A data feature extraction and analysis method and system based on infrared thermal imaging

Abstract

The present invention discloses a data feature extraction and analysis method and system based on infrared thermal imaging, and relates to the technical field of core body temperature monitoring. The method comprises: establishing a database; if the monitoring scene is judged to be a multi-person rapid monitoring scene, the coefficient of the facial temperature-ambient temperature coupling feature expression is obtained by fitting according to the data of all users in the database, a core body temperature prediction model is established, and the core body temperature of the target user is predicted; if the monitoring scene is judged to be a single-person long-term monitoring scene, the fitting coefficient of the coupling feature expression is obtained based on the similarity between the personal characteristics of the target user and the database user, and a personalized core body temperature prediction model is established to predict the core body temperature of the target user. The present invention proposes a facial temperature-ambient temperature coupling feature, realizes the accurate measurement of human core body temperature based on infrared thermal imaging, and can provide the best monitoring solution for multi-person rapid monitoring and single-person long-term monitoring scenes.

Description

Technical Field

[0001] The present invention relates to the field of body temperature monitoring, and in particular to a method and system for extracting and analyzing data features based on infrared thermal imaging. Background Art

[0002] Non-contact detection methods of core body temperature, represented by infrared thermal imaging, have played an important role in the rapid screening and control of some infectious diseases, especially in hospitals, schools, airports and other crowded public places. However, non-contact detection of human core body temperature often has problems such as low accuracy and susceptibility to environmental influences. Since infrared thermal imaging only infers core body temperature through body surface temperature, the difficult-to-quantify difference between surface temperature and core body temperature is the key reason why infrared thermal imaging is difficult to accurately predict core body temperature. The factors affecting this difference are complex and closely related to environmental factors, individual characteristics, personal status and other factors. The application environment temperature range of infrared thermal imaging body temperature detection recommended by existing standards is relatively limited, only 20-24°C, which greatly limits the application scenarios of infrared thermal imaging body temperature detection.

[0003] The existing technology studies and improves the above problems, however, there are generally the following shortcomings:

[0004] (1) In the process of infrared thermal imaging detecting core body temperature, the difference between skin temperature and core body temperature cannot be accurately quantified, and the role of ambient temperature in this process (coupling effect between ambient temperature and skin temperature) is ignored, which limits the application range of ambient temperature for infrared thermal imaging body temperature monitoring;

[0005] (2) Existing technologies ignore the impact of individual differences on the surface-core body temperature pattern. In the process of calculating core body temperature, errors caused by differences in personal characteristics are not considered, which will affect the adaptability of infrared thermal imaging body temperature monitoring to the population.

[0006] Therefore, it is urgent to propose effective solutions to the above problems. Summary of the invention

[0007] The purpose of the present invention is to address the deficiencies in the prior art and to propose a data feature extraction and analysis method and system based on infrared thermal imaging.

[0008] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a data feature extraction and analysis method based on infrared thermal imaging, the method comprising the following steps:

[0009] Obtain user data, including infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics, and real core body temperature, and establish a database; determine whether it is a multi-person rapid monitoring scenario or a single-person long-term monitoring scenario;

[0010] If it is a multi-person rapid monitoring scenario, after the database is established, all user data in the database are used as a data set, and the fitting coefficient K is obtained by fitting the first coupling feature expression. The fitting coefficient K is substituted into the second coupling feature expression to calculate the coupling features of all users in the database; based on all user data in the database as a training data set, coupling features, facial temperature distribution features, personal features, and environmental features are used as inputs, and the real core body temperature is used as a label, training is performed through a machine learning method to establish a core body temperature prediction model; the infrared thermal imaging facial temperature distribution features, personal features, and environmental features of the target user are obtained, and the coupling features of the target user are obtained based on the second coupling feature expression after substituting the fitting coefficient K, and the coupling features are input into the core body temperature prediction model to predict the core body temperature of the target user, and the core body temperature of multiple target users is obtained by repeating the above steps multiple times;

