Human skin temperature prediction method and system fusing informer and STGCN

Abstract

The invention relates to the technical field of human body temperature prediction, and provides a human body skin temperature prediction method and system fusing an informer and STGCN. The method comprises the following steps: according to a topological structure among a plurality of measurement points of acquired skin temperature data of a target object, based on a space-time diagram convolutional network, extracting space time sequence characteristics of the measurement points, and based on a first informer module of a self-attention mechanism, respectively according to a data matrix of the skin temperature data and a data matrix of acquired environmental parameter data; respectively extracting temperature time sequence characteristics and environment time sequence characteristics of the measurement points; and then, fusing the space time sequence characteristics, the temperature time sequence characteristics and the environment time sequence characteristics of the measurement points in a second informer module based on a self-attention mechanism according to a time sequence, inputting an obtained characteristic matrix into a full-connection neural network, and predicting the skin temperature of a target measurement point.

Description

Technical Field

[0001] The present application relates to the technical field of human body temperature prediction, and in particular to a method and system for predicting human skin temperature by integrating informer and STGCN. Background Art

[0003] When the ambient temperature is too low, it is easy to cause human skin damage (for example, frostbite). In a low temperature environment, the blood vessels in the human skin are prone to spasm, causing tissue damage, thus causing frostbite; at the same time, the peripheral blood vessels of the human body will contract, causing blood blockage and increased blood pressure, leading to increased heart load, and the chances of cerebral hemorrhage and myocardial infarction will also be greatly increased. Therefore, extreme low temperature environments increase the risk of frostbite for outdoor workers, posing a major threat to human life safety and health.

[0004] Therefore, there is an urgent need to provide a technical solution to the above-mentioned deficiencies in the prior art. Summary of the invention

[0005] The purpose of this application is to provide a human skin temperature prediction method and system that integrates informer and STGCN to solve or alleviate the problems existing in the above-mentioned prior art.

[0006] In order to achieve the above objectives, this application provides the following technical solutions: The present application provides a human skin temperature prediction method integrating informer and STGCN, comprising: step S101, according to the topological structure between multiple measurement points corresponding to the skin temperature data of the target object, based on the spatiotemporal graph convolutional network, the spatial temporal characteristics of the measurement points are analyzed. Perform extraction; And, the first informer module based on the self-attention mechanism, according to the data matrix of the skin temperature data and a data matrix of environmental parameter data corresponding to the skin temperature data obtained , to respectively analyze the temperature timing characteristics of the measurement points , Environmental time series characteristics Perform extraction; Step S102: The spatial temporal features , the temperature time series characteristics , the environmental time series characteristics The features are fused in a second informer module based on a self-attention mechanism in time sequence, and the obtained feature matrix is ​​input into a fully connected neural network to predict the skin temperature of the measuring point of the target object.

[0007] Preferably, in step S101, an adjacency matrix is ​​constructed based on the topological structure between multiple measurement points corresponding to the skin temperature data of the target object. Perform normalization to obtain the normalized adjacency matrix , and based on the graph convolution operation, according to the formula:

[0008] Determine the measuring point at Skin temperature feature matrix at the moment , to construct the spatial temporal characteristics of the measurement points ; In the formula, is the ReLU activation function; For the target object The skin temperature feature matrix at the moment, is the skin temperature feature matrix The corresponding weight matrix Preferably, in step S101, according to the formula:

[0009] The temperature time series characteristics of the measurement point , Environmental time series characteristics Perform extraction; In the formula, is the data matrix of the skin temperature data, The data matrix for the skin temperature data The corresponding weight matrix of the query matrix, The data matrix for the skin temperature data The weight matrix corresponding to the bond matrix, The data matrix for the skin temperature data The weight matrix corresponding to the value matrix; The data matrix for the skin temperature data The corresponding vector dimensions in the key matrix; A data matrix of environmental parameter data corresponding to the skin temperature data; is a data matrix of environmental parameter data corresponding to the skin temperature data The corresponding weight matrix of the query matrix, is a data matrix of environmental parameter data corresponding to the skin temperature data The weight matrix corresponding to the bond matrix, is a data matrix of environmental parameter data corresponding to the skin temperature data The weight matrix corresponding to the value matrix; is a data matrix of environmental parameter data corresponding to the skin temperature data The corresponding vector dimension in the key matrix.

