Intelligent forehead thermometer temperature measurement precision optimization method and system based on AI

Abstract

The invention relates to the technical field of artificial intelligence, and particularly discloses an AI-based intelligent forehead thermometer temperature measurement precision optimization method and system, and the method comprises the following steps: S1, sensor data collection: collecting temperature data, environment data and image data in real time through a sensor collection terminal; s2, data preprocessing and optimization: data preprocessing is carried out, and dynamic compensation and local caching of temperature data are carried out; s3, depth analysis and model optimization: performing data storage, depth analysis and model optimization, and feeding back the optimized model to the edge computing node; according to the invention, the sensor acquisition terminal acquires temperature data by using an infrared temperature sensor, acquires environment data and image data, and eliminates the influence of hair and clothes of the forehead on the measurement accuracy in combination with the temperature measured by the forehead thermometer in real time, so that the accuracy and reliability of the acquired data are ensured; and an effective data support is provided for subsequent data analysis.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and more specifically, to an AI-based intelligent forehead thermometer temperature measurement accuracy optimization method and system. Background Art

[0002] The infrared forehead thermometer measures the forehead surface temperature through a thermopile sensor and estimates the internal body temperature through a fixed algorithm. The accuracy is affected by temperature and measurement distance, resulting in inevitable errors. The global demand for fast, non-contact temperature screening is growing rapidly. With the advancement of artificial intelligence and sensor technology, multivariate data of ambient temperature, humidity, and distance can be integrated to dynamically calibrate the measurement results. The forehead thermometer based on artificial intelligence has achieved technological progress from single-point calibration to dynamic adaptation through multi-sensor fusion, deep learning algorithms, and edge computing. While ensuring public health safety, it reduces the consumption of medical resources and provides a new health management for large populations. Provides an efficient and reliable solution.

[0003] Traditional forehead thermometers have some disadvantages in actual use. First, the temperature measurement is inaccurate. Traditional forehead thermometers rely on fixed calibration before leaving the factory and cannot dynamically compensate for environmental changes. There is no dynamic compensation algorithm and they are easily affected by environmental factors and blocked by clothes, which reduces the accuracy of temperature detection and makes the forehead thermometer ineffective in complex environments. In addition, the use of traditional forehead thermometers relies on manual operation, takes a long time to align, and has low recording efficiency. Second, the lack of dynamic compensation capability. Traditional forehead thermometers lack the support of edge computing and AI algorithms, and cannot optimize and process the data collected in real time. They cannot dynamically compensate the temperature measurement results according to environmental parameters, resulting in large measurement errors. In addition, traditional forehead thermometers cannot learn and optimize the temperature measurement model based on historical data. The accuracy of the thermometer will gradually decrease with the increase of usage time. Third, the data management capability is weak. Traditional forehead thermometers cannot store data online, and cannot quickly generate statistical reports and trace fever history based on historically measured temperature data. Manual data monitoring is required, which increases labor costs. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides an AI-based intelligent forehead thermometer temperature measurement accuracy optimization method and system, which collects temperature data and environmental data and image data that affect the measured temperature through a sensor acquisition terminal, pre-processes and optimizes the collected data through an edge computing node, and calibrates the measured temperature using a model deployed on the edge computing node. The model is stored and optimized through a cloud platform, the measured temperature is fed back to the user through a user interaction terminal, and the forehead thermometer is calibrated and optimized through a calibration maintenance terminal. The system effectively solves the problems of inaccurate temperature measurement, lack of dynamic compensation capability, and weak data management capability proposed in the background technology.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: an AI-based intelligent forehead thermometer temperature measurement accuracy optimization method, including a sensor acquisition terminal, an edge computing node, a cloud platform, a user interaction terminal and a calibration maintenance terminal, wherein the sensor acquisition terminal is connected to the edge computing node via Bluetooth, and the edge computing node, the cloud platform, the user interaction terminal and the calibration maintenance terminal are connected via the Internet, including the following steps: S1: Sensor data collection: The sensor collection terminal collects temperature data, environmental data, and image data in real time, and transmits the collected data to the edge computing node; S2: Data preprocessing and optimization: The collected data is preprocessed through edge computing nodes, dynamic temperature data compensation is performed, and local caching is performed, and the preprocessed data is uploaded to the cloud platform; S3: Deep analysis and model optimization: Data storage, deep analysis and model optimization are performed through the cloud platform, and the optimized model is fed back to the edge computing node; S4: Data display and feedback: The user interaction terminal obtains the temperature measurement results from the cloud platform, displays the temperature measurement data to the user, and provides an alarm function and feedback mechanism; S5: System calibration and maintenance: Regularly calibrate the smart forehead thermometer through the calibration and maintenance terminal, remotely monitor the system's operating status, perform fault diagnosis, and perform software upgrades and parameter adjustments on the system.