[0011] If it is a single-person long-term monitoring scenario, after establishing the database, obtain the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics, and environmental characteristics; extract user data with similar personal characteristics in the database through the target user's personal characteristics as a data set, and obtain the fitting coefficient K through the first coupling feature expression, substitute the fitting coefficient K into the second coupling feature expression, and calculate the coupling characteristics of all users in the database and the target user; assign weights to all users in the database according to the similarity with the target user's personal characteristics, and perform adaptive adjustments based on the weights as a training data set, with coupling characteristics, facial temperature distribution characteristics, personal characteristics, and environmental characteristics as inputs, and real core body temperature as a label, and train through machine learning methods to establish a personalized core body temperature prediction model; input the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics, and coupling characteristics into the personalized core body temperature prediction model to predict the target user's core body temperature;

[0012] The first coupling characteristic expression is: T c =(K+1)T max -KT a , the second coupling characteristic expression is: T op =(K+1)T max -KT a , where T max is the maximum temperature of the facial ROI, T a is the ambient temperature, T c is the true core body temperature, T op is the coupling characteristic, and K is the fitting coefficient.

[0013] Furthermore, the infrared thermal imaging facial temperature distribution characteristics include the maximum temperature T of the facial ROI max , forehead ROI maximum temperature T FEmax , Maximum temperature of inner canthus ROI T CEmax , Maximum temperature of mouth ROI TMmax ;

[0014] The personal characteristics include gender, age, race, BMI, and body fat percentage; wherein gender and race are assigned values ​​{0, 1, ..., n} by category, and age, BMI, and body fat percentage are assigned values ​​{0, 1, ..., n} by numerical interval;

[0015] The environmental characteristics include the ambient temperature T a .

[0016] Furthermore, the infrared thermal imaging facial temperature distribution features of the users in the database and the target user are obtained by performing face extraction, facial region division, and temperature value calculation on the infrared thermal imaging images of the users;

[0017] The personal characteristics of the users in the database are obtained by artificial collection, and the personal characteristics of the target user are obtained by artificial collection or prediction based on the user's infrared thermal image by machine learning method;

[0018] The environmental characteristics of the users and target users in the database are collected by auxiliary environmental parameter collection equipment;

[0019] The actual core body temperature T of the user in the database c Measured by a medical thermometer.

[0020] In some specific embodiments, the method further includes: performing feature selection on features in a training data set to obtain a key feature set, training through a machine learning method, and using the key feature set of the target user as an input feature of a core body temperature prediction model or a personalized core body temperature prediction model to predict the core body temperature of the target user; wherein the key feature set includes at least a coupling feature.

[0021] Preferably, the feature selection method includes: Pearson correlation coefficient and recursive feature elimination.

[0022] Preferably, the machine learning method includes: random forest, neural network, support vector machine.

[0023] Furthermore, in a single-person long-term monitoring scenario, extracting user data with similar personal characteristics from a database as a data set through the personal characteristics of the target user includes: calculating the similarity between the personal characteristics of all users in the database and the target user, setting a threshold, and including user data with a similarity greater than the threshold into the data set;

[0024] The method assigns weights to all users in the database according to the similarity with the personal characteristics of the target user, and performs adaptive adjustment based on the weights to obtain a training data set, including: normalizing the personal characteristics, calculating the similarity between the personal characteristics of all users in the database and the personal characteristics of the target user, assigning weights to all users in the database based on the similarity so that users with relatively large similarities have relatively high weights, and resampling according to the weights to obtain the training data set;

[0025] The similarity calculation method includes: weighted Euclidean distance, Manhattan distance, Mahalanobis distance, Langevin distance, and Minkowski distance.

[0026] In some specific embodiments, the method further includes: after predicting the core body temperature of the target user, using a medical thermometer to measure the target user's actual core body temperature, incorporating the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics and actual core body temperature into a database for updating.

