Plantar pressure analysis method, device, equipment, storage medium and program product

Abstract

The invention discloses a plantar pressure analysis method and device, equipment, a storage medium and a program product, and relates to the technical field of data processing.The plantar pressure analysis method comprises the steps that pressure data of a plurality of plantar areas of a target user is obtained; integrating the plurality of pressure data to obtain a pressure data matrix; inputting the pressure data matrix into a pre-trained plantar pressure analysis model, and performing head marking processing and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix; performing multi-head attention processing on the target matrix, and extracting to obtain a pressure feature vector; and obtaining a plantar pressure analysis result of the target user according to the pressure feature vector. According to the method, the pressure data of different sole areas in the target matrix can be concerned, the characteristics of the pressure data are captured from multiple angles, when the method is applied to high arch foot recognition, the sole pressure analysis result corresponds to the high arch foot recognition result, and the effect of improving the high arch foot recognition accuracy is achieved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a plantar pressure analysis method, device, equipment, storage medium and program product. Background Art

[0002] In the related art, the traditional feature extraction algorithm is usually used to extract the plantar pressure features based on the overall information of the foot pressure data to monitor and evaluate the high arch foot. The data characteristics of different plantar areas are not taken into account, resulting in a low accuracy rate for high arch foot recognition. Therefore, how to perform plantar pressure analysis and improve the accuracy rate of high arch foot recognition needs to be solved urgently. Summary of the invention

[0003] The main purpose of this application is to provide a plantar pressure analysis method, device, equipment, storage medium and program product, aiming to solve the technical problem of how to perform plantar pressure analysis and improve the accuracy of high arch foot identification.

[0004] To achieve the above purpose, the present application proposes a plantar pressure analysis method, which includes:

[0005] Obtain pressure data of multiple plantar areas of a target user;

[0006] Integrate multiple pressure data to obtain a pressure data matrix;

[0007] Input the pressure data matrix into the pre-trained plantar pressure analysis model, perform header labeling and position embedding processing on each data vector in the pressure data matrix, and obtain the target matrix;

[0008] Perform multi-head attention processing on the target matrix to extract the pressure feature vector;

[0009] According to the pressure feature vector, the plantar pressure analysis result of the target user is obtained.

[0010] In one embodiment, the step of integrating a plurality of pressure data to obtain a pressure data matrix includes:

[0011] For each pressure data, the pressure data is segmented according to the gait cycle to obtain at least one sub-pressure data;

[0012] For each gait cycle, the sub-pressure data of multiple plantar regions are integrated to obtain a sub-pressure data matrix;

[0013] At least one sub-pressure data matrix is ​​integrated to obtain a pressure data matrix.

[0014] In one embodiment, the plurality of plantar regions include a heel region, an arch region, a 1st metatarsal region, a 2nd metatarsal region, a 3rd-5th metatarsal region, a 1st phalange region, a 2nd phalange region, and a 3rd-5th phalange region.

[0015] In one embodiment, before the step of performing header marking and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix, the plantar pressure analysis method further includes:

[0016] According to the preset data dimension, the pressure data matrix is ​​dimensionally mapped to obtain an intermediate matrix;

[0017] The step of performing header marking and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix includes:

[0018] Each intermediate data vector in the intermediate matrix is ​​subjected to header marking and position embedding processing to obtain a target matrix.

[0019] In one embodiment, the step of obtaining the plantar pressure analysis result of the target user according to the feature vector includes:

[0020] The classification predictor of the plantar pressure analysis model is used to perform dimension mapping and normalization on the pressure feature vector to obtain the plantar pressure analysis result of the target user.

[0021] In one embodiment, the plantar pressure analysis method further includes:

[0022] Acquire pressure data sample sets of multiple users to obtain pressure data matrix sample sets, wherein one pressure data sample set includes pressure data samples of multiple sole areas of one user;

[0023] Construct an initial domain adversarial transfer network;

[0024] The initial domain adversarial transfer network is trained using the pressure data matrix sample set to obtain the plantar pressure analysis model.