[0006] An AI-based intelligent forehead thermometer temperature measurement accuracy optimization system, including a sensor acquisition terminal, an edge computing node, a cloud platform, a user interaction terminal, and a calibration maintenance terminal; The sensor acquisition terminal is used to collect temperature data, environmental data and image data in real time, and transmit the collected data to the edge computing node; The edge computing node is used to pre-process the collected data, perform dynamic compensation of temperature data, and perform local caching, and upload the pre-processed data to the cloud platform; The cloud platform is used for data storage, in-depth analysis and model optimization, and feeds back the optimized model to the edge computing node; The user interaction terminal is used to obtain temperature measurement results from the cloud platform, display temperature measurement data to users, and provide alarm functions and feedback mechanisms; The calibration and maintenance terminal is used to regularly calibrate the intelligent forehead thermometer, remotely monitor the operating status of the system, perform fault diagnosis, and perform software upgrades and parameter adjustments on the system.

[0007] Technical effects and advantages of the present invention: 1. The present invention collects temperature data by using an infrared temperature sensor through a sensor acquisition terminal, and collects environmental data and image data, and combines the temperature measured in real time by the forehead thermometer, the temperature and humidity in the measurement environment, the measurement distance, and whether the measurement site is the forehead. The influence of the environmental temperature and humidity, the measurement distance, and the inaccurate measurement site on the measured temperature is eliminated, and the influence of forehead hair and clothing on the measurement accuracy is eliminated, thereby ensuring the accuracy and reliability of the collected data and providing effective data support for subsequent data analysis; 2. The present invention pre-processes the temperature data, environmental data and image data collected by the sensor acquisition terminal through the edge computing node, and dynamically compensates the forehead temperature of the person to be measured monitored in real time in combination with the environmental temperature and humidity and the measurement distance, and judges whether the detection part is accurate in combination with the average radiation temperature in the real-time collected image data. The detection part is adjusted in real time according to the judgment result, which improves the quality and reliability of the data, and locally caches the real-time collected and pre-processed data to ensure that the data will not be lost when the network is unstable, providing a complete data basis for subsequent cloud analysis; 3. The present invention uses a cloud platform to perform structured storage on real-time detected and pre-processed data, and records historical data and scene measurement labels, centrally processes and analyzes a large amount of data, digs out the rules between the data, and continuously optimizes the temperature measurement correction algorithm, so that the edge computing node can perform temperature correction and compensation more accurately in subsequent temperature measurements, thereby improving the temperature measurement accuracy of the forehead thermometer, and continuously optimizes the model through user feedback from the user interaction terminal, so that the temperature measured by the forehead thermometer can better meet the needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a schematic diagram of the method steps of the present invention.

[0009] Figure 2 It is a schematic diagram of the overall structure of the present invention. DETAILED DESCRIPTION

[0010] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0011] As attached Figure 1 An AI-based intelligent forehead thermometer temperature measurement accuracy optimization method and system is shown, including a sensor acquisition terminal, an edge computing node, a cloud platform, a user interaction terminal and a calibration maintenance terminal, wherein the sensor acquisition terminal is connected to the edge computing node via Bluetooth, and the edge computing node, the cloud platform, the user interaction terminal and the calibration maintenance terminal are connected via the Internet.