[0010] Preferably, in step S102, the spatial temporal characteristics of the measurement points are calculated respectively , Temperature time series characteristics , Environmental time series characteristics Pearson correlation coefficient with the skin temperature data , and according to the Pearson correlation coefficient Calculate the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The importance ratio of According to the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The importance ratio of the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics Sort by importance to build a judgment matrix ; According to the judgment matrix , respectively calculate the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The weight vector is used to calculate the spatial and temporal characteristics of the measurement points. , Temperature time series characteristics , Environmental time series characteristics Perform feature fusion to obtain the feature matrix .

[0011] Preferably, according to the formula:

[0012] Calculate the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The importance ratio of ;in, , are all positive integers, is the judgment matrix The diagonal elements in Characterizes the measurement point The time series characteristics and The importance ratio of the time series features; Characterizes the measurement point The time series characteristics and The importance ratio of the time series features.

[0013] Preferably, according to the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics Sort the importance of the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics With the judgment matrix The row elements of each row correspond one to one to determine the spatial temporal characteristics of the measurement point , Temperature time series characteristics , Environmental time series characteristics The weight vector of .

[0014] Preferably, according to the formula:

[0015] Calculate the spatial temporal characteristics of the measurement points respectively , Temperature time series characteristics , Environmental time series characteristics The weight vector of for OK The judgment matrix of columns The The weight vector of the time series features corresponding to the row elements, is the judgment matrix The Line The elements of the column, ; All are positive integers.

[0016] Preferably, according to the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The weight vector of the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics Perform weighted fusion to obtain the feature matrix of the measurement point .

[0017] The embodiment of the present application also provides a human skin temperature prediction system integrating informer and STGCN, including: The feature extraction unit is configured to extract the spatial temporal features of the measurement points based on the topological structure between the multiple measurement points corresponding to the skin temperature data of the target object obtained, based on the spatiotemporal graph convolution network. Extraction; and, based on the first informer module of the self-attention mechanism, respectively according to the data matrix of the skin temperature data and a data matrix of environmental parameter data corresponding to the skin temperature data obtained , to respectively analyze the temperature timing characteristics of the measurement points , Environmental time series characteristics Perform extraction; The temperature prediction unit is configured to convert the spatial temporal characteristics , the temperature time series characteristics , the environmental time series characteristics The features are fused in a second informer module based on a self-attention mechanism in time sequence, and the obtained feature matrix is ​​input into a fully connected neural network to predict the skin temperature of the measuring point of the target object.

[0018] Beneficial effects: The human skin temperature prediction method provided by the embodiment of the present application integrates informer and STGCN, and according to the topological structure between multiple measurement points corresponding to the skin temperature data of the acquired target object, the spatial temporal characteristics of the measurement points are analyzed based on the spatiotemporal graph convolutional network. Extract, and based on the first informer module of the self-attention mechanism, according to the data matrix of skin temperature data And the data matrix of environmental parameter data corresponding to the skin temperature data , respectively, the temperature time series characteristics of the measurement points , Environmental time series characteristics Extraction; then, the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The features are fused in the second informer module based on the self-attention mechanism in time sequence, and the obtained feature matrix is ​​input into the fully connected neural network to predict the skin temperature of the target measurement point.

[0019] Therefore, through the topological structure of multiple measurement points with physiological connection relationships, the spatial timing features, temperature timing features and environmental timing features are comprehensively extracted based on the spatiotemporal graph convolutional network and the informer module based on the self-attention mechanism, and are fused in the informer module based on the self-attention mechanism according to the time sequence, effectively establishing the spatial dependency between different measurement points, overcoming the constraints of the existing technology data quality on the model performance and the insufficient applicability of the traditional human body thermal balance equation method in dynamic scenarios, enhancing the generalization ability of skin temperature prediction, and using the obtained multi-dimensional features as the input of the fully connected neural network to provide sufficient information for subsequent skin temperature prediction, effectively improving the accuracy and robustness of temperature prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. Among them: Figure 1 A schematic diagram of a flow chart of a human skin temperature prediction method integrating informer and STGCN according to some embodiments of the present application; Figure 2 A schematic diagram of multiple measurement points of a target object provided according to some embodiments of the present application; Figure 3 for Figure 2 A schematic diagram of the topological structure corresponding to the multiple measurement points shown; Figure 4 A schematic diagram of the structure of an Informer module based on a self-attention mechanism according to some embodiments of the present application; Figure 5 A logic diagram of a human skin temperature prediction method integrating informer and STGCN provided according to some embodiments of the present application; Figure 6 A schematic diagram of the structure of a human skin temperature prediction system that integrates informer and STGCN according to some embodiments of the present application. DETAILED DESCRIPTION

