Rope skipping counting method and system

Abstract

The invention provides a rope skipping counting method and system, and the method comprises the steps: collecting a visible light image and an infrared image in a rope skipping process, carrying out the preprocessing of the visible light image and the infrared image, obtaining the preprocessing data, carrying out the image fusion of the preprocessing data, obtaining image fusion data, taking the image fusion data as a training set, and obtaining a rope skipping counting result; a rope skipping action recognition model is obtained through target detection network training, and rope skipping actions are recognized and counted through the rope skipping action recognition model. According to the invention, fusion training is carried out after the visible light image and the infrared image are preprocessed, so that the robustness and the precision of rope skipping action detection and identification are improved, and the accuracy of rope skipping counting is improved to a great extent.

Description

Technical Field

[0001] The invention relates to the field of sports detection, and in particular to a rope skipping counting method; in addition, the invention also relates to a rope skipping counting system. Background Art

[0002] In the field of autonomous movement and sports competitions, accurate identification and tracking of skipping ropes is crucial for accurate skipping counts and evaluation of sports performance. However, traditional skipping rope detection methods often do not work well in complex environments, such as under different lighting conditions, resulting in poor accuracy in detecting skipping movements and high skipping rope counting errors. Summary of the invention

[0003] In order to solve the problems existing in the prior art, at least one embodiment of the present invention provides a rope skipping counting method, which greatly improves the accuracy of rope skipping action detection in different environments, thereby greatly improving the accuracy of rope skipping counting. To this end, at least one embodiment of the present invention also provides a rope skipping counting system.

[0004] In a first aspect, an embodiment of the present invention provides a rope skipping counting method, comprising: Collect visible light images and infrared images during rope skipping; Preprocessing the visible light image and the infrared image to obtain preprocessed data; Performing image fusion on the preprocessed data to obtain image fusion data; The image fusion data is used as a training set, and the rope skipping action recognition model is obtained through object detection network training; The rope skipping action recognition model is used to identify and count rope skipping actions.

[0005] In some embodiments, the present invention provides a rope skipping counting method, wherein preprocessing the visible light image and the infrared image to obtain preprocessed data includes: Perform image alignment, denoising, and contrast enhancement on visible light and infrared images.

[0006] In some embodiments, the present invention provides a method for counting rope skipping, wherein image fusion is performed on preprocessed data to obtain image fusion data, including: Convolutional neural networks are used to perform parallel feature extraction on visible light images and infrared images, and the features of visible light images are spliced ​​with the features of infrared images.

[0007] In some embodiments, a rope skipping counting method provided by the present invention includes performing parallel feature extraction on visible light images and infrared images through a convolutional neural network, including: The visible light image and infrared image are decomposed into multiple scales respectively, and features are extracted and merged at each scale.

[0008] In some embodiments, the present invention provides a method for counting rope skipping, wherein feature extraction and merging are performed at each scale, including: Feature enhancement is performed at each scale via a diffusion model.

[0009] In some embodiments, a rope skipping counting method provided by the present invention uses image fusion data as a training set, and obtains a rope skipping action recognition model through target detection network training, including: During the training process, the training parameters are optimized by dynamically adjusting the gradient.

[0010] In a second aspect, an embodiment of the present invention further provides a rope skipping counting system, comprising: An acquisition module, used for acquiring visible light images and infrared images during rope skipping; A preprocessing module, used for preprocessing visible light images and infrared images to obtain preprocessed data; A fusion module, used for performing image fusion on the preprocessed data to obtain image fusion data; The training module is used to use the image fusion data as a training set and obtain a rope skipping action recognition model through object detection network training; The recognition module is used to recognize and count the rope skipping action through the rope skipping action recognition model.

[0011] In a third aspect, an embodiment of the present invention further provides a rope skipping counting device, comprising at least one processor; a memory coupled to the at least one processor, the memory storing executable instructions, which when executed by the at least one processor implement the steps of any one of the methods of the first aspect above.

[0012] In a fourth aspect, an embodiment of the present invention further provides a chip for executing the steps of the method in the first aspect. Specifically, the chip includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip is used to execute the steps of the method in the first aspect.

[0013] In a fifth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any one of the methods in the first aspect above are implemented.

