AI-driving-based auxiliary diagnosis method, device, equipment, medium and product for gynecological 3D detection mirror

Abstract

The invention relates to an AI drive-based auxiliary diagnosis method, device, equipment, medium and product for a gynecological 3D detection scope, and belongs to the technical field of image processing. The method comprises the following steps: receiving image data about a target object output by an endoscope; inputting the image data into a preset task deep learning model to obtain first feature information of focus features of the target object; performing filling processing on the lesion feature through a preset adaptive region-of-interest model to obtain second feature information of the processed lesion feature; determining third feature information about the lesion feature in a target object case associated with the target object; and according to a difference value reflected by the third feature information and the second feature information, determining a variable quantity of a focus feature of the target object.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the technical field of image processing, and more specifically, to an AI-driven 3D gynecological detection mirror assisted diagnosis method, device, equipment, medium, and product. Background Art

[0002] With the rapid development of image processing technology, the early screening and accurate diagnosis of gynecological diseases increasingly rely on artificial intelligence. Currently, the lesion characteristics of gynecological diseases such as cervical cancer and endometriosis are relatively subtle in the early stage, and traditional manual film reading methods have problems such as strong subjectivity, low efficiency, and easy missed diagnosis. Summary of the Invention

[0003] An object of embodiments of the present disclosure is to provide a new technical solution for AI-driven 3D gynecological detection mirror assisted diagnosis.

[0004] According to a first aspect of the present disclosure, there is provided an AI-driven 3D gynecological detection mirror assisted diagnosis method, the method comprising: Receiving image data of a target object output by the endoscope; Inputting the image data into a preset task deep learning model to obtain first feature information of the lesion characteristics of the target object; Performing filling processing on the lesion characteristics through a preset adaptive region of interest model to obtain second feature information of the processed lesion characteristics; Determining third feature information of the lesion characteristics in a target object case associated with the target object; Determining a change amount of the lesion characteristics of the target object according to a difference reflected by the third feature information and the second feature information.

[0005] Optionally, the task deep learning model includes a backbone network, a classification branch network, and a localization branch network; The inputting the image data into a preset task deep learning model to obtain first feature information of the lesion characteristics of the target object includes: Inputting the image data into the backbone network to extract image features in the image data; Inputting the image features into the classification branch network to determine the lesion type of the lesion characteristics of the target object; Inputting the image features into the localization branch network to output the lesion location of the lesion characteristics of the target object, and taking the lesion type and the lesion location as the first feature information of the lesion characteristics of the target object.

[0006] Optionally, before inputting the image data into the backbone network and extracting the image features from the image data, the method further includes: Determining the task complexity of the image data through a preset gradient balance algorithm; Setting the loss weight of the task deep learning model according to the task complexity.

[0007] Optionally, the filling process of the lesion features through a preset adaptive region of interest model to obtain the second feature information of the processed lesion features includes: Determining the relative position information of the lesion features in the target object through a three-dimensional image database in the preset adaptive region of interest model; Performing a filling process on the lesion features according to the relative position information and an extraction module in the preset adaptive region of interest model to obtain the second feature information of the processed lesion features.

[0008] Optionally, after performing a filling process on the lesion features according to the relative position information and an extraction module in the preset adaptive region of interest model to obtain the second feature information of the processed lesion features, the method further includes: Obtaining the second feature information after fine segmentation through an image segmentation model in the preset adaptive region of interest model.

[0009] Optionally, the case recognition library includes object cases of multiple objects, and the target object case and the object identifier of the target object are associated and stored in the case recognition library.

[0010] According to a second aspect of the present disclosure, there is also provided an AI-driven 3D gynecological endoscope-assisted diagnosis device, the device includes: A receiving module, configured to receive image data of a target object output by the endoscope; A first obtaining module, configured to input the image data into a preset task deep learning model to obtain first feature information of the lesion features of the target object; A second obtaining module, configured to perform a filling process on the lesion features through a preset adaptive region of interest model to obtain second feature information of the processed lesion features; A first determining module, configured to determine third feature information about the lesion features in the object case associated with the target object; A second determining module, configured to determine the change amount of the lesion features of the target object according to the difference reflected by the third feature information and the second feature information.

[0011] According to a third aspect of the present disclosure, an electronic device is further provided, including a memory and a processor. The memory is used for storing a computer program; the processor is used for executing the computer program to implement the method according to the first aspect of the present disclosure.