[0027] As a second aspect, the present invention provides a core body temperature prediction system based on infrared thermal imaging, the system comprising:

[0028] The data collection unit includes a personal feature collection module, a facial temperature data collection module, an environmental data collection module, and a core body temperature collection module, which are respectively used to collect personal features, infrared thermal imaging facial temperature distribution features, environmental features, and real core body temperature data measured by a medical thermometer, and establish a database; in a single-person long-term monitoring scenario, the data in the database is input into a first similarity measurement module, and in a multi-person rapid monitoring scenario, the data is input into a coupling feature calculation module;

[0029] The feature extraction unit includes a first similarity measurement module and a coupling feature calculation module. The first similarity measurement module is used to calculate the similarity between the target user and the personal features of the user in the database in a single-person long-term monitoring scenario, set a threshold, and select user data with a similarity greater than the threshold to input into the coupling feature calculation module to calculate the coupling feature; the coupling feature calculation module is used to calculate the coupling feature according to the facial temperature distribution feature and the environmental feature; the data processed by the feature extraction unit is input into the model establishment unit as a training data set in a multi-person rapid monitoring scenario, and is input into the data adaptive adjustment unit in a single-person long-term monitoring scenario;

[0030] A data adaptive adjustment unit is used to receive the data processed by the feature extraction unit in a single-person long-term monitoring scenario, perform data adaptive adjustment, and obtain a training data set; the data adaptive adjustment unit includes a second similarity measurement module and a data resampling module, the second similarity measurement module evaluates the similarity between the target user and the personal features of the users in the database, and the data resampling module resamples according to the similarity to obtain the training data set, and users with relatively large similarity have relatively high weights;

[0031] A model building unit, used to build a core body temperature prediction model by training based on a training data set through a machine learning method, wherein the training data set consists of personal characteristics, facial temperature distribution characteristics, environmental characteristics, coupling characteristics, and real core body temperature;

[0032] The model application unit includes a core body temperature calculation module, which calculates the predicted core body temperature based on the core body temperature prediction model according to the personal characteristics, facial temperature distribution characteristics, environmental characteristics, and coupling characteristics of the target user.

[0033] As a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, is used to implement the above-mentioned data feature extraction and analysis method based on infrared thermal imaging.

[0034] Compared with the prior art, the present invention has at least the following beneficial effects:

[0035] (1) The facial temperature-ambient temperature coupling feature proposed in the present invention is an important feature obtained after analyzing the human-environment heat transfer mechanism. It quantifies the difference between the body surface temperature affected by the ambient temperature and the core body temperature, thereby improving the accuracy of calculating the core body temperature using infrared thermal imaging facial temperature at different ambient temperatures, and improving the environmental adaptability of infrared thermal imaging body temperature monitoring technology;

[0036] (2) The personalized core body temperature prediction model proposed in the present invention can adaptively adjust the model structure to obtain the prediction model that best matches the target user. Taking into account the differences between different individuals, it is possible to establish the optimal prediction model for different individuals, thereby improving the individual adaptability of infrared thermal imaging body temperature monitoring technology.

[0037] (3) The present invention can meet the needs of different monitoring scenarios. For single-person long-term monitoring, it can provide the coupling characteristics and personalized model that are most suitable for the target user, ensuring the most accurate body temperature prediction results; for multi-person rapid monitoring scenarios, it can achieve rapid and real-time body temperature prediction while ensuring accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1This is a flow chart of the data feature extraction and analysis method based on infrared thermal imaging proposed by the present invention;

[0039] Figure 2 This is a structural block diagram of the core body temperature prediction system based on infrared thermal imaging in a multi-person rapid monitoring scenario proposed by the present invention;

[0040] Figure 3 This is a structural block diagram of the core body temperature prediction system based on infrared thermal imaging in a single-person long-term monitoring scenario proposed by the present invention. DETAILED DESCRIPTION

[0041] Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings.

[0042] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0043] The present invention proposes a data feature extraction and analysis method based on infrared thermal imaging, including a multi-person rapid monitoring scenario and a single-person long-term monitoring scenario. The multi-person rapid monitoring scenario does not require time to retrain the model. The key is to quickly obtain the prediction results. Therefore, a universal prediction model can be used to quickly and in real time obtain the prediction results while ensuring a certain degree of accuracy. Single-person long-term monitoring will establish a personalized core body temperature prediction model that best suits the individual's personal characteristics. After first learning about the individual's characteristic data, it will take a certain amount of time to complete the adaptive adjustment of the model training data set and the model training process. After the model training is completed, it can become the best prediction model for the target user, thereby improving the accuracy of the entire monitoring process.