[0014] It can be seen that a skipping rope counting method and system according to an embodiment of the present invention improves the robustness and accuracy of skipping rope action detection and recognition by preprocessing visible light images and infrared images and performing fusion training, thereby greatly improving the accuracy of skipping rope counting. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] Figure 1 Shown is a flow chart of a rope skipping counting method according to an embodiment of the present invention; Figure 2 Shown is a schematic diagram of the framework of a rope skipping counting system in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. In this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0019] [Example 1] The inventor of this solution has found that in the prior art, in the field of autonomous sports and sports competitions, accurately identifying and tracking the skipping rope is crucial for accurately counting skipping ropes and evaluating sports performance. However, traditional skipping rope detection methods often do not work well in complex environments, such as under different lighting conditions, resulting in poor accuracy in detecting skipping rope movements and a high error rate in skipping rope counting. Embodiment 1 of the present invention provides the following solution: like Figure 1As shown, this embodiment provides a rope skipping counting method, the method comprising the following steps: Step 101, collect visible light images and infrared images during rope skipping. It should be noted that the visible light images and infrared images are collected synchronously in time.

[0020] In some embodiments, the visible light images during rope skipping are collected using a digital camera, a smart phone, or an industrial camera. During the collection process, the visible light camera uses a lens system to focus the light in the rope skipping scene onto the sensor to generate an image. The infrared images during rope skipping are collected using an infrared camera. The infrared camera can be a thermal imaging camera or a near-infrared camera. During the collection process, the thermal imaging camera detects the infrared radiation released during rope skipping and converts it into an image.

[0021] Step 102: pre-process the visible light image and the infrared image to obtain pre-processed data.

[0022] In some embodiments, image alignment, noise removal, and contrast enhancement are performed on the visible light image and the infrared image.

[0023] It should be noted that when aligning visible light images and infrared images, geometric correction and registration are performed on the two images to ensure that their perspectives and scales are consistent to ensure the effectiveness of subsequent fusion. When denoising visible light images and infrared images, denoising algorithms such as Gaussian filtering or homogenization filtering can be used to improve image quality. When enhancing the contrast of visible light images and infrared images, the contrast of the image can be enhanced by methods such as histogram equalization or gamma correction.

[0024] Step 103: performing image fusion on the preprocessed data to obtain image fusion data.

[0025] In some embodiments, a convolutional neural network is used to perform parallel feature extraction on the visible light image and the infrared image, and the features of the visible light image are spliced ​​with the features of the infrared image.

[0026] It should be noted that the visible light image and the infrared image are weighted averaged and their respective brightness information is combined. The wavelet transform technology is used to decompose and reconstruct the image, and the information of different frequencies is merged to improve the details of the image. The features extracted from the visible light image and the infrared image in parallel through the convolutional neural network can be edges, textures, color histograms, etc. In addition, the SIFT algorithm (scale-invariant feature transformation algorithm) or the SURF algorithm improved by the SIFT algorithm can also be used to extract features from the visible light image and the infrared image in parallel.

[0027] In some embodiments, multi-scale decomposition is performed on the visible light image and the infrared image respectively, and feature extraction and merging are performed at each scale.

[0028] It should be noted that by performing multi-scale decomposition on visible light images and infrared images, feature information at different levels is extracted to better understand the structure and content of the image. Specifically, the image is decomposed at multiple levels through wavelet transform, and key features such as edges, textures, and color histograms are extracted at each scale. The features extracted at each scale are combined to form the final feature representation to enhance the overall feature description capability.

[0029] In some embodiments, feature enhancement is performed at each scale via a diffusion model.

[0030] It should be noted that the diffusion model is an image enhancement algorithm based on partial differential equations. It smoothes the pixel values ​​in visible light images and infrared images to make the edges and details in visible light images and infrared images more obvious. The grayscale value of the pixel is adjusted according to the difference between pixels to make the details of the image more obvious. Specifically, when using the diffusion model for image generation, some noise is added to an initial image, and then in the simulation process, the value of each pixel in the image is continuously adjusted according to the partial differential equation so that it gradually tends to the final image.

[0031] Step 104, using the image fusion data as a training set, and obtaining a rope skipping action recognition model through target detection network training.

[0032] It should be noted that the target detection network can use YOLO, such as YOLOv3, YOLOv4, YOLOv5 or YOLOv7. Divide the image fusion data into training set, validation set and test set, and prepare the file structure required by YOLO, such as configuration file, dataset path, etc. Load the pre-trained model as a basis to speed up training and improve accuracy, such as the model trained by the COCO dataset. Adjust the parameters of YOLO, select appropriate parameters such as learning rate, batch size, etc., and start the training process. Based on target detection, real-time tracking of rope skipping and athletes is performed through Kalman filtering, SORT, Deep SORT, etc.