[0012] According to a fourth aspect of the present disclosure, a computer-readable storage medium is further provided. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method according to the first aspect of the present disclosure is implemented.

[0013] According to a fifth aspect of the present disclosure, a computer program product is further provided, including a computer program, and when the computer program is executed by a processor, the method according to the first aspect of the present disclosure is implemented.

[0014] One beneficial effect of the embodiments of the present disclosure is that the AI-driven gynecological 3D detection mirror-assisted diagnosis method provided by the present invention can receive image data of the uterus of a certain patient output by an endoscope. The image data is input into a preset task deep learning model to obtain first feature information of the lesion features of the uterus. Through a preset adaptive region of interest model, the lesion features are filled to obtain second feature information of the processed lesion features. The third feature information regarding the lesion features in the target object case associated with a certain patient is determined. According to the difference reflected by the third feature information and the second feature information, the change amount of the lesion features of the uterus of a certain patient is determined. In other words, through the task deep learning model, the adaptive region of interest model, and the case recognition library, the efficiency of gynecological disease diagnosis can be improved and effective assistance can be provided.

[0015] Through the following detailed description of the exemplary embodiments of the present disclosure with reference to the accompanying drawings, other features and advantages of the embodiments of the present disclosure will become clear. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings incorporated in and constituting a part of this specification illustrate embodiments of the present disclosure and, together with the description, are used to explain the principles of the embodiments of the present disclosure.

[0017] Figure 1 is a schematic flowchart of an AI-driven gynecological 3D detection mirror-assisted diagnosis method according to an embodiment; Figure 2 is a schematic block diagram of an AI-driven gynecological 3D detection mirror-assisted diagnosis device according to an embodiment; Figure 3 is a schematic hardware structure diagram of an electronic device according to an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that: unless otherwise specifically stated, the relative arrangements of the parts and steps, numerical expressions and values set forth in these embodiments do not limit the scope of the present invention.

[0019] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way a limitation on the present invention or its application or use.

[0020] Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and devices should be regarded as part of the specification.

[0021] In all the examples shown and discussed herein, any specific values should be construed as merely exemplary and not as limitations. Accordingly, other examples of the exemplary embodiments may have different values.

[0022] It should be noted that: like reference numerals and letters denote like items in the following drawings, and thus, once an item is defined in one drawing, further discussion thereof is not required in subsequent drawings.

[0023] <Method Embodiment> Figure 1 is a schematic flowchart of an AI-driven 3D gynecological endoscope-assisted diagnosis method according to an embodiment. The implementation subject is a control terminal, and the control terminal can be communicatively connected to the endoscope.

[0024] As Figure 2 shown, the AI-driven 3D gynecological endoscope-assisted diagnosis method of this embodiment may include the following steps S110 to step S150: Step S110, receiving image data of a target object output by the endoscope.

[0025] In this embodiment, the target object may be an organ of a certain patient, such as the uterus, etc.

[0026] Step S120, inputting the image data into a preset task deep learning model to obtain first feature information of the lesion features of the target object.

[0027] In some embodiments, in order to improve the accuracy of the first feature information output by the task deep learning model, the task deep learning model includes a backbone network, a classification branch network, and a localization branch network. This step S120 may include the following steps S210 to step S230: Step S210, inputting the image data into the backbone network to extract image features in the image data.

[0028] In this embodiment, the backbone network uses MobileNetV3-Large as the feature extractor, and the number of channels is optimized (the number of channels in the last layer is compressed from 960 to 512) to effectively reduce the computational load.

[0029] Step S220: Input the image features into the classification branch network to determine the lesion type of the lesion features of the target object.

[0030] In this embodiment, this type of branch network is followed by a Transformer encoder (4 layers, number of heads = 8) after the backbone network to implement lesion type recognition (such as cervical cancer, polyps, etc.).

[0031] Step S230: Input the image features into the localization branch network, output the lesion location of the lesion features of the target object, and use the lesion type and the lesion location as the first feature information of the lesion features of the target object.

[0032] In this embodiment, this localization branch combines a lightweight FPN (Feature Pyramid Network) to output the 3D bounding box coordinates and confidence of the lesion features as the lesion location of the lesion features of the target object.