[0025] In addition, to achieve the above purpose, the present application also proposes a plantar pressure analysis device, which includes:

[0026] An acquisition module, used to acquire pressure data of multiple sole areas of a target user;

[0027] An integration module, used to integrate multiple pressure data to obtain a pressure data matrix;

[0028] A processing module is used to input the pressure data matrix into a pre-trained plantar pressure analysis model, perform header marking and position embedding processing on each data vector in the pressure data matrix, and obtain a target matrix;

[0029] The extraction module is used to perform multi-head attention processing on the target matrix and extract the pressure feature vector;

[0030] The analysis module is used to obtain the plantar pressure analysis result of the target user according to the pressure feature vector.

[0031] In addition, to achieve the above objectives, the present application also proposes a plantar pressure analysis device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the plantar pressure analysis method as described above.

[0032] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the plantar pressure analysis method as described above are implemented.

[0033] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the plantar pressure analysis method as described above are implemented.

[0034] One or more technical solutions proposed in this application have at least the following technical effects:

[0035] The present application provides a plantar pressure analysis method, apparatus, equipment, storage medium and program product, which integrate the pressure data of different plantar areas of a target user to obtain a pressure data matrix, perform header marking processing and position embedding processing on the pressure data matrix, and after obtaining a target matrix, perform multi-head attention processing to extract a pressure feature vector and obtain a plantar pressure analysis result. Since the target matrix after header marking processing and position embedding processing is processed by a multi-head attention mechanism, the pressure data of different plantar areas in the target matrix can be focused on, the characteristics of the pressure data can be captured from multiple angles, the accuracy of the pressure feature vector is improved, and thus the accuracy of the plantar pressure analysis result is improved. When applied to high-arch foot recognition, the plantar pressure analysis result corresponds to the high-arch foot recognition result, thereby achieving the effect of improving the accuracy of high-arch foot recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 This is a flow chart of the first embodiment of the plantar pressure analysis method of the present application;

[0039] Figure 2 This is a schematic diagram of the modules of the first embodiment of the plantar pressure analysis device of the present application;

[0040] Figure 3 Schematic diagram of the device structure of the hardware operating environment involved in the plantar pressure analysis method in the embodiment of the present application.

[0041] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0042] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0043] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0044] The main solution of the embodiment of the present application is: obtain pressure data of multiple plantar areas of the target user; integrate multiple pressure data to obtain a pressure data matrix; input the pressure data matrix into a pre-trained plantar pressure analysis model, perform header labeling and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix; perform multi-head attention processing on the target matrix to extract a pressure feature vector; and obtain the plantar pressure analysis result of the target user based on the pressure feature vector.

[0045] High arches is a foot deformity characterized by an arch that is significantly higher than normal. This condition can have many effects and potential harms on the feet, lower limb joints, spine and other parts of the human body, which in turn affects the individual's quality of life. High arches are very likely to cause serious health problems if they are not intervened for a long time. Early prevention and intervention can greatly reduce the harm caused by high arches, ensure body balance and gait stability, and reduce the risk of joint inflammation and other problems in the future. Therefore, real-time monitoring and evaluation of high arches symptoms is the key to the successful prevention and treatment of high arches.

[0046] In the early stage, high arches may not have obvious symptoms. Patients may only have slight discomfort or even no subjective discomfort and ignore seeking medical treatment. In addition, high arches in children are often easily considered to be caused by the normal development stage, and thus are not given enough attention. Professional high arch foot examination and diagnosis usually involves gait analysis, arch height measurement and foot mechanics analysis, which requires experienced podiatrist to perform. The whole process also takes a lot of time and energy from both doctors and patients. In addition, high-precision arch monitoring equipment has not yet been fully popularized nationwide due to high costs and long time consumption. Therefore, it is difficult to achieve frequent detection of high arch foot symptoms during daily monitoring and reexamination. Therefore, the design of a portable high-precision arch monitoring system is one of the current research focuses in this field.

[0047] At present, the research in the field of high arch foot health monitoring is mainly focused on the development and improvement of related monitoring hardware equipment, which requires the monitoring equipment to have the functions of integrated and automated foot data collection and feature analysis. Existing equipment includes gait analysis platforms equipped with multiple high-resolution pressure sensor arrays, pressure detectors that capture foot dynamic parameters, etc. However, based on these hardware devices, related technologies usually use traditional feature extraction algorithms to extract plantar pressure features based on the overall information of foot pressure data to monitor and evaluate high arch foot symptoms, without considering the data characteristics of different plantar areas, resulting in low accuracy in high arch foot recognition.