[0033] In some embodiments, the image fusion loss and the target detection loss are represented by a loss function, a predetermined weight factor is used to balance the importance of the task, the loss function is optimized by a gradient descent method, and the parameters of the image fusion and target detection networks are updated simultaneously, as represented by the following formula 1: I will , θd = arg min Oh Lu ( u , Φ( x , y ; I will)) + (1 − oh ) Ld ( d , Ψ( x , y ; θd ))+ S ( i * ) Formula 1; in, I will and θd Represent the parameters of image fusion and target detection networks respectively, oh is a predefined weight factor used to balance task training; S( ⋅ ) To jointly optimize the constraints of the two tasks; Lu ( u , Φ( x , y ; I will )) is the first loss function, indicating the output u With the model Φ (parameters are I will ) the difference between the predicted values ​​generated; Ld ( d , Ψ( x , y ; i d )) is the second loss function, which represents the output d With the model Ψ (parameters are θd ) to find the difference between the predicted values ​​generated by

[0034] It should be noted that by setting the weight oh , a trade-off can be made between the losses of the two tasks. For example, if oh The larger the value, the more attention is paid to the first loss. Lu ; if 1− oh The larger the value, the more attention is paid to the second loss. Ld .

[0035] In some embodiments, training parameters are optimized by dynamically adjusting gradients during training.

[0036] It should be noted that the gradient of each task is calculated and the gradient matrix is ​​constructed. The gradient matrix is ​​analyzed by singular value decomposition to determine the main gradient direction. The gradient direction is adjusted so that the gradients of different tasks are orthogonal to each other to reduce conflicts. The adjusted gradient is used to update the network parameters to optimize the performance of multi-task learning.

[0037] Step 105, identifying the rope skipping action through the rope skipping action recognition model and counting.

[0038] In some embodiments, the detection results and tracking paths are drawn on the image using the OpenCV library, and the bounding box of the skipping rope and its tracking trajectory are marked, thereby visualizing the detection and tracking and improving the counting accuracy.

[0039] [Example 2] like Figure 2 As shown, this embodiment provides a rope skipping counting system, comprising: The acquisition module 201 is used to acquire visible light images and infrared images during the rope skipping process. It should be noted that the acquisition of visible light images and infrared images is synchronized in time.

[0040] In some embodiments, the visible light images during rope skipping are collected using a digital camera, a smart phone, or an industrial camera. During the collection process, the visible light camera uses a lens system to focus the light in the rope skipping scene onto the sensor to generate an image. The infrared images during rope skipping are collected using an infrared camera. The infrared camera can be a thermal imaging camera or a near-infrared camera. During the collection process, the thermal imaging camera detects the infrared radiation released during rope skipping and converts it into an image.

[0041] The preprocessing module 202 is used to preprocess the visible light image and the infrared image to obtain preprocessed data.

[0042] In some embodiments, image alignment, noise removal, and contrast enhancement are performed on the visible light image and the infrared image.

[0043] It should be noted that when aligning visible light images and infrared images, geometric correction and registration are performed on the two images to ensure that their perspectives and scales are consistent to ensure the effectiveness of subsequent fusion. When denoising visible light images and infrared images, denoising algorithms such as Gaussian filtering or homogenization filtering can be used to improve image quality. When enhancing the contrast of visible light images and infrared images, the contrast of the image can be enhanced by methods such as histogram equalization or gamma correction.

[0044] The fusion module 203 is used to perform image fusion on the pre-processed data to obtain image fusion data.

[0045] In some embodiments, a convolutional neural network is used to perform parallel feature extraction on the visible light image and the infrared image, and the features of the visible light image are spliced ​​with the features of the infrared image.

[0046] It should be noted that the visible light image and the infrared image are weighted averaged and their respective brightness information is combined. The wavelet transform technology is used to decompose and reconstruct the image, and the information of different frequencies is merged to improve the details of the image. The features extracted from the visible light image and the infrared image in parallel through the convolutional neural network can be edges, textures, color histograms, etc. In addition, the SIFT algorithm (scale-invariant feature transformation algorithm) or the SURF algorithm improved by the SIFT algorithm can also be used to extract features from the visible light image and the infrared image in parallel.

[0047] In some embodiments, multi-scale decomposition is performed on the visible light image and the infrared image respectively, and feature extraction and merging are performed at each scale.

[0048] It should be noted that by performing multi-scale decomposition on visible light images and infrared images, feature information at different levels is extracted to better understand the structure and content of the image. Specifically, the image is decomposed at multiple levels through wavelet transform, and key features such as edges, textures, and color histograms are extracted at each scale. The features extracted at each scale are combined to form the final feature representation to enhance the overall feature description capability.