[0012] In a more specific application of this embodiment, the sensor acquisition terminal is connected to the edge computing node via Bluetooth, and includes an infrared temperature sensor, an ambient temperature and humidity sensor, and a visible light camera. The infrared temperature sensor is used to measure the infrared radiation of the human forehead, convert it into an electrical signal, and obtain the temperature of the human body surface. The ambient temperature and humidity sensor is used to monitor the temperature and humidity of the measurement environment in real time. The visible light camera is used to identify whether the measurement part is the head and whether the measurement distance is appropriate. The sensor acquisition terminal is used to collect temperature data of the human forehead and temperature and humidity data of the measurement environment, and provide image data to assist in judging the accuracy and effectiveness of the measurement.

[0013] The edge computing node receives data from the sensor acquisition terminal and connects to the cloud platform through the Internet. It is used to pre-process the data transmitted by the sensor acquisition terminal, including data filtering and calibration, noise and interference removal, dynamic compensation for the temperature measured by the infrared temperature sensor in combination with the ambient temperature and humidity data, and local storage of the processed data.

[0014] The cloud platform is used to receive data uploaded by edge computing nodes and is connected to user interaction terminals and calibration maintenance terminals through the Internet. It includes a server cluster and a database. The server cluster uses a cloud server with powerful computing and storage capabilities. It is used to store a large amount of temperature measurement data and run complex AI algorithms. The database is used to store historical temperature data, equipment information, and calibration records. The cloud platform is used to receive and store a large amount of temperature measurement data and related environmental data, conduct in-depth analysis and mining of the data, optimize the temperature measurement model, and realize data query, statistics, and analysis.

[0015] The user interaction terminal is connected to the cloud platform through the Internet and receives data and services provided by the cloud platform. The specific device can be a smart phone, tablet computer, or computer, which is used to display the measured temperature in real time, provide historical temperature query, and support users to set alarm thresholds. When the measured temperature exceeds the set value, an alarm is issued.

[0016] The calibration and maintenance terminal is used to achieve remote calibration and maintenance of the forehead thermometer temperature measurement accuracy optimization system. It is connected to the cloud platform through a VPN security network. It is used to regularly calibrate the forehead thermometer, remotely monitor the system's operating status, and continuously optimize system performance.

[0017] The specific implementation of the present invention comprises the following steps: S1: Sensor data acquisition: The sensor acquisition terminal collects temperature data, environmental data, and image data in real time, and transmits the collected data to the edge computing node.

[0018] Furthermore, the sensor acquisition terminal includes an infrared thermopile sensor, a temperature and humidity sensor, a distance sensor, an infrared camera, and an RGB camera. The infrared thermopile sensor is used to collect temperature data, and the temperature data is the initial temperature T raw , collect the ambient temperature T in the environmental data through the temperature and humidity sensor env and relative humidity RH, the distance d in the environmental data collected by the distance sensor, and the average radiation temperature T in the image data collected by the infrared camera image , obtain the temperature measurement points of the forehead thermometer through the RGB camera, and get the original data set D raw ={T raw , T env , RH, d, T image}.

[0019] In this embodiment, it is necessary to specifically explain that the original forehead temperature radiation value is obtained by using an infrared thermopile sensor, and is converted into a corresponding temperature value, i.e., the initial temperature, using the Stefan-Boltzmann law. Generally, the human body temperature range is 32°C-42°C. The ambient temperature during the measurement of body temperature is collected by the thermistor in the temperature and humidity sensor, and the relative humidity during the measurement of body temperature is collected by the humidity-sensitive capacitor. The humidity-sensitive capacitor uses a polymer film as a medium. When the ambient humidity changes, water molecules are adsorbed to change the dielectric constant of the capacitor, causing the capacitance value to change. RH=(C X -C 0 ) / (C 100 -C 0 )×100%, where C X Real-time capacitance value, C 0 and C 100The capacitance at 0% and 100% humidity respectively. The distance between the forehead thermometer and the person being measured is measured through the distance sensor to ensure that the measurement distance is within the optimal measurement range, which is 2-4cm. The infrared camera is used to obtain the facial image of the subject and extract the average radiation temperature of the forehead area. The RGB camera uses the key point positioning algorithm to determine the detection point of the forehead thermometer and judge whether the monitoring point is the target point, that is, the forehead. The present invention collects temperature data, environmental data and image data through the sensor terminal, which is the basis of the entire temperature measurement accuracy optimization system. It comprehensively collects data related to temperature measurement. Temperature data is the core measurement target. Environmental data is used to compensate for environmental influences on temperature measurement results. Image data is used to eliminate interference factors in the measurement process, ensuring the accuracy and reliability of the collected data, and providing effective data support for subsequent data analysis.