[0033] In some embodiments, in order to effectively avoid bias in the above-mentioned task deep learning model, before step S210, the method further includes the following steps S310 and S320: Step S310: Determine the task complexity of the image data through a preset gradient balancing algorithm.

[0034] In this embodiment, the gradient balancing algorithm can be the GradNorm algorithm (Gradient Normalization). The GradNorm algorithm is a gradient balancing algorithm for multi-task learning, which effectively solves the problems of large differences in gradient magnitudes and unbalanced training among different tasks in multi-task learning. Moreover, by dynamically adjusting the gradient weights of each task, all tasks can learn at a similar speed, thereby improving the overall performance to achieve the determination of the task complexity of the image data.

[0035] Step S320: Set the loss weights of the task deep learning model according to the task complexity.

[0036] In this embodiment, the task complexity can be set manually. For example, when there are more than a set number of polyps in the image data, the task complexity is complex; when there are less than a set number of polyps in the image data, the task complexity is simple.

[0037] In this embodiment, the total number of model parameters of the task deep learning model is 4.2M (occupying 50MB of memory after compression), and it has an inference speed with a single-frame processing time of ≤10ms for 4K video streams.

[0038] Step S130, through a preset adaptive region of interest model, perform filling processing on the lesion features to obtain the second feature information of the processed lesion features.

[0039] In some embodiments, this step S130 may include the following steps S410 and S420: Step S410, through the three-dimensional image database in the preset adaptive region of interest model, determine the relative position information of the lesion features in the target object.

[0040] In this embodiment, the three-dimensional image database can store 100,000 cases of cervical 3D image data, and the three-dimensional image database can be a multi-scale template library constructed based on the extraction of 20 key features such as the morphology of the cervical os and the vascular distribution.

[0041] Step S420, according to the relative position information and the extraction module in the preset adaptive region of interest model, perform filling processing on the lesion features to obtain the second feature information of the processed lesion features.

[0042] In this embodiment, the extraction module can perform initial ROI extraction on the lesion features, that is, edge detection and region screening. That is, during the edge detection process, an improved Canny operator (dual-threshold adaptive adjustment) is used, combined with morphological closing operation to fill holes, that is, perform filling processing on the lesion features. Then, in the region screening, based on the templates in the three-dimensional image database, match the lesion templates with (SSIM similarity ≥ 0.85), determine the lesion features that conform to the lesion templates in the image data and the second feature information of the lesion features.

[0043] In some embodiments, in order to improve the fineness of the obtained second feature information, after this step S420, the method further includes the following steps S510 and S520: Step S510, through the image segmentation model in the preset adaptive region of interest model, obtain the second feature information after fine segmentation.

[0044] In this embodiment, the image segmentation model in the adaptive region of interest model can be a lightweight U-Net model. This image segmentation model can perform secondary segmentation on the second feature information and support dynamic adjustment of the segmentation threshold to adapt to individual differences (such as texture changes caused by cervical erosion) to obtain the second feature information after fine segmentation.

[0045] Step S140, determining third feature information about lesion features in the target object case associated with the target object.

[0046] In some embodiments, the case identification library includes target cases of multiple objects, and the target object cases and the object identifiers of the target objects are associated and stored in the case identification library.

[0047] In this embodiment, the data source of the case identification library can be a joint operation of multiple hospitals to collect gynecological 3D endoscopic video streams (4K@30fps) and synchronous fluorescence imaging data. It includes 5 typical scenarios: normal tissue, early lesions (CIN I-II), advanced cancer (CIN III), polyps, and endometriosis. In addition, the data source can be annotated with multi-dimensional data, such as structural information, functional information, and timing information. Among them, the structural information includes the 3D bounding box of the lesion (length, width, and height), surface curvature, and depth information. Functional information includes fluorescence intensity distribution (abnormal areas are marked as highlighted). Timing information includes dynamic lesion evolution sequences (such as polyp growth trajectories).

[0048] In this embodiment, the annotation process and tools of the data sources summarized in the case identification library can be with the help of a semi-automatic annotation platform. Level 1 annotation: pre-annotation by AI (preliminary delineation of the lesion area based on the pre-trained model). Level 2 verification: cross-examination by 3 deputy chief physicians, and the annotation consistency must be ≥95% before it can be entered into the library. Dispute resolution: Resolve divergent cases through a multi-expert consultation system. In addition, it supports 3D point cloud editing, fluorescence thermogram overlay, timeline playback, and a built-in standardized terminology library (such as FIGO staging, lesion morphology classification).