[0021] The present application will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments. Each example is provided by way of explanation of the present application and does not limit the present application. In fact, it will be clear to those skilled in the art that modifications and variations may be made in the present application without departing from the scope or spirit of the present application. For example, a feature shown or described as a part of an embodiment may be used in another embodiment to produce yet another embodiment. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the embodiments of the present invention should belong to the scope of protection of the embodiments of the present invention.

[0022] In an existing technology for predicting human skin temperature, a video of the subject's skin changing over time and the true value of the skin temperature measured by a temperature sensor are collected, a feature matrix is ​​generated using video sampling and saturation extraction, and the true value of the skin temperature is interpolated to generate a label, thereby constructing a training set; then, the constructed training set is input into the improved ResNet50V2 network for training to obtain an optimized model. In the improved ResNet50V2 network, the feature outputs of the convolutional layer, residual block one, residual block two, and residual block three are extracted, and global pooling is performed respectively to form feature matrices of four feature extraction paths; at the same time, these features are spliced ​​with the feature matrix extracted from the high-level layers of the original network.

[0023] In another existing technology for predicting human skin temperature, a human thermal response data set is constructed by conducting simulation experiments in an artificial environment test chamber, which includes parameters such as skin temperature, heart rate and core temperature. Then, based on the data set, a regression model is used to invert the relationship between simulated skin temperature, simulated heart rate and simulated core temperature. At the same time, the model parameters in the Kalman filter are automatically adjusted through a deep learning algorithm to optimize and correct the measured data of the target individual; the core temperature of the target individual is accurately predicted using a long-term prediction network model.

[0024] The above-mentioned prediction technology for human skin temperature, on the one hand, mainly relies on physiological data such as skin temperature and heart rate. However, in a complex low-temperature environment, the impact of environmental factors on skin temperature cannot be ignored. However, in the existing public technology, the impact of integrated environmental factors such as environmental characteristics (such as temperature, humidity, wind speed, etc.) on skin temperature prediction is not clearly defined; on the other hand, when predicting human skin temperature, the measurement points are treated as independent individuals, and the spatial relationship between the measurement points is lacking; however, the temperature changes between various parts of the human body are interrelated, and the lack of spatial dependence in the prediction process makes the prediction results of the model inaccurate.

[0025] Based on this, the embodiment of the present application proposes a method for predicting human skin temperature by integrating informer and STGCN to overcome the problem that the data quality of the prior art has a large impact on the performance of the model and the traditional human thermal balance equation method is not applicable enough in dynamic scenarios. Through the powerful nonlinear modeling capabilities of the deep learning algorithm, it adapts to diverse input data (including data of subjects of different heights, weights, ages, genders and different environmental data), mines complex patterns and feature associations in the data, enhances the generalization ability of the model, and improves the accuracy of temperature prediction and the robustness of the model. At the same time, through multi-task learning and data enhancement technology, the prediction error caused by noise and inconsistency is reduced, and it adapts to diverse dynamic working scenarios, overcomes the limitations of traditional methods, and achieves more efficient and accurate temperature prediction.

[0026] like Figures 1 to 5 As shown, the human skin temperature prediction method integrating informer and STGCN includes: Step S101: Based on the topological structure between multiple measurement points corresponding to the skin temperature data of the target object, the spatial temporal characteristics of the measurement points are analyzed based on the spatiotemporal graph convolutional network. Extraction; and, based on the first informer module of the self-attention mechanism, respectively, according to the data matrix of the skin temperature data And the data matrix of environmental parameter data corresponding to the skin temperature data , to respectively measure the temperature time series characteristics of the measurement points , Environmental time series characteristics Extract.