[0044] like Figure 1 The flowchart of the data feature extraction and analysis method based on infrared thermal imaging proposed by the present invention is shown, which specifically includes the following steps:

[0045] S1. Establish a user database. The user data in the database includes infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics, and real core body temperature data measured by a medical thermometer; obtain the facial temperature-ambient temperature coupling characteristic expression of the coefficient to be fitted based on the heat transfer mechanism.

[0046] S2. Obtain monitoring scenario requirements. If it is a multi-person rapid monitoring scenario, jump to step S3 and end after completing step S7; if it is a single-person long-term monitoring scenario, jump to step S8.

[0047] S3. Using the first coupling feature expression, a fitting coefficient K of the coupling feature expression is obtained according to all user data in the database; based on the second coupling feature expression, coupling features of users in the database are calculated according to facial temperature distribution features and environmental features.

[0048] S4. Based on the machine learning algorithm, all user data in the database is used as a training data set to establish a core body temperature prediction model; the input of the model training is all the features in steps S1 and S3 in the data set, and the output of the model training is the actual core body temperature in the data set.

[0049] In some of the embodiments, during the model building process, feature selection can be performed on the input features to remove features that have little impact on the core body temperature prediction, and retain important features to form a key feature set as the final input features for model training.

[0050] S5. Obtain the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics and environmental characteristics, substitute the fitting coefficient K obtained in step S3 based on the second coupling characteristic expression, and calculate the coupling characteristics of the target user according to the facial temperature distribution characteristics and environmental characteristics.

[0051] S6. Obtain the infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics, and coupling characteristics of the target user in step S4 as model input characteristics, and predict the core body temperature of the target user according to the established core body temperature prediction model.

[0052] In some embodiments, the model input features are the key feature set obtained through feature selection in step S4.

[0053] S7. Repeat steps S5 and S6 to calculate the core body temperatures of different target users.

[0054] S8. Obtain the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics and environmental characteristics.

[0055] S9. Extract user data with similar personal features from a database based on the personal features of the target user, and obtain the fitting coefficient K of the coupling feature expression of the target user based on the first coupling feature expression; and calculate the coupling features of the target user and the coupling features of users in the database based on the second coupling feature expression according to facial temperature distribution features and environmental features.

[0056] S10. According to the similarity of personal characteristics between the target user and the users in the database, weights are assigned to different samples in the database, and the data distribution in the database is adaptively adjusted to form a training data set. A personalized core body temperature prediction model is established based on the machine learning algorithm and the adjusted training data set. The input of the model training is all the features in steps S1 and S9 in the data set, and the output of the model training is the actual core body temperature in the data set.

[0057] In some of the embodiments, during the model building process, feature selection can be performed on the input features to remove features that have little impact on the core body temperature prediction, and retain important features to form a key feature set as the final input features for model training.

[0058] S11. Obtain infrared thermal imaging facial temperature distribution features, personal features, environmental features, and coupling features of the target user as model input features, and predict the core body temperature of the target user based on the established personalized core body temperature prediction model.

[0059] In some embodiments, the model input features are key feature sets obtained through feature selection in step S10.

[0060] In some of the embodiments, S12 is also included, and a medical thermometer can be selectively used to measure the real core body temperature of the target user, and all data of the target user can be saved in a database.

[0061] Repeat step S11, continuously input updated features of the target user, and calculate the core body temperature of the target user in real time.

[0062] In step S1, the infrared thermal imaging facial temperature distribution features include the maximum temperature (T max ), forehead ROI maximum temperature (T FEmax ), maximum temperature of the inner canthus ROI (T CEmax ), maximum temperature of mouth ROI (T Mmax ), the facial temperature distribution features are input into the database after face extraction, facial area division and temperature value calculation of infrared thermal imaging images; the personal features include gender, age, race, BMI (Body Mass Index), body fat percentage, and the format is converted into a format readable by the model, gender and race are assigned values ​​{0,1,…,n} by category, age, BMI, body fat percentage are assigned values ​​{0,1,…,n} by numerical interval, and the personal features are collected manually and input into the database; the environmental features include the ambient temperature (T a ), which is collected by the auxiliary environmental parameter collection device and then input into the database; the real core body temperature data T cMeasured by a medical thermometer and entered into the database.