[0048] Therefore, the present application provides a solution, which integrates the pressure data of different plantar areas of the target user to obtain a pressure data matrix, performs header marking processing and position embedding processing on the pressure data matrix, and after obtaining the target matrix, performs multi-head attention processing, extracts the pressure feature vector, and obtains the plantar pressure analysis result. Since the target matrix after header marking processing and position embedding processing is processed by the multi-head attention mechanism, the pressure data of different plantar areas in the target matrix can be paid attention to, and the characteristics of the pressure data are captured from multiple angles, which improves the accuracy of the pressure feature vector, thereby improving the accuracy of the plantar pressure analysis result. When applied to high-arch foot recognition, the plantar pressure analysis result corresponds to the high-arch foot recognition result, thereby achieving the effect of improving the accuracy of high-arch foot recognition.

[0049] It should be noted that the execution subject of this embodiment can be a computing service device with plantar pressure analysis, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a plantar pressure analysis device, etc. The following takes the plantar pressure analysis device as an example to illustrate this embodiment and the following embodiments.

[0050] Based on this, the present application embodiment provides a method for analyzing plantar pressure, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the plantar pressure analysis method of the present application.

[0051] In this embodiment, the plantar pressure analysis method may include steps S100 to S500:

[0052] Step S100, obtaining pressure data of multiple sole areas of the target user.

[0053] Step S200, integrating multiple pressure data to obtain a pressure data matrix.

[0054] Step S300: Input the pressure data matrix into a pre-trained plantar pressure analysis model, perform header labeling and position embedding processing on each data vector in the pressure data matrix, and obtain a target matrix.

[0055] Step S400, performing multi-head attention processing on the target matrix to extract the pressure feature vector.

[0056] Step S500: obtaining a foot pressure analysis result of the target user according to the pressure feature vector.

[0057] It should be noted that the target user is any user who is to undergo plantar pressure analysis, and the target task of the plantar pressure analysis may be arch type detection. The arch type detection may include high arch foot detection, and correspondingly, the target user is any user who is to undergo high arch foot detection.

[0058] The plantar region may be divided according to the task type of the target task of the plantar pressure analysis. For example, the plantar region may include an arch region, a heel region, a metatarsal region, and the like.

[0059] In this embodiment, the plurality of plantar regions may include a heel region, an arch region, a 1st metatarsal region, a 2nd metatarsal region, a 3rd to 5th metatarsal region, a 1st phalange region, a 2nd phalange region, and a 3rd to 5th phalange region.

[0060] It should be noted that for the arch type detection task, especially the high arch foot detection task, the sole area is divided into the heel area, the arch area, the first metatarsal area, the second metatarsal area, the third to fifth metatarsal areas, the first phalanx area, the second phalanx area, and the third to fifth phalanx areas. The left and right feet of the target user are divided in the same way.

[0061] It is understandable that the pressure data may include pressure signals collected by a wearable sole pressure collection device during the target user's walking process. The pressure data may include pressure signals of the target user's left foot and / or right foot. Preferably, the pressure data may include pressure signals of the target user's left foot and right foot.

[0062] Taking the target task of high arch foot detection as an example, if the pressure data includes the pressure signal of the target user's left foot, the plantar pressure analysis result can be the high arch foot detection result of the target user's left foot; if the pressure data includes the pressure signal of the target user's right foot, the plantar pressure analysis result can be the high arch foot detection result of the target user's right foot; if the pressure data includes the pressure signals of the target user's left and right feet, the plantar pressure analysis result can be the high arch foot detection result of the target user's left and right feet.

[0063] In addition, it is understandable that before inputting the pressure data into the plantar pressure analysis model for analysis, multiple pressure data need to be integrated into a pressure data matrix, wherein in the pressure data matrix, a data vector may include pressure signals of different foot regions in the same gait cycle.