[0027] By collecting the skin temperature data of multiple measurement points with physiological connections (such as head, hands, arms, chest, thighs, calves, feet, etc.) of the target object, combined with external conditions such as ambient temperature, humidity, wind speed (environmental parameter data), the spatial, environmental, temperature and other temporal characteristics related to the environment in which the target object is located are extracted to provide multi-dimensional data input for the prediction of the target object's skin temperature in the future.

[0028] According to the adjacency relationship between multiple measurement points with physiological connection relationship, the topological structure between multiple measurement points of the target object is constructed to effectively capture the spatial dependency between different measurement points and provide spatial dependency information for the prediction of skin temperature of the target object in the future. Here, according to the relative position between the measurement points, the spatial dependency between the skin temperature measurement points is established. For example, there is a strong temperature transfer relationship between the adjacent parts such as the head and the arm, the hand and the arm, the foot and the calf, and the thigh. Through the relative position and connectivity between the measurement points, the spatial relationship between the nodes of the topological structure is established.

[0029] In this application, the spatial adjacency relationship between the skin temperature measurement points of the target object is described by the adjacency matrix. Specifically, the adjacency matrix as follows:

[0030] Among them, for the measurement points that do not have a direct adjacency relationship, the corresponding element in the adjacency matrix is ​​0, and for the measurement points that have a direct adjacency relationship, the corresponding element in the adjacency matrix is ​​1. Then, the adjacency matrix is ​​normalized to effectively ensure that the degree of each node (measurement point) in the topological structure has a balanced impact on the spatiotemporal graph convolutional network used for feature extraction. Specifically, according to the formula:

[0031] Adjacency Matrix Perform normalization to obtain the normalized adjacency matrix In the formula, the degree matrix It is a diagonal matrix, each diagonal element of which represents the degree of the corresponding measurement point (the number of direct connections between a measurement point and other measurement points). According to the adjacency matrix The degree of each node (measurement point) can be calculated (for example, if the head is connected to the chest, the degree of the head is 1; if the chest is connected to the head, the forearm and the thigh, the degree of the chest is 3), and the corresponding degree matrix can be constructed. , specifically,

[0032] After getting the normalized adjacency matrix After that, based on the graph convolution operation, according to the formula:

[0033] Determine the measuring point at Skin temperature feature matrix at the moment ; In the formula, is the ReLU activation function; Target The skin temperature feature matrix at the moment, its elements are Skin temperature data at each measurement point at all times, is the skin temperature feature matrix In the graph convolutional neural network, the features of each measurement point are combined with the features of the neighboring nodes, the temperature features of each measurement point are updated, and the weight matrix is ​​automatically learned by combining the neighboring node information. And adjust the features to finally get the updated features of each measurement point Then, the skin temperature feature matrix of all historical moments of the measurement point is integrated to obtain the spatial temporal characteristics of the measurement point: .

[0034] The skin temperature data of the target object is obtained by measuring multiple measurement points at multiple times. Each measurement point has a measured temperature value at each time. For example, the skin temperature data of the target object is obtained by measuring the temperature of the head, chest, arm, hand, thigh, calf, and foot of the target object at multiple times, and the data matrix of the skin temperature data is constructed accordingly. ,as follows:

[0035] In the formula, Indicates The measurement points are Skin temperature measurement at the moment.

[0036] Then, the first informer module (informer-1) based on the self-attention mechanism learns the dependency between the data (temperature measurement value) at each moment and the data (temperature measurement value) at other moments. By stacking multiple self-attention mechanisms together, the complex and long-term dependency of skin temperature over time is deeply mined. The data matrix of skin temperature data is input into informer-1. After each layer of self-attention learning, the correlation between the data (temperature measurement value) at any moment and the data (temperature measurement value) at other moments is judged to generate high-level time series features. Specifically, according to the formula:

[0037] Temperature time series characteristics of the measurement point Extraction is performed to obtain the variation pattern and long-term dependence of skin temperature data over time. In the formula, is the data matrix of skin temperature data, The data matrix for skin temperature data The corresponding weight matrix of the query matrix, The data matrix for skin temperature data The weight matrix corresponding to the bond matrix, The data matrix for skin temperature data The weight matrix corresponding to the value matrix; The data matrix for skin temperature data The corresponding vector dimension in the key matrix.

