when the program is executed by a processor, it implements the sperm tracking method according to any one of claims 1-5.

Images