[0049] The case identification database may be a time series database that stores multiple examination data of the same patient (such as cervical cancer follow-up) to support lesion evolution analysis.

[0050] Step S150, determining the change amount of the lesion feature of the target object according to the difference reflected by the third feature information and the second feature information.

[0051] In this embodiment, the information reflected by the third characteristic information and the second characteristic information is, for example, the three-dimensional coordinates of the lesion characteristics, and the difference reflected by the third characteristic information and the second characteristic information is the distance deviation of the three-dimensional coordinate points.

[0052] In this embodiment, the AI-driven 3D gynecological detection mirror-assisted diagnosis method can receive the image data of a patient's uterus output by the endoscope. The image data is input into a preset task deep learning model to obtain the first feature information of the lesion features of the uterus. Through the preset adaptive region of interest model, the lesion features are filled to obtain the second feature information of the processed lesion features. Determine the third feature information about the lesion features in the target object case associated with a certain patient. According to the difference reflected by the third feature information and the second feature information, determine the change amount of the lesion features of a certain patient's uterus. In other words, through the task deep learning model, the adaptive region of interest model, and the case recognition library, the efficiency of gynecological disease diagnosis can be improved and effective assistance can be provided.

[0053] <Device Embodiment 1> Figure 2 is a schematic block diagram of an AI-driven 3D gynecological detection mirror-assisted diagnosis device according to an embodiment. As Figure 2 shown, the AI-driven 3D gynecological detection mirror-assisted diagnosis device 200 may include: A receiving module 210 for receiving the image data of the target object output by the endoscope; A first obtaining module 220 for inputting the image data into a preset task deep learning model to obtain the first feature information of the lesion features of the target object; A second obtaining module 230 for filling the lesion features through a preset adaptive region of interest model to obtain the second feature information of the processed lesion features; A first determining module 240 for determining the third feature information about the lesion features in the object case associated with the target object; A second determining module 250 for determining the change amount of the lesion features of the target object according to the difference reflected by the third feature information and the second feature information.

[0054] In some embodiments, the first obtaining module 220 is further configured to input the image data into the backbone network to extract the image features in the image data; input the image features into the classification branch network to determine the lesion type of the lesion features of the target object; input the image features into the localization branch network to output the lesion location of the lesion features of the target object, and use the lesion type and the lesion location as the first feature information of the lesion features of the target object.

[0055] In some embodiments, the AI-driven 3D gynecological detection mirror-assisted diagnosis device 200 further includes a setting module for determining the task complexity of the image data through a preset gradient balance algorithm; setting the loss weight of the task deep learning model according to the task complexity.

[0056] In some embodiments, the second obtaining module 230 is further configured to determine the relative position information of the lesion feature in the target object through the three-dimensional image database in the preset adaptive region of interest model; and perform filling processing on the lesion feature according to the relative position information and the extraction module in the preset adaptive region of interest model, so as to obtain the second feature information of the processed lesion feature.

[0057] In some embodiments, the AI-driven gynecological 3D detection mirror-assisted diagnosis device 200 further includes an annotation module, which is configured to obtain the second feature information after fine segmentation through the image segmentation model in the preset adaptive region of interest model.

[0058] <Device Embodiment 2> Figure 3 It is a schematic hardware structure diagram of an electronic device according to another embodiment.

[0059] As Figure 3 shown, the electronic device 300 includes a processor 310 and a memory 320. The memory 320 is used to store executable computer programs, and the processor 310 is configured to execute the methods in any of the above method embodiments according to the control of the computer programs.

[0060] Each module of the above AI-driven gynecological 3D detection mirror-assisted diagnosis device 200 can be implemented by the processor 310 executing the computer program stored in the memory 320, or can be implemented by other structures, which is not limited herein.

[0061] The present invention can be a system, a method, and / or a computer program product. The computer program product may include a computer-readable storage medium having thereon computer-readable program instructions for causing a processor to implement various aspects of the present invention.

[0062] A computer-readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include: 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), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanically encoded device, such as a punched card or raised structures in grooves storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage medium as used herein is not construed to be a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., an optical pulse through an optical fiber cable), or an electrical signal transmitted through a wire.

[0063] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to respective computing / processing devices, or can be downloaded to an external computer or an external storage device through a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include a copper transmission cable, an optical fiber transmission, a wireless transmission, a router, a firewall, a switch, a gateway computer, and / or an edge server. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing / processing device.