[0038] While measuring the skin temperature data of the target object, the environmental variables (such as temperature, humidity, wind speed, etc.) are recorded at each moment to construct a data matrix of environmental parameter data. ,as follows:

[0039] In the formula, Indicates The environment variables are Environmental measurements at the moment.

[0040] Then, the first informer module (informer-1), which is also based on the self-attention mechanism, learns the dependency between the data (environmental measurement values) at each moment and the data (environmental measurement values) at other moments. By stacking multiple self-attention mechanisms together, it deeply explores the complex and long-term dependency of environmental variables over time. The data matrix of environmental parameter data is input into informer-1. After each layer of self-attention learning, the correlation between the data (environmental measurement values) at any moment and the data (environmental measurement values) at other moments is judged to generate high-level time series features. Specifically, according to the formula:

[0041] Environmental timing characteristics of the measurement point Extraction is performed to obtain the changing patterns and long-term dependence of environmental parameter data over time. In the formula, is the data matrix of environmental parameter data corresponding to skin temperature data; is the data matrix of environmental parameter data corresponding to skin temperature data The corresponding weight matrix of the query matrix, is the data matrix of environmental parameter data corresponding to skin temperature data The weight matrix corresponding to the bond matrix, is the data matrix of environmental parameter data corresponding to skin temperature data The weight matrix corresponding to the value matrix; is the data matrix of environmental parameter data corresponding to skin temperature data The corresponding vector dimension in the key matrix.

[0042] Step S102: spatial temporal features , Temperature time series characteristics , Environmental time series characteristics The features are fused in the second informer module based on the self-attention mechanism in time sequence, and the obtained feature matrix is ​​input into the fully connected neural network to predict the skin temperature of the measurement point of the target object.

[0043] In this application, through the spatial temporal characteristics Reflect the spatial dependence between the skin temperature measurement points of the target object through the temperature time series characteristics Reflects the change of the target object's skin temperature over time, through the environmental time series characteristics It reflects the influence of environmental factors on the change of the target object's skin temperature over time, and characterizes the change law of environmental variables in the area where the target object is located when the temperature is measured. , Temperature time series characteristics , Environmental time series characteristics They are fused in time sequence in the second informer module (informer-2) based on the self-attention mechanism.

[0044] First, calculate the spatial and temporal characteristics of the measurement points respectively , Temperature time series characteristics , Environmental time series characteristics Pearson correlation coefficient with skin temperature data ;in, , , , Characterize the spatial and temporal characteristics of the measurement points Correlation with skin temperature data, Characterize the temperature time series characteristics of the measurement point Correlation with skin temperature data, Characterize the environmental timing characteristics of the measurement point Correlation with skin temperature data.

[0045] Then, according to the spatial and temporal characteristics , Temperature time series characteristics , Environmental time series characteristics Pearson correlation coefficient with skin temperature data , according to the formula:

[0046] Calculate the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The importance ratio of , All are positive integers; Characterize the measurement point The time series characteristics and The importance ratio of the time series features.

[0047] Next, according to the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The importance ratio of the spatial and temporal characteristics of the measurement points is calculated in order of size. , Temperature time series characteristics , Environmental time series characteristics Sort by importance and build a judgment matrix Among them, the importance ratio As a judgment matrix The elements in the judgment matrix The diagonal elements in are reciprocals of each other, that is:

[0048] Characterize the measurement point The time series characteristics and The importance ratio of the time series features.

[0049] In the judgment matrix In the example, each row element is divided into two parts according to the spatial and temporal characteristics. , Temperature time series characteristics , Environmental time series characteristics The importance ratios are sorted by spatial and temporal features. , Temperature time series characteristics , Environmental time series characteristics That is, according to the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics Sort the importance of the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics With the judgment matrix The elements of each row correspond one to one.