[0064] Furthermore, plantar pressure analysis can be implemented through a pre-trained plantar pressure analysis model. The plantar pressure analysis model may include a feature extractor and a classification predictor. The feature extractor can extract a pressure feature vector from the pressure data features, and the classification predictor can predict the arch type of the target user based on the pressure feature vector.

[0065] Continuing with the example of the target task of high-arched foot detection, the feature extractor can extract the arch feature vector, and the classification predictor can obtain the classification probability that the target user has high-arched foot based on the arch feature vector.

[0066] In this embodiment, the feature extractor can be implemented by the Transformer framework, and the feature extractor may include a first LayerNorm layer, a linear layer, a second LayerNorm layer, a header marking layer, a PositionEmbedding layer, a Dropout layer, a Transformer encoding layer, a third LayerNorm layer and an output layer connected in sequence. In a specific implementation, after the pressure data matrix is ​​input into the feature extractor, the header marking layer can be used to perform header marking processing on each data vector in the pressure data matrix, and a classification mark is added to each data vector. Through the Position Embedding layer, the position information of the header mark of each data vector in the pressure data matrix after the header marking can be embedded, and the corresponding position vector is added to each data vector to obtain the target matrix. Then, after the calculation of the multi-head self-attention mechanism, residual connection and multi-layer perceptron modules in the Transformer encoding layer, each header in the output matrix of the Transformer encoding layer is taken out to obtain the pressure feature vector. Finally, the pressure feature vector is predicted by the classification predictor to obtain the foot pressure analysis result. Preferably, the strategy of the Position Embedding layer is 1D-Position Embedding, that is, one-dimensional position embedding. Compared with other strategies, one-dimensional position embedding requires less computational complexity and is concise and effective.

[0067] In a feasible implementation, step S200 may include: for each pressure data, segmenting the pressure data according to the gait cycle to obtain at least one sub-pressure data; for each gait cycle, integrating the sub-pressure data of multiple plantar areas to obtain a sub-pressure data matrix; integrating at least one sub-pressure data matrix to obtain a pressure data matrix.

[0068] It should be noted that the pressure data matrix may include pressure signals within one gait cycle, or may include pressure signals within multiple gait cycles. When the pressure data matrix includes pressure signals within one gait cycle, the plantar pressure analysis model may process multiple pressure data matrices to obtain corresponding multiple classification probabilities, and use the largest classification probability among the multiple classification probabilities as the plantar pressure analysis result. When the pressure data matrix includes pressure signals within multiple gait cycles, the plantar pressure analysis model may directly obtain the corresponding classification probability as the plantar pressure analysis result based on the pressure data matrix.

[0069] It is understandable that the pressure data of a plantar area collected by the wearable plantar pressure collection device may include a pressure signal within at least one gait cycle, and for each pressure signal, the pressure signal may be divided into at least one sub-pressure signal according to the gait cycle of the target user. Among them, a sub-pressure signal is a pressure signal of a plantar area within a gait cycle. For a gait cycle, the sub-pressure signals of each plantar area within the gait cycle are integrated to obtain a sub-pressure data matrix, and the number of columns of the sub-pressure data matrix may be the number of plantar areas. In the case where the pressure data matrix includes pressure signals within multiple gait cycles, the sub-pressure data matrices of multiple gait cycles are integrated to obtain a pressure data matrix, and the number of columns of the pressure data matrix may be the number of plantar areas, and the number of rows may be the number of gait cycles.

[0070] In a feasible implementation, before "performing header marking and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix", the plantar pressure analysis method may also include: performing dimension mapping on the pressure data matrix according to a preset data dimension to obtain an intermediate matrix.

[0071] Correspondingly, “performing header marking and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix” includes: performing header marking and position embedding processing on each intermediate data vector in the intermediate matrix to obtain a target matrix.

[0072] It should be noted that after the pressure data matrix is ​​input into the plantar pressure analysis model, it is necessary to first perform dimension mapping on the pressure data matrix to convert the pressure data matrix into a dimensional form followed by the internal calculation of the feature extractor that is suitable for the plantar pressure analysis model. The dimension mapping is implemented through the linear layer of the Transformer framework, and the preset data dimension can be determined according to the model scale of the plantar pressure analysis model and the purpose of the plantar pressure analysis.