[0063] In step S1, the user data in the database includes facial temperature distribution characteristics, personal characteristics, ambient temperature and real core body temperature. In the database establishment stage, a large amount of user real core body temperature data measured by a medical thermometer is recorded and saved for model training and coefficient fitting of coupling characteristics.

[0064] In steps S1, S3 and S9, the second coupling characteristic (T op ) is expressed as T op =(K+1)T max -KT a , where T max is the maximum temperature of the facial ROI, T a The coefficient K to be fitted of the coupling feature is obtained by fitting the following equation (the first coupling feature expression) based on the least squares method or other algorithms: c =(K+1)T max -KT a , where T max is the maximum temperature of the facial ROI, T a is the ambient temperature, T c is the real core body temperature. In the scenario of multi-person rapid monitoring, in order to improve efficiency and save model training time, the K value is set to the T of all users in the database. c 、T max 、T a The result of data fitting is that the coupling features of all users in the database can be calculated before obtaining the target user data, and the core body temperature prediction model can be trained. In the single-person long-term monitoring scenario, the fitting coefficient K of the coupling feature is based on the T of users with similar personal characteristics in the database. c 、T max 、T a The data is fitted to maximize the guarantee that the obtained coupling characteristics are the most accurate fit for the target individual.

[0065] In steps S4 and S10, the key feature set at least includes coupling features, and the most relevant and discriminative key feature set can be screened out by feature selection methods such as Pearson correlation coefficient and recursive feature elimination. The machine learning algorithm can adopt a machine learning algorithm that can fit various complex nonlinear mappings, including random forests, neural networks, and support vector machines.

[0066] In steps S6 and S11, the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics, and coupling characteristics are obtained through real-time processing of infrared thermal images and auxiliary environmental parameter acquisition equipment. Personal characteristics may not be manually collected in a multi-person rapid monitoring scenario, and the personal characteristics of the target user (such as gender, age, etc.) can be predicted by machine learning algorithms with the help of infrared thermal images. In a single-person monitoring scenario, the target user's personal characteristic data can be manually collected in advance and input into the model, and other data can be obtained through real-time processing of infrared thermal images and auxiliary environmental parameter acquisition equipment.

[0067] In step S9, the user data with similar personal characteristics in the database is extracted according to the personal characteristics of the target user, the similarity between all users in the database and the personal characteristics of the target user is calculated, a threshold is set, and the user data with similarity greater than the threshold is included in the data set.

[0068] In step S10, the similarity of personal characteristics is calculated by normalizing characteristics such as gender, age, race, BMI, body fat percentage, etc., and then obtained by weighted Euclidean distance, Manhattan distance, Mahalanobis distance, Langevin distance, Minkowski distance or other similarity calculation methods; based on the similarity calculation results, different weights are assigned to the data of different users in the database so that users with similar characteristics have larger weights, and the data are resampled according to the weights to form a training data set for the model.

[0069] The present invention also provides a core body temperature prediction system based on infrared thermal imaging, comprising the following modules: a data acquisition unit, including a personal feature acquisition module, a facial temperature data acquisition module, an environmental data acquisition module, and a core body temperature acquisition module, which are respectively used to collect personal features, infrared thermal imaging facial temperature distribution features, environmental features, and real core body temperature data measured by a medical thermometer, and establish a database; in a single-person long-term monitoring scenario, the data in the database are input into a first similarity measurement module, and in a multi-person rapid monitoring scenario, the data are input into a coupling feature calculation module;

[0070] The feature extraction unit includes a first similarity measurement module and a coupling feature calculation module. The first similarity measurement module is used to calculate the similarity between the target user and the personal features of the user in the database in a single-person long-term monitoring scenario, set a threshold, and select user data with a similarity greater than the threshold to input into the coupling feature calculation module to calculate the coupling feature; the coupling feature calculation module is used to calculate the coupling feature according to the facial temperature distribution feature and the environmental feature; the data processed by the feature extraction unit is input into the model establishment unit as a training data set in a multi-person rapid monitoring scenario, and is input into the data adaptive adjustment unit in a single-person long-term monitoring scenario;