[0020] S2: Data preprocessing and optimization: The collected data is preprocessed through edge computing nodes, dynamic temperature data compensation is performed, and local caching is performed, and the preprocessed data is uploaded to the cloud platform.

[0021] Furthermore, the data preprocessing of the edge computing node uses a data filtering algorithm to process the collected raw data, which includes temperature data, environmental data, and image data, to remove noise and interference in the data.

[0022] It should be specifically explained in this embodiment that the temperature data and environmental data can be preprocessed by moving average filtering. By calculating the average value of the collected data in a specific time window, the time series data can be smoothed to reduce random fluctuations. For example, the moving average method with a window of 5 can be used to effectively eliminate occasional pulse interference; the image data can be preprocessed by median filtering, Gaussian filtering, bilateral filtering and fast edge-preserving filtering. Among them, the median filtering replaces the current pixel value with the median in the pixel field, which can effectively remove salt and pepper noise. The Gaussian filtering weights the pixel value in the field according to the Gaussian distribution, smoothes the image while retaining the edge, and can effectively remove Gaussian noise. The bilateral filtering combines spatial proximity and pixel value similarity to smooth the image while keeping the edge clear, which is suitable for edge computing scenarios that need to retain details. The fast edge-preserving filtering calculates the local mean square error by integrating the image to achieve fast blurring with edge retention, which is suitable for resource-constrained edge computing nodes. The appropriate filtering method is selected according to the image data and filtering requirements.

[0023] Furthermore, the temperature data dynamic compensation specifically includes the following steps: A1: The time for the forehead thermometer to measure the subject is divided into n time zones of equal length, marked as 1, 2, ..., i, ..., n in sequence, and the initial temperature T in each time zone is collected in real time through the data collection terminal. rawi 、Ambient temperature T envi , relative humidity RH i and the distance d i , and collect the average radiation temperature T image ; A2: Calculate the initial temperature mean, ambient temperature mean, relative humidity mean and distance mean through the initial temperature, ambient temperature, relative humidity and distance in any time zone, and use the formula to convert the initial temperature mean, ambient temperature mean and relative humidity mean: , Calculate the calibration temperature T comp1 , where T cal Indicates the ambient calibration temperature, RH cal represents the environmental calibration humidity, α and β represent the calibration coefficients of the ambient temperature and relative humidity respectively; A3: Use the radiation attenuation model formula to calculate the average of the temperature and distance of a calibration: , Calculate the secondary calibration temperature T comp2 , where γ is the radiation attenuation coefficient, Δd represents the distance mean, and d 0 represents the standard distance; A4: Compare the absolute value of the difference between the secondary calibration temperature and the average radiation temperature with the temperature error threshold. If it is greater than the temperature error threshold, it means that the temperature measurement of the forehead thermometer is abnormal, and the measurement position is realigned. If it is less than the temperature error threshold, upload the preprocessed data set D pro ={T comp2 , T env , RH, timestamp, userID} to the cloud platform, where timestamp represents the measurement timestamp and userID represents the identification of the person to be measured.

[0024] In this embodiment, it is necessary to specifically explain that the initial temperature measured by the forehead thermometer is compensated for ambient temperature and relative humidity to quantify the influence of ambient temperature and relative humidity on the original temperature, wherein the setting of α and β requires the influence coefficient obtained by measuring the influence of ambient temperature and relative humidity on the initial temperature for multiple times; by correcting the distance factor, the influence of the distance between the forehead thermometer and the forehead of the person to be measured on the measured temperature is eliminated, and by comparing the secondary calibration temperature with the average radiation temperature, the interference of hair occlusion or measurement position deviation on the measurement result is eliminated; The present invention improves the quality and reliability of data by preliminarily processing and optimizing the collected raw data, removing noise interference, and reducing the influence of environmental factors and measurement distance on the temperature measurement results. It also locally caches the real-time collected and preprocessed data to ensure that the data will not be lost when the network is unstable, thereby providing a complete data foundation for subsequent cloud analysis.