[0073] Continuing with the target task of high arch foot detection as an example, if the pressure data of the target user's left foot in a gait cycle is {L1, L2...L8}, and the pressure data of the target user's right foot is {R1, R2...R8}, then the sub-pressure data matrix is ​​{L1, L2...L8, R1, R2...R8}, and the specific size can be 16×T. After dimension preset, the sub-pressure data matrix can be mapped into a matrix of size 16×512, and then the matrix is ​​subjected to header tagging and position embedding processing to obtain a 17×512 target matrix. The target matrix is ​​subjected to multi-head attention processing to extract the pressure feature vector.

[0074] In a feasible implementation manner, step S500 may include: using a classification predictor of a plantar pressure analysis model to perform dimension mapping and normalization processing on a pressure feature vector to obtain a plantar pressure analysis result of a target user.

[0075] It should be noted that the classification predictor of the plantar pressure analysis model may include a fully connected layer and a Softmax layer. The fully connected layer can perform dimension mapping on the pressure feature vector, and the Softmax layer can obtain the classification probability, i.e., the foot pressure analysis result, based on the pressure feature vector after dimension mapping. Among them, the fully connected layer of the classification predictor can perform dimension mapping on the pressure feature vector according to the purpose of the actual plantar pressure analysis, and obtain the result vector corresponding to the target dimension as the input of the Softmax layer of the classification predictor.

[0076] In a feasible implementation manner, the plantar pressure analysis method may also include: acquiring pressure data sample sets of multiple users to obtain a pressure data matrix sample set, where a pressure data sample set includes pressure data samples of multiple plantar areas of a user; constructing an initial domain adversarial transfer network; and using the pressure data matrix sample set to train the initial domain adversarial transfer network to obtain a plantar pressure analysis model.

[0077] It should be noted that the plantar pressure analysis model can be DANN (Domain-Adversarial Neural Networks). The initial domain adversarial transfer network can include a feature extractor, a classification predictor, and a domain discriminator; during the training process, the domain discriminator can be used to classify the pressure data matrix samples, form an adversarial relationship with the initial feature extractor, promote the feature extractor to extract domain invariant features, and reduce the impact of differences in foot pressure data distribution between individuals on the actual recognition results.

[0078] In the specific training process, the pressure data matrix samples are passed through the feature extractor G f After mapping, it is transformed into the feature vector corresponding to the input, and the feature vector is passed through the classification predictor G y And the domain discriminator G d The internal mapping of the sample plantar pressure analysis results and sample classification results are obtained respectively, and the sample classification label d is substituted into i , we can get the domain discriminator loss function L d , as shown in Formula 1; similarly, substitute the sample label y i The classification predictor loss function L can be obtained yAfter adding the gradient reversal layer, the total loss function of the overall initial domain adversarial transfer network is as follows: Formula 2; the final optimization goal of the total loss function E is as follows: Formula 3, which needs to minimize the loss of the classification predictor loss function and maximize the loss of the classification predictor loss function. Thus, the classification predictor G y And the domain discriminator G d There is a relationship similar to generation and discrimination between them. The former makes the extracted feature vector make the classification predictor classification effect better, and the latter makes the domain discriminator more resolvable. The two confront each other, and finally the loss function of the classification predictor and the loss function of the domain discriminator are balanced.

[0079] Formula 1 is:

[0080]

[0081] Formula 2 is:

[0082]

[0083] Formula three is:

[0084]

[0085] x i is the i-th pressure data matrix sample, θ f is the model parameter of the feature extractor, θ y are the model parameters of the classification predictor, θ d is the model parameter of the domain discriminator, λ is the weight parameter, is the prediction result of the classification predictor, is the classification result of the domain discriminator, is the target model parameter of the feature extractor, are the target model parameters of the classification predictor, are the target model parameters of the domain discriminator.

[0086] It can be understood that the domain discriminator includes a gradient reversal layer and a fully connected layer. During the training process, as shown in Formula 4, the gradient reversal layer is used to automatically reverse the gradient direction during the back propagation process, and to achieve the identity transformation during the forward propagation process, thereby achieving the confrontation between the feature extractor and the domain discriminator. Formula 4 is:

[0087] R(x)=x

[0088]

[0089] Among them, x is the input data of the gradient reversal layer, R is the function of the gradient reversal layer, and I is the identity matrix.