[0071] A data adaptive adjustment unit is used to receive the data processed by the feature extraction unit in a single-person long-term monitoring scenario, perform data adaptive adjustment, and obtain a training data set; the data adaptive adjustment unit includes a second similarity measurement module and a data resampling module, the second similarity measurement module evaluates the similarity between the target user and the personal features of the users in the database, and the data resampling module resamples according to the similarity to obtain the training data set, so that users with relatively large similarity have relatively high weights;

[0072] A model building unit, used to build a core body temperature prediction model by training based on a training data set through a machine learning method, wherein the training data set consists of personal characteristics, facial temperature distribution characteristics, environmental characteristics, coupling characteristics, and real core body temperature;

[0073] The model application unit includes a core body temperature calculation module, which calculates the predicted core body temperature based on the core body temperature prediction model according to the personal characteristics, facial temperature distribution characteristics, environmental characteristics, and coupling characteristics of the target user.

[0074] In some of the embodiments, modules may be added or reduced according to different monitoring scenario requirements to provide a corresponding system structure.

[0075] (1) In the scenario of multiple people fast monitoring, such as Figure 2 As shown, the core body temperature prediction system based on infrared thermal imaging includes a data acquisition unit, a feature extraction unit, a model building unit, and a model application unit;

[0076] The data acquisition unit includes a personal feature acquisition module, a facial temperature data acquisition module, an environmental data acquisition module, and a core body temperature acquisition module, which are respectively used to collect personal features, infrared thermal imaging facial temperature distribution features, environmental features, and real core body temperature data measured by a medical thermometer;

[0077] The feature extraction unit includes a coupling feature calculation module, which calculates the facial temperature-ambient temperature coupling feature according to the facial temperature distribution feature and the environmental feature;

[0078] The model building unit includes a training data set, a machine learning algorithm, and a core body temperature prediction model. The training data set consists of personal characteristics, facial temperature distribution characteristics, environmental characteristics, coupling characteristics, and core body temperature. The core body temperature prediction model is established based on the machine learning algorithm and the training data set.

[0079] The model application unit includes a core body temperature calculation module, which calculates the predicted core body temperature based on the core body temperature prediction model according to the personal characteristics, facial temperature distribution characteristics, environmental characteristics, and coupling characteristics of multiple users.

[0080] (2) In the scenario of single-person long-term monitoring, such as Figure 3 As shown, the core body temperature prediction system based on infrared thermal imaging includes a data acquisition unit, a feature extraction unit, a data adaptive adjustment unit, a model building unit, and a model application unit;

[0081] The data acquisition unit includes a personal feature acquisition module, a facial temperature data acquisition module, an environmental data acquisition module, and a core body temperature acquisition module, which are respectively used to collect personal features, infrared thermal imaging facial temperature distribution features, environmental features, and real core body temperature data measured by a medical thermometer;

[0082] The feature extraction unit includes a similarity measurement algorithm and a coupling feature calculation module. The similarity measurement algorithm is used to evaluate the similarity between the target user and the personal features of the users in the database, and select the user data with greater similarity to fit the coefficient of the coupling feature expression of the target user. The coupling feature calculation module calculates the facial temperature-ambient temperature coupling feature according to the facial temperature distribution feature and the environmental feature;

[0083] The data adaptive adjustment unit includes a similarity measurement algorithm and a data resampling algorithm. The similarity measurement algorithm is used to evaluate the similarity between the target user and the personal characteristics of the users in the database. The data resampling algorithm resamples according to the similarity to obtain a training data set, and the user data with greater similarity has a higher weight;

[0084] The model building unit includes a training data set, a machine learning algorithm, and a personalized core body temperature prediction model. The training data set is obtained after being processed by the data adaptive adjustment unit and consists of personal characteristics, facial temperature distribution characteristics, environmental characteristics, coupling characteristics, and core body temperature. The personalized core body temperature prediction model is established based on the machine learning algorithm and the training data set;

[0085] The model application unit includes a core body temperature calculation module, which calculates the predicted core body temperature based on the personalized core body temperature prediction model according to the personal characteristics, facial temperature distribution characteristics, environmental characteristics, and coupling characteristics of a single user.