[0025] S3: Deep analysis and model optimization: Data storage, deep analysis and model optimization are performed through the cloud platform, and the optimized model is fed back to the edge computing node.

[0026] Furthermore, data storage specifically involves structured storage of data transmitted by edge computing nodes, and recording of historical data and scene measurement labels; deep analysis specifically involves error analysis and cluster analysis of data transmitted by edge computing nodes; model optimization specifically involves using a neural network model to update compensation model parameters, and pushing the optimized model parameters to edge computing nodes for subsequent temperature measurements.

[0027] What needs to be specifically explained in this embodiment is that the measurement scene information specifically includes indoor / outdoor and season, etc. Error analysis specifically refers to quantitatively evaluating the accuracy and stability of the measurement results by calculating the error between the secondary calibration temperature and the reference temperature, and statistically calculating the mean and variance of the error. Cluster analysis specifically refers to identifying high-frequency error scenes through clustering algorithms, and locating key factors that affect measurement accuracy under specific conditions, such as negative deviation scenes where the measured values ​​are generally low in a high humidity environment. Model optimization specifically refers to updating the compensation model parameters by minimizing the mean absolute error, where the objective function is: , Where N is the number of samples, T rawi , T envi RH i and i are the initial temperature, ambient temperature, relative humidity and distance respectively. f(x) is the model function. By minimizing the loss function, the accuracy of the model's temperature correction is improved.

[0028] The present invention centrally processes and analyzes a large amount of data through a cloud platform, digs out the patterns between the data, and uses a machine learning model to continuously optimize the temperature measurement correction algorithm. Through continuous updating of the model, the edge computing node can perform temperature correction and compensation more accurately in subsequent temperature measurements, thereby improving the temperature measurement accuracy of the forehead thermometer.

[0029] S4: Data display and feedback: The user interaction terminal obtains the temperature measurement results from the cloud platform, displays the temperature measurement data to the user, and provides an alarm function and feedback mechanism.

[0030] Furthermore, the user interaction terminal obtains the temperature measurement results from the cloud platform through the Internet, and displays the temperature measurement data to the user in a digital form. The user can view the temperature measurement data at any time, build an alarm mechanism, and set the temperature alarm threshold. When the measured temperature exceeds the temperature alarm threshold, the user interaction terminal sends an alarm message to remind relevant personnel to take solutions. The user provides feedback on the temperature measurement results through the user interaction terminal, and the user feedback information is transmitted to the cloud platform to further optimize the model.

[0031] What needs to be specifically explained in this embodiment is that the user interaction terminal has the functions of data storage and real-time data refresh. When the user views single data, the real-time temperature measurement information can be retrieved from the cloud platform. When the user needs to obtain long-term data, the user interaction terminal retrieves the historical storage data from the cloud platform for the user to view the relevant content in real time; when the measured temperature exceeds the temperature alarm threshold, the user interaction terminal can issue an alarm message through sound, vision or vibration, where the sound is a "di di" alarm sound, the vision is a flashing screen and a red warning mark, and the vibration is a vibration of the mobile phone; the solution is that the user can measure again to confirm the physical condition and seek medical treatment in time. When applied to large-scale scene screening, it reminds the administrator to conduct further physical examinations on the person being measured to avoid potential health risks.

[0032] The present invention provides users with a convenient interactive experience through a user interaction terminal, which is convenient for users to understand the temperature measurement results of the forehead thermometer in time. The alarm function helps users to discover abnormal body temperature in time and ensure the timeliness of user health monitoring. The user's usage needs are collected through a feedback mechanism, and more optimization data are provided for the optimization model, so that the temperature measured by the forehead thermometer can better meet the needs of users.