[0090] In addition, the fully connected layer is mainly used to extract feature mapping. The length of the mapping result is a vector of the number of individual subjects in the sample. The vector is converted into sample classification probability through the Softmax layer. Finally, the sample with the largest classification probability is selected to obtain the individual subject number in the sample.

[0091] Among them, the calculation formula of the Softmax layer is as follows:

[0092]

[0093] z j Input the j-th dimension data in the Softmax layer input vector.

[0094] Therefore, this embodiment provides a plantar pressure analysis method, which integrates the pressure data of different plantar areas of the target user to obtain a pressure data matrix, performs header marking processing and position embedding processing on the pressure data matrix, and after obtaining the target matrix, performs multi-head attention processing to extract the pressure feature vector and obtain the plantar pressure analysis result. Since the target matrix after header marking processing and position embedding processing is processed by the multi-head attention mechanism, the pressure data of different plantar areas in the target matrix can be paid attention to, and the characteristics of the pressure data are captured from multiple angles, which improves the accuracy of the pressure feature vector, thereby improving the accuracy of the plantar pressure analysis result. When applied to high-arch foot recognition, the plantar pressure analysis result corresponds to the high-arch foot recognition result, thereby achieving the effect of improving the accuracy of high-arch foot recognition.

[0095] The present application provides a plantar pressure analysis device, such as Figure 2 As shown, the plantar pressure analysis device may include:

[0096] An acquisition module 10 is used to acquire pressure data of multiple sole areas of a target user;

[0097] An integration module 20, used for integrating a plurality of pressure data to obtain a pressure data matrix;

[0098] The processing module 30 is used to input the pressure data matrix into the pre-trained plantar pressure analysis model, perform header marking and position embedding processing on each data vector in the pressure data matrix, and obtain a target matrix;

[0099] An extraction module 40 is used to perform multi-head attention processing on the target matrix to extract a pressure feature vector;

[0100] The analysis module 50 is used to obtain the plantar pressure analysis result of the target user according to the pressure feature vector.

[0101] For more implementation details of the specific implementation of the above-mentioned plantar pressure analysis device, please refer to the description of the specific implementation of the plantar pressure analysis method in the above-mentioned embodiment. For the sake of brevity of the specification, they will not be repeated here.

[0102] The present application provides a plantar pressure analysis device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the plantar pressure analysis method in the above-mentioned embodiment one.

[0103] Reference below Figure 3 , which shows a schematic diagram of the structure of a plantar pressure analysis device suitable for implementing the embodiment of the present application. The plantar pressure analysis device in the embodiment of the present application may include but is not limited to mobile terminals such as laptop computers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions: tablet computers), etc., and fixed terminals such as desktop computers, etc. Figure 3 The plantar pressure analysis device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0104] like Figure 3As shown, the plantar pressure analysis device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the plantar pressure analysis device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the plantar pressure analysis device to communicate with other devices wirelessly or by wire to exchange data. Although the plantar pressure analysis device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0105] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0106] The plantar pressure analysis device provided by the present application adopts the plantar pressure analysis method in the above embodiment, which can solve the technical problem of how to perform plantar pressure analysis and improve the accuracy of high arch foot identification. Compared with the relevant technology, the beneficial effects of the plantar pressure analysis device provided by the present application are the same as the beneficial effects of the plantar pressure analysis method provided by the above embodiment, and the other technical features in the plantar pressure analysis device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0107] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0108] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0109] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the plantar pressure analysis method in the above-mentioned embodiment.

[0110] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with 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 embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0111] The computer-readable storage medium may be included in the plantar pressure analysis device; or may exist independently without being assembled into the plantar pressure analysis device.

[0112] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the plantar pressure analysis device, the plantar pressure analysis device: obtains pressure data of multiple plantar areas of the target user; integrates multiple pressure data to obtain a pressure data matrix; inputs the pressure data matrix into a pre-trained plantar pressure analysis model, performs header labeling and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix; performs multi-head attention processing on the target matrix to extract a pressure feature vector; and obtains the plantar pressure analysis result of the target user based on the pressure feature vector.