[0049] In some embodiments, feature enhancement is performed at each scale via a diffusion model.

[0050] It should be noted that the diffusion model is an image enhancement algorithm based on partial differential equations. It smoothes the pixel values ​​in visible light images and infrared images to make the edges and details in visible light images and infrared images more obvious. The grayscale value of the pixel is adjusted according to the difference between pixels to make the details of the image more obvious. Specifically, when using the diffusion model for image generation, some noise is added to an initial image, and then in the simulation process, the value of each pixel in the image is continuously adjusted according to the partial differential equation so that it gradually tends to the final image.

[0051] The training module 204 is used to use the image fusion data as a training set and obtain a rope skipping action recognition model through target detection network training.

[0052] It should be noted that the target detection network can use YOLO, such as YOLOv3, YOLOv4, YOLOv5 or YOLOv7. Divide the image fusion data into training set, validation set and test set, and prepare the file structure required by YOLO, such as configuration file, dataset path, etc. Load the pre-trained model as a basis to speed up training and improve accuracy, such as the model trained by the COCO dataset. Adjust the parameters of YOLO, select appropriate parameters such as learning rate, batch size, etc., and start the training process. Based on target detection, real-time tracking of rope skipping and athletes is performed through Kalman filtering, SORT, Deep SORT, etc.

[0053] In some embodiments, the image fusion loss and the target detection loss are represented by a loss function, a predetermined weight factor is used to balance the importance of the task, the loss function is optimized by a gradient descent method, and the parameters of the image fusion and target detection networks are updated simultaneously, as represented by the following formula 1: I will , θd = arg min Oh Lu ( u , Φ( x , y ; I will )) + (1 − oh ) Ld ( d , Ψ( x , y ; θd ))+ S ( i * ) Formula 1; in, I will and θd Represent the parameters of image fusion and target detection networks respectively, oh is a predefined weight factor used to balance task training; S( ⋅ ) To jointly optimize the constraints of the two tasks; Lu ( u , Φ( x , y ; I will )) is the first loss function, indicating the output u With the model Φ (parameters are I will ) the difference between the predicted values ​​generated; Ld ( d , Ψ( x , y ; i d )) is the second loss function, which represents the output dWith the model Ψ (parameters are θd ) to find the difference between the predicted values ​​generated by

[0054] It should be noted that by setting the weight oh , a trade-off can be made between the losses of the two tasks. For example, if oh The larger the value, the more attention is paid to the first loss. Lu ; if 1− oh The larger the value, the more attention is paid to the second loss. Ld .

[0055] In some embodiments, training parameters are optimized by dynamically adjusting gradients during training.

[0056] It should be noted that the gradient of each task is calculated and the gradient matrix is ​​constructed. The gradient matrix is ​​analyzed by singular value decomposition to determine the main gradient direction. The gradient direction is adjusted so that the gradients of different tasks are orthogonal to each other to reduce conflicts. The adjusted gradient is used to update the network parameters to optimize the performance of multi-task learning.

[0057] The recognition module 205 is used to recognize and count the rope skipping action through the rope skipping action recognition model.

[0058] In some embodiments, the detection results and tracking paths are drawn on the image using the OpenCV library, and the bounding box of the skipping rope and its tracking trajectory are marked, thereby visualizing the detection and tracking and improving the counting accuracy.

[0059] [Example 3] This embodiment provides a rope skipping counting device, comprising: At least one processor; a memory coupled to the at least one processor, the memory storing executable instructions, wherein the executable instructions, when executed by the at least one processor, implement the method steps of embodiment 1 of the present invention.

[0060] In a rope skipping counting device provided by an embodiment of the present invention, a processor and a memory may be provided separately or integrated together.

[0061] For example, the memory may include a random access memory, a flash memory, a read-only memory, a programmable read-only memory, a non-volatile memory or a register, etc. The processor may be a central processing unit (CPU), etc., or a graphic processing unit (GPU). The memory may store executable instructions. The processor may execute the executable instructions stored in the memory to implement the various processes described herein.

[0062] It can be understood that the memory in this embodiment can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a ROM (Read-Only Memory), a PROM (Programmable ROM), an EPROM (Erasable PROM), an EEPROM (Electrically EPROM), or a flash memory. The volatile memory can be a RAM (Random Access Memory), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as SRAM (Static RAM), DRAM (Dynamic RAM), SDRAM (Synchronous DRAM), DDR SDRAM (Double Data Rate SDRAM), ESDRAM (Enhanced SDRAM), SLDRAM (Synchlink DRAM), and DRRAM (Direct Rambus RAM). The memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.