[0086] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it is used to implement the data feature extraction and analysis method based on infrared thermal imaging.

[0087] The computer-readable storage medium may be a hard disk or a memory, or any other device with data processing capabilities, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc., equipped on the device.

[0088] Furthermore, the computer-readable storage medium may include both an internal storage unit of any device with data processing capabilities and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and may also be used to temporarily store data that has been output or is to be output.

[0089] Those skilled in the art will readily appreciate other embodiments of the present application after considering the description and practicing the contents disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application. The description and examples are intended to be exemplary only.

[0090] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A data feature extraction and analysis method based on infrared thermal imaging, characterized in that: The method comprises the following steps: Obtain user data, including infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics, and real core body temperature, and establish a database; determine whether it is a multi-person rapid monitoring scenario or a single-person long-term monitoring scenario; If it is a multi-person rapid monitoring scenario, after the database is established, all user data in the database are used as a data set, and the fitting coefficient K is obtained by fitting the first coupling feature expression. The fitting coefficient K is substituted into the second coupling feature expression to calculate the coupling features of all users in the database; based on all user data in the database as a training data set, coupling features, facial temperature distribution features, personal features, and environmental features are used as inputs, and the real core body temperature is used as a label, training is performed through a machine learning method to establish a core body temperature prediction model; the infrared thermal imaging facial temperature distribution features, personal features, and environmental features of the target user are obtained, and the coupling features of the target user are obtained based on the second coupling feature expression after substituting the fitting coefficient K, and the coupling features are input into the core body temperature prediction model to predict the core body temperature of the target user, and the core body temperature of multiple target users is obtained by repeating the above steps multiple times; If it is a single-person long-term monitoring scenario, after establishing the database, obtain the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics, and environmental characteristics; extract user data with similar personal characteristics in the database through the target user's personal characteristics as a data set, and obtain the fitting coefficient K through the first coupling feature expression, substitute the fitting coefficient K into the second coupling feature expression, and calculate the coupling characteristics of all users in the database and the target user; assign weights to all users in the database according to the similarity with the target user's personal characteristics, and perform adaptive adjustments based on the weights as a training data set, with coupling characteristics, facial temperature distribution characteristics, personal characteristics, and environmental characteristics as inputs, and real core body temperature as a label, and train through machine learning methods to establish a personalized core body temperature prediction model; input the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics, and coupling characteristics into the personalized core body temperature prediction model to predict the target user's core body temperature; The first coupling characteristic expression is: T c =(K+1)T max -KT a , the second coupling characteristic expression is: T op =(K+1)T max -KT a , where T max is the maximum temperature of the facial ROI, T a is the ambient temperature, T c is the true core body temperature, T op is a coupling feature.

2. The data feature extraction and analysis method based on infrared thermal imaging according to claim 1 is characterized in that: The infrared thermal imaging facial temperature distribution features include the maximum temperature T of the facial ROI max , forehead ROI maximum temperature T FEmax , Maximum temperature of inner canthus ROI T CEmax , Maximum temperature of mouth ROI T Mmax ; The personal characteristics include gender, age, race, BMI, and body fat percentage; wherein gender and race are assigned values ​​{0, 1, ..., n} by category, and age, BMI, and body fat percentage are assigned values ​​{0, 1, ..., n} by numerical interval; The environmental characteristics include the ambient temperature T a .

3. The data feature extraction and analysis method based on infrared thermal imaging according to claim 1 is characterized in that: The infrared thermal imaging facial temperature distribution features of the users in the database and the target users are obtained by performing face extraction, facial area division, and temperature value calculation on the infrared thermal imaging images of the users; The personal characteristics of the users in the database are obtained by artificial collection, and the personal characteristics of the target user are obtained by artificial collection or prediction based on the user's infrared thermal image by machine learning method; The environmental characteristics of the users and target users in the database are collected by auxiliary environmental parameter collection equipment; The actual core body temperature T of the user in the database c Measured by a medical thermometer.