[0064] The computer program instructions for carrying out the operations of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state-setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages - such as Smalltalk, C++, etc., and conventional procedural programming languages - such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may 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, alternatively, may be connected to an external computer (e.g., through the Internet using an Internet service provider). In some embodiments, by using the state information of the computer-readable program instructions to customize an electronic circuit, such as a programmable logic circuit, a field-programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit can execute the computer-readable program instructions to implement various aspects of the present invention.

[0065] Aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0066] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, when executed by the processor of the computer or other programmable data processing apparatus, create a means for implementing the functions / acts specified in one or more blocks of the flowchart illustrations and / or block diagrams. These computer-readable program instructions can also be stored in a computer-readable storage medium, which instructions cause a computer, a programmable data processing apparatus, and / or other devices to operate in a particular manner, such that the computer-readable medium storing the instructions comprises a manufacture, which includes instructions for implementing various aspects of the functions / acts specified in one or more blocks of the flowchart illustrations and / or block diagrams.

[0067] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices, causing a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other devices to generate a computer-implemented process such that the instructions executed on the computer, other programmable data processing apparatus, or other devices implement the functions / acts specified in one or more boxes of the flowchart and / or block diagram.

[0068] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two consecutive blocks may in fact be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block of the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or acts, or by a combination of dedicated hardware and computer instructions. As is well known to those skilled in the art, implementation by hardware, implementation by software, and implementation by a combination of software and hardware are equivalent.

[0069] The embodiments of the present invention have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein. The scope of the present invention is defined by the appended claims.

Claims

1. An AI-driven 3D gynecological detection mirror-assisted diagnosis method, characterized in that, The method includes: Receiving image data of a target object output by the endoscope; Inputting the image data into a preset task deep learning model to obtain first feature information of the lesion features of the target object; Performing filling processing on the lesion features through a preset adaptive region of interest model to obtain second feature information of the processed lesion features; Determining third feature information about the lesion features in the target object case associated with the target object; Determining the change amount of the lesion features of the target object according to the difference reflected by the third feature information and the second feature information.

2. The method according to claim 1, wherein The task deep learning model includes a backbone network, a classification branch network, and a localization branch network; The step of inputting the image data into a preset task deep learning model to obtain first feature information of the lesion features of the target object includes: Inputting the image data into the backbone network to extract image features in the image data; Inputting the image features into the classification branch network to determine the lesion type of the lesion features of the target object; Inputting the image features into the localization branch network to output the lesion location of the lesion features of the target object, and using the lesion type and the lesion location as the first feature information of the lesion features of the target object.

3. The method according to claim 1, characterized in that, The step of performing filling processing on the lesion features through a preset adaptive region of interest model to obtain second feature information of the processed lesion features includes: Determining the relative position information of the lesion features in the target object through a three-dimensional image database in the preset adaptive region of interest model; Performing filling processing on the lesion features according to the relative position information and an extraction module in the preset adaptive region of interest model to obtain second feature information of the processed lesion features.

4. The method according to claim 3, wherein After performing filling processing on the lesion features according to the relative position information and an extraction module in the preset adaptive region of interest model to obtain second feature information of the processed lesion features, the method further includes: Obtaining finely segmented second feature information through an image segmentation model in the preset adaptive region of interest model.

5. The method according to claim 1, wherein The case recognition library includes object cases of multiple objects, and the target object case is associated and stored with the object identifier of the target object in the case recognition library.

6. An AI-driven 3D gynecological detection mirror-assisted diagnosis device, characterized in that, The device includes: A receiving module, configured to receive image data of a target object output by the endoscope; A first obtaining module, configured to input the image data into a preset task deep learning model to obtain first feature information of the lesion features of the target object; A second obtaining module, configured to perform filling processing on the lesion features through a preset adaptive region of interest model to obtain second feature information of the processed lesion features; A first determining module, configured to determine third feature information about the lesion features in the object case associated with the target object; A second determination module, configured to determine a change amount of a lesion feature of the target object according to a difference reflected by the third feature information and the second feature information.

7. An electronic device, characterized in that, Comprising a memory and a processor, the memory is configured to store a computer program; the processor is configured to execute the computer program to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

9. A computer program product, characterized in that, Comprising a computer program, when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images