[0063] In some embodiments, the memory stores the following elements, upgrade packages, executable units or data structures, or a subset thereof, or an extended set thereof: an operating system and an application program.

[0064] The operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application program includes various application programs, which are used to implement various application services. The program for implementing the method of the embodiment of the present invention can be included in the application program.

[0065] In the embodiment of the present invention, the processor calls a program or instruction stored in the memory, specifically, a program or instruction stored in an application program, and the processor is used to execute the method steps of the embodiment 1 of the present invention.

[0066] [Example 4] This embodiment provides a chip for executing the method of the above-mentioned embodiment 1 of the present invention. Specifically, the chip includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip is used to execute the method of the embodiment 1 of the present invention.

[0067] [Example 5] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method of Embodiment 1 of the present invention are implemented.

[0068] For example, machine-readable storage media may include, but are not limited to, various known and unknown types of non-volatile memory.

[0069] In summary, embodiments 1-5 of the present invention provide a rope skipping counting method and system, which improves the robustness and accuracy of rope skipping action detection and recognition by preprocessing visible light images and infrared images and performing fusion training, thereby greatly improving the accuracy of rope skipping counting.

[0070] It will be appreciated by those skilled in the art that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0071] In the embodiments of the present application, the disclosed systems, devices and methods can be implemented in other ways. For example, the division of units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system. In addition, the coupling between the various units can be direct coupling or indirect coupling. In addition, the various functional units in the embodiments of the present application can be integrated into a processing unit, or can be separate physical existences, etc.

[0072] It should be understood that in the various embodiments of the present application, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0073] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a machine-readable storage medium. Therefore, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a machine-readable storage medium, which can include several instructions to enable an electronic device to perform all or part of the technical solution described in the embodiments of the present application. The above-mentioned storage medium may include various media that can store program codes, such as ROM, RAM, removable disk, hard disk, magnetic disk or optical disk.

[0074] The above contents are only specific implementation methods of the present application, and the protection scope of the present application is not limited thereto. Those skilled in the art may make changes or substitutions within the technical scope disclosed in the present application, and these changes or substitutions shall be within the protection scope of the present application.

Claims

1. A method for counting rope skipping, characterized in that: include: Collect visible light images and infrared images during rope skipping; Preprocessing the visible light image and the infrared image to obtain preprocessed data; Performing image fusion on the preprocessed data to obtain image fusion data; The image fusion data is used as a training set, and a rope skipping action recognition model is obtained through target detection network training; The rope skipping action is recognized and counted by the rope skipping action recognition model.

2. The rope skipping counting method according to claim 1, characterized in that: The preprocessing of the visible light image and the infrared image to obtain preprocessed data comprises: The visible light image and the infrared image are aligned, denoised and contrast enhanced.

3. The rope skipping counting method according to claim 1, characterized in that: The performing image fusion on the pre-processed data to obtain image fusion data comprises: The visible light image and the infrared image are subjected to parallel feature extraction through a convolutional neural network, and the features of the visible light image and the features of the infrared image are spliced.

4. The rope skipping counting method according to claim 3, characterized in that: The parallel feature extraction of the visible light image and the infrared image by using a convolutional neural network comprises: The visible light image and the infrared image are respectively decomposed at multiple scales, and features are extracted and merged at each scale.

5. The rope skipping counting method according to claim 4, characterized in that: The feature extraction and merging at each scale includes: Feature enhancement is performed at each scale via a diffusion model.

6. The rope skipping counting method according to claim 1, characterized in that: The image fusion data is used as a training set, and a rope skipping action recognition model is obtained by training a target detection network, including: During the training process, the training parameters are optimized by dynamically adjusting the gradient.

7. A skipping rope counting system, characterized in that: include: An acquisition module, used for acquiring visible light images and infrared images during rope skipping; A preprocessing module, used for preprocessing the visible light image and the infrared image to obtain preprocessed data; A fusion module, used for performing image fusion on the preprocessed data to obtain image fusion data; A training module, used for taking the image fusion data as a training set and obtaining a rope skipping action recognition model through target detection network training; The recognition module is used to recognize and count the rope skipping action through the rope skipping action recognition model.

8. A skipping rope counting device, comprising at least one processor; a memory coupled to the at least one processor, the memory storing executable instructions, characterized in that: The executable instructions, when executed by the at least one processor, enable the steps of the method according to any one of claims 1 to 6 to be implemented.

9. A chip, characterized in that: It comprises a processor, which is used to call and run a computer program from a memory, so that a device equipped with the chip executes the steps of the method as claimed in any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images