4. The data feature extraction and analysis method based on infrared thermal imaging according to claim 1 is characterized in that: The method also includes: performing feature selection on features in the training data set to obtain a key feature set, training through a machine learning method, and using the key feature set of the target user as an input feature of a core body temperature prediction model or a personalized core body temperature prediction model to predict the core body temperature of the target user; wherein the key feature set includes at least a coupling feature.

5. The data feature extraction and analysis method based on infrared thermal imaging according to claim 4 is characterized in that: The feature selection method includes: Pearson correlation coefficient and recursive feature elimination.

6. The data feature extraction and analysis method based on infrared thermal imaging according to claim 1 is characterized in that: The machine learning methods include: random forest, neural network, and support vector machine.

7. The data feature extraction and analysis method based on infrared thermal imaging according to claim 1 is characterized in that: In a single-person long-term monitoring scenario, extracting user data with similar personal characteristics from a database as a data set by using the personal characteristics of the target user includes: calculating the similarity between the personal characteristics of all users in the database and the target user, setting a threshold, and including user data with a similarity greater than the threshold into the data set; The method assigns weights to all users in the database according to the similarity with the personal characteristics of the target user, and performs adaptive adjustment based on the weights to obtain a training data set, including: normalizing the personal characteristics, calculating the similarity between the personal characteristics of all users in the database and the personal characteristics of the target user, assigning weights to all users in the database based on the similarity so that users with relatively large similarities have relatively high weights, and resampling according to the weights to obtain the training data set; The similarity calculation method includes: weighted Euclidean distance, Manhattan distance, Mahalanobis distance, Langevin distance, and Minkowski distance.

8. The data feature extraction and analysis method based on infrared thermal imaging according to claim 1 is characterized in that: The method also includes: after predicting the core body temperature of the target user, using a medical thermometer to measure the real core body temperature of the target user, and incorporating the target user's infrared thermal imaging facial temperature distribution characteristics, personal characteristics, environmental characteristics and real core body temperature into a database for updating.

9. A core body temperature prediction system based on infrared thermal imaging, characterized in that: It is used to implement the method according to any one of claims 1 to 8, and the system comprises: The data collection unit includes a personal feature collection module, a facial temperature data collection module, an environmental data collection module, and a core body temperature collection module, which are respectively used to collect personal features, infrared thermal imaging facial temperature distribution features, environmental features, and real core body temperature data measured by a medical thermometer, and establish a database; in a single-person long-term monitoring scenario, the data in the database is input into a first similarity measurement module, and in a multi-person rapid monitoring scenario, the data is input into a coupling feature calculation module; The feature extraction unit includes a first similarity measurement module and a coupling feature calculation module. The first similarity measurement module is used to calculate the similarity between the target user and the personal features of the user in the database in a single-person long-term monitoring scenario, set a threshold, and select user data with a similarity greater than the threshold to input into the coupling feature calculation module to calculate the coupling feature; the coupling feature calculation module is used to calculate the coupling feature according to the facial temperature distribution feature and the environmental feature; the data processed by the feature extraction unit is input into the model establishment unit as a training data set in a multi-person rapid monitoring scenario, and is input into the data adaptive adjustment unit in a single-person long-term monitoring scenario; A data adaptive adjustment unit is used to receive the data processed by the feature extraction unit in a single-person long-term monitoring scenario, perform data adaptive adjustment, and obtain a training data set; the data adaptive adjustment unit includes a second similarity measurement module and a data resampling module, the second similarity measurement module evaluates the similarity between the target user and the personal features of the users in the database, and the data resampling module resamples according to the similarity to obtain the training data set, and users with relatively large similarity have relatively high weights; A model building unit, used to build a core body temperature prediction model by training based on a training data set through a machine learning method, wherein the training data set consists of personal characteristics, facial temperature distribution characteristics, environmental characteristics, coupling characteristics, and real core body temperature; The model application unit includes a core body temperature calculation module, which calculates the predicted core body temperature based on the core body temperature prediction model according to the personal characteristics, facial temperature distribution characteristics, environmental characteristics, and coupling characteristics of the target user.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it is used to implement the method according to any one of claims 1 to 8.

Images