[0113] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate 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 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 may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0114] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0115] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0116] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned plantar pressure analysis method, and can solve the technical problem of how to perform plantar pressure analysis and improve the accuracy of high arch foot identification. Compared with the related art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the plantar pressure analysis method provided in the above-mentioned embodiment, and will not be repeated here.

[0117] The present application also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the plantar pressure analysis method as described above are implemented.

[0118] The computer program product provided by the present application can solve the technical problem of how to perform plantar pressure analysis and improve the accuracy of high arch foot identification. Compared with the related art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the plantar pressure analysis method provided by the above embodiment, which will not be repeated here.

[0119] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for analyzing plantar pressure, characterized in that: The plantar pressure analysis method comprises: Obtain pressure data of multiple plantar areas of a target user; Integrate the plurality of pressure data to obtain a pressure data matrix; Input the pressure data matrix into a pre-trained plantar pressure analysis model, perform header labeling and position embedding processing on each data vector in the pressure data matrix, and obtain a target matrix; Performing multi-head attention processing on the target matrix to extract a pressure feature vector; A plantar pressure analysis result of the target user is obtained according to the pressure feature vector.

2. The plantar pressure analysis method according to claim 1, characterized in that: The step of integrating the plurality of pressure data to obtain a pressure data matrix comprises: For each of the pressure data, segment the pressure data according to the gait cycle to obtain at least one sub-pressure data; For each of the gait cycles, integrating the sub-pressure data of the plurality of plantar regions to obtain a sub-pressure data matrix; At least one of the sub-pressure data matrices is integrated to obtain the pressure data matrix.

3. The plantar pressure analysis method according to claim 1, characterized in that: The plurality of plantar regions include a heel region, an arch region, a 1st metatarsal region, a 2nd metatarsal region, a 3rd to 5th metatarsal region, a 1st phalange region, a 2nd phalange region, and a 3rd to 5th phalange region.

4. The plantar pressure analysis method according to claim 1, characterized in that: Before the step of performing header marking and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix, the plantar pressure analysis method further includes: According to the preset data dimension, dimension mapping is performed on the pressure data matrix to obtain an intermediate matrix; The step of performing header marking and position embedding processing on each data vector in the pressure data matrix to obtain a target matrix includes: Header marking processing and position embedding processing are performed on each intermediate data vector in the intermediate matrix to obtain a target matrix.

5. The plantar pressure analysis method according to claim 1, characterized in that: The step of obtaining the plantar pressure analysis result of the target user according to the feature vector comprises: The classification predictor of the plantar pressure analysis model is used to perform dimension mapping and normalization processing on the pressure feature vector to obtain the plantar pressure analysis result of the target user.

6. The plantar pressure analysis method according to any one of claims 1 to 5, characterized in that: The plantar pressure analysis method further comprises: Acquire pressure data sample sets of multiple users to obtain a pressure data matrix sample set, wherein one pressure data sample set includes pressure data samples of multiple sole areas of one user; Construct an initial domain adversarial transfer network; The initial domain adversarial transfer network is trained using the pressure data matrix sample set to obtain the plantar pressure analysis model.

7. A plantar pressure analysis device, characterized in that: The plantar pressure analysis device comprises: An acquisition module, used to acquire pressure data of multiple sole areas of a target user; An integration module, used for integrating the plurality of pressure data to obtain a pressure data matrix; A processing module, used for inputting the pressure data matrix into a pre-trained plantar pressure analysis model, performing header marking and position embedding processing on each data vector in the pressure data matrix, and obtaining a target matrix; An extraction module, used for performing multi-head attention processing on the target matrix to extract a pressure feature vector; The analysis module is used to obtain the plantar pressure analysis result of the target user according to the pressure feature vector.

8. A plantar pressure analysis device, characterized in that: The plantar pressure analysis device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, and is configured to implement the steps of the plantar pressure analysis method according to any one of claims 1 to 6 through the computer program.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the plantar pressure analysis method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the plantar pressure analysis method according to any one of claims 1 to 6 are implemented.

Images