[0033] S5: System calibration and maintenance: Regularly calibrate the smart forehead thermometer through the calibration and maintenance terminal, remotely monitor the system's operating status, perform fault diagnosis, and perform software upgrades and parameter adjustments on the system.

[0034] Furthermore, the calibration and maintenance terminal uses high-precision calibration equipment to regularly calibrate the forehead thermometer, remotely monitors the temperature measurement accuracy of the intelligent forehead thermometer through the network to optimize the operating status of the system, obtains the performance indicators of the system in real time, and diagnoses the system when abnormal performance indicators are found to find out the cause of the fault, and performs software upgrades and parameter adjustments on the system based on the cause of the fault.

[0035] What needs to be specifically explained in this embodiment is that the high-precision calibration equipment can be a Fluke blackbody furnace, which can provide a stable and accurate standard reference temperature. By comparing the temperature measured by the forehead thermometer with the standard temperature value of the blackbody furnace, the calibration parameters inside the forehead thermometer are adjusted according to the deviation value to ensure the accuracy of the forehead thermometer measurement; by establishing a secure connection with the help of VPN, the stable interaction of monitoring instructions and data is ensured. The performance indicators can be the temperature measurement response time, the temperature measurement error range, the device online rate and the data upload success rate, which reflect the operation of each component in the system in real time. For example, if the temperature measurement error range exceeds the allowable range, the forehead thermometer needs to be calibrated, the device online rate reflects whether the device is operating normally, and the data upload success rate reflects whether the status of cloud communication is normal. Software adjustment specifically refers to the method of remotely pushing upgrade packages to repair software vulnerabilities, optimize algorithms and add new functions to improve the overall performance of the system; parameter adjustment refers to adjusting relevant parameters according to the type of fault to restore the system to normal operation and optimize measurement parameters.

[0036] The present invention ensures that the smart forehead thermometer maintains high-precision performance through the calibration and maintenance terminal. Regular calibration of the forehead thermometer can eliminate errors that occur in the sensor during long-term use and ensure the accuracy of the measurement results. Remote monitoring and fault diagnosis can promptly discover problems that occur during system operation and eliminate the impact of inaccurate measurements caused by faults, so that the system can continuously adapt to new environments and requirements and improve the performance and stability of the system.

[0037] like Figure 2 The embodiment shown provides an AI-based intelligent forehead thermometer temperature measurement accuracy optimization system, including a sensor acquisition terminal, an edge computing node, a cloud platform, a user interaction terminal, and a calibration and maintenance terminal.

[0038] Sensor acquisition terminal: collects temperature data, environmental data and image data in real time, and transmits the collected data to the edge computing node.

[0039] Edge computing node: pre-processes the collected data, performs dynamic compensation for temperature data, caches it locally, and uploads the pre-processed data to the cloud platform.

[0040] Cloud platform: performs data storage, in-depth analysis, and model optimization, and feeds the optimized model back to the edge computing node.

[0041] User interaction terminal: obtains temperature measurement results from the cloud platform, displays temperature measurement data to users, and provides alarm function and feedback mechanism.

[0042] Furthermore, the user interaction terminal specifically includes a smart phone, a tablet computer, a personal computer and a medical display interface, which realizes the functions of temperature data display, alarm setting and user feedback through an application or a network interface.

[0043] Calibration and maintenance terminal: Regularly calibrate the smart forehead thermometer, remotely monitor the system's operating status, perform fault diagnosis, and perform software upgrades and parameter adjustments on the system.