[0050] For example, spatial temporal features are sorted in descending order of importance ratio. , Temperature time series characteristics , Environmental time series characteristics When sorting, the judgment matrix The first row of elements in is the data corresponding to the time series feature with the largest importance ratio, and the last row of elements is the data corresponding to the time series feature with the smallest importance ratio; Furthermore, according to the judgment matrix The time series features corresponding to each row element in , respectively calculate the weight vector of the corresponding time series features. That is, according to the judgment matrix , respectively calculate the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The weight vector of . Specifically, according to the formula:

[0051] Calculate the spatial temporal characteristics of the measurement points separately , Temperature time series characteristics , Environmental time series characteristics The weight vector of for OK Column Judgment Matrix The The weight vector of the time series features corresponding to the row elements, is the judgment matrix The Line The elements of the column, ; All are positive integers.

[0052] Finally, according to the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The weight vector of the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics Perform weighted fusion to obtain the fusion features of the measurement points, that is, the feature matrix ; and the obtained feature matrix The fully connected neural network is input to predict the skin temperature of each measurement point of the target object at any time in the future.

[0053] Therefore, through the topological structure of multiple measurement points with physiological connection relationships, the spatial time series features, temperature time series features and environmental time series features are comprehensively extracted based on the spatiotemporal graph convolutional network and the informer-1 module based on the self-attention mechanism, and are fused in the informer-2 module based on the self-attention mechanism according to the time series, effectively establishing the spatial dependency between different measurement points, overcoming the constraints of the existing technology data quality on the model performance and the insufficient applicability of the traditional human body heat balance equation method in dynamic scenarios, enhancing the generalization ability of skin temperature prediction, and using the obtained multi-dimensional features as the input of the fully connected neural network to provide sufficient information for the subsequent skin temperature prediction, effectively improving the accuracy and robustness of temperature prediction.

[0054] The present application also provides a human skin temperature prediction system integrating informer and STGCN, such as Figure 6 As shown, the prediction system includes: The feature extraction unit 601 is configured to extract the spatial temporal features of the measurement points based on the topological structure between the multiple measurement points corresponding to the skin temperature data of the target object obtained, based on the spatiotemporal graph convolution network. Extraction; and, based on the first informer module of the self-attention mechanism, respectively, according to the data matrix of the skin temperature data And the data matrix of environmental parameter data corresponding to the skin temperature data , to respectively measure the temperature time series characteristics of the measurement points , Environmental time series characteristics Perform extraction; The temperature prediction unit 602 is configured to convert the spatial temporal features , Temperature time series characteristics , Environmental time series characteristics The features are fused in the second informer module based on the self-attention mechanism in time sequence, and the obtained feature matrix is ​​input into the fully connected neural network to predict the skin temperature of the measurement point of the target object.

[0055] The human skin temperature prediction system integrating informer and STGCN provided in the embodiments of the present application can implement the steps and processes of the human skin temperature prediction method integrating informer and STGCN in any of the above embodiments, and achieve the same technical effects, which will not be repeated here one by one.

[0056] In the description of the present invention, it is to be understood that the terms “center”, “longitudinal”, “lateral”, “length”, “width”, “thickness”, “up”, “down”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, “clockwise”, “counterclockwise”, “axial”, “radial”, “circumferential”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0057] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0058] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral one; it can be a mechanical connection, an electrical connection, or communication with each other; it can be a direct connection, or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0059] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0060] In the present invention, the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0061] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A human skin temperature prediction method integrating informer and STGCN, characterized in that: include: Step S101: Based on the topological structure between multiple measurement points corresponding to the skin temperature data of the target object, the spatial temporal characteristics of the measurement points are analyzed based on the spatiotemporal graph convolution network. Perform extraction; And, the first informer module based on the self-attention mechanism, according to the data matrix of the skin temperature data and a data matrix of environmental parameter data corresponding to the skin temperature data obtained , to respectively analyze the temperature timing characteristics of the measurement points , Environmental time series characteristics Perform extraction; Step S102: The spatial temporal features , the temperature time series characteristics , the environmental time series characteristics The features are fused in a second informer module based on a self-attention mechanism in time sequence, and the obtained feature matrix is ​​input into a fully connected neural network to predict the skin temperature of the measuring point of the target object.

2. The human skin temperature prediction method integrating informer and STGCN according to claim 1 is characterized in that: In step S101, An adjacency matrix constructed based on the topological structure between multiple measurement points corresponding to the skin temperature data of the target object Perform normalization to obtain the normalized adjacency matrix , and based on the graph convolution operation, according to the formula: ； Determine the measuring point at Skin temperature feature matrix at the moment , to construct the spatial temporal characteristics of the measurement points ; In the formula, is the ReLU activation function; For the target object The skin temperature feature matrix at the moment, is the skin temperature feature matrix The corresponding weight matrix.