[0044] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other; Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing the temperature measurement accuracy of an intelligent forehead thermometer based on AI, characterized in that: It includes a sensor acquisition terminal, an edge computing node, a cloud platform, a user interaction terminal and a calibration maintenance terminal, wherein the sensor acquisition terminal is connected to the edge computing node via Bluetooth, and the edge computing node, the cloud platform, the user interaction terminal and the calibration maintenance terminal are connected via the Internet, including the following steps: S1: Sensor data collection: The sensor collection terminal collects temperature data, environmental data, and image data in real time, and transmits the collected data to the edge computing node; S2: Data preprocessing and optimization: The collected data is preprocessed through edge computing nodes, dynamic temperature data compensation is performed, and local caching is performed, and the preprocessed data is uploaded to the cloud platform; S3: Deep analysis and model optimization: Data storage, deep analysis and model optimization are performed through the cloud platform, and the optimized model is fed back to the edge computing node; S4: Data display and feedback: The user interaction terminal obtains the temperature measurement results from the cloud platform, displays the temperature measurement data to the user, and provides an alarm function and feedback mechanism; S5: System calibration and maintenance: Regularly calibrate the smart forehead thermometer through the calibration and maintenance terminal, remotely monitor the system's operating status, perform fault diagnosis, and perform software upgrades and parameter adjustments on the system.

2. According to claim 1, a method for optimizing temperature measurement accuracy of an AI-based intelligent forehead thermometer is characterized in that: The sensor acquisition terminal includes an infrared thermopile sensor, a temperature and humidity sensor, a distance sensor, an infrared camera and an RGB camera. The infrared thermopile sensor is used to collect temperature data. The temperature data is the initial temperature T raw , collect the ambient temperature T in the environmental data through the temperature and humidity sensor env and relative humidity RH, the distance d in the environmental data collected by the distance sensor, and the average radiation temperature T in the image data collected by the infrared camera image , obtain the temperature measurement points of the forehead thermometer through the RGB camera, and get the original data set D raw ={T raw , T env , RH, d, T image }.

3. According to the AI-based intelligent forehead thermometer temperature measurement accuracy optimization method of claim 1, it is characterized by: The data preprocessing of the edge computing node uses a data filtering algorithm to process the collected raw data, which includes temperature data, environmental data, and image data, to remove noise and interference in the data.

4. According to claim 1, a method for optimizing temperature measurement accuracy of an AI-based intelligent forehead thermometer is characterized in that: The temperature data dynamic compensation specifically comprises the following steps: A1: The time for the forehead thermometer to measure the subject is divided into n time zones of equal length, marked as 1, 2, ..., i, ..., n in sequence, and the initial temperature T in each time zone is collected in real time through the data collection terminal. rawi 、Ambient temperature T envi , relative humidity RH i and the distance d i , and collect the average radiation temperature T image ; A2: Calculate the initial temperature mean, ambient temperature mean, relative humidity mean and distance mean through the initial temperature, ambient temperature, relative humidity and distance in any time zone, and use the formula to convert the initial temperature mean, ambient temperature mean and relative humidity mean: ， Calculate the calibration temperature T comp1 , where T cal Indicates the ambient calibration temperature, RH cal represents the environmental calibration humidity, α and β represent the calibration coefficients of the ambient temperature and relative humidity respectively; A3: Use the radiation attenuation model formula to calculate the average of the temperature and distance of a calibration: ， Calculate the secondary calibration temperature T comp2 , where γ is the radiation attenuation coefficient, Δd represents the mean distance, and d0 represents the standard distance; A4: Compare the absolute value of the difference between the secondary calibration temperature and the average radiation temperature with the temperature error threshold. If it is greater than the temperature error threshold, it means that the temperature measurement of the forehead thermometer is abnormal, and the measurement position is realigned. If it is less than the temperature error threshold, upload the preprocessed data set D pro ={T comp2 , T env , RH, timestamp, userID} to the cloud platform, where timestamp represents the measurement timestamp and userID represents the identification of the person to be measured.

5. According to the AI-based intelligent forehead thermometer temperature measurement accuracy optimization method of claim 1, it is characterized by: The data storage specifically includes structured storage of data transmitted by edge computing nodes, and recording of historical data and scene measurement labels; the deep analysis specifically includes error analysis and cluster analysis of data transmitted by edge computing nodes; the model optimization specifically includes using a neural network model to update compensation model parameters, and pushing the optimized model parameters to the edge computing nodes for subsequent temperature measurements.

6. According to the AI-based intelligent forehead thermometer temperature measurement accuracy optimization method of claim 1, it is characterized by: The user interaction terminal obtains the temperature measurement result