3. The human skin temperature prediction method integrating informer and STGCN according to claim 1 is characterized in that: In step S101, according to the formula: ； The temperature time series characteristics of the measurement point , Environmental time series characteristics Perform extraction; In the formula, is the data matrix of the skin temperature data, The data matrix for the skin temperature data The corresponding weight matrix of the query matrix, The data matrix for the skin temperature data The weight matrix corresponding to the bond matrix, The data matrix for the skin temperature data The weight matrix corresponding to the value matrix; The data matrix for the skin temperature data The corresponding vector dimensions in the key matrix; A data matrix of environmental parameter data corresponding to the skin temperature data; is a data matrix of environmental parameter data corresponding to the skin temperature data The corresponding weight matrix of the query matrix, is a data matrix of environmental parameter data corresponding to the skin temperature data The weight matrix corresponding to the bond matrix, is a data matrix of environmental parameter data corresponding to the skin temperature data The weight matrix corresponding to the value matrix; is a data matrix of environmental parameter data corresponding to the skin temperature data The corresponding vector dimensions in the key matrix.

4. The human skin temperature prediction method integrating informer and STGCN according to claim 1 is characterized in that: In step S102, Calculate the spatial temporal characteristics of the measurement points respectively , Temperature time series characteristics , Environmental time series characteristics Pearson correlation coefficient with the skin temperature data , and according to the Pearson correlation coefficient Calculate the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The importance ratio of According to the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The importance ratio of the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics Sort by importance to build a judgment matrix ; According to the judgment matrix , respectively calculate the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The weight vector is used to calculate the spatial and temporal characteristics of the measurement points. , Temperature time series characteristics , Environmental time series characteristics Perform feature fusion to obtain the feature matrix .

5. The human skin temperature prediction method integrating informer and STGCN according to claim 4 is characterized in that: According to the formula: ； Calculate the spatial temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The importance ratio of ; in, , are all positive integers, is the judgment matrix The diagonal elements in Characterizes the measurement point The time series characteristics and The importance ratio of the time series features; Characterizes the measurement point The time series characteristics and The importance ratio of the time series features.

6. The human skin temperature prediction method integrating informer and STGCN according to claim 4 is characterized in that: According to the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics Sort the importance of the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics With the judgment matrix The row elements of each row correspond one to one to determine the spatial temporal characteristics of the measurement point , Temperature time series characteristics , Environmental time series characteristics The weight vector of .

7. The human skin temperature prediction method integrating informer and STGCN according to claim 6 is characterized in that: According to the formula: ； Calculate the spatial temporal characteristics of the measurement points respectively , Temperature time series characteristics , Environmental time series characteristics The weight vector of In the formula, for OK The judgment matrix of columns The The weight vector of the time series features corresponding to the row elements, is the judgment matrix The Line Elements of the column, ; All are positive integers.

8. According to claim 1, the method for predicting human skin temperature by integrating informer and STGCN is characterized in that: According to the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics The weight vector of the spatial and temporal characteristics of the measurement points , Temperature time series characteristics , Environmental time series characteristics Perform weighted fusion to obtain the feature matrix of the measurement point .

9. A human skin temperature prediction system integrating informer and STGCN, characterized in that: include: The feature extraction unit is configured to extract the spatial temporal features of the measurement points based on the topological structure between the multiple measurement points corresponding to the skin temperature data of the target object obtained, based on the spatiotemporal graph convolution network. Extraction; and, based on the first informer module of the self-attention mechanism, respectively according to the data matrix of the skin temperature data and a data matrix of environmental parameter data corresponding to the skin temperature data obtained , to respectively analyze the temperature timing characteristics of the measurement points , Environmental time series characteristics Perform extraction; The temperature prediction unit is configured to convert the spatial temporal characteristics , the temperature time series characteristics , the environmental time series characteristics The features are fused in a second informer module based on a self-attention mechanism in time sequence, and the obtained feature matrix is ​​input into a fully connected neural network to predict the skin temperature of the measuring point of the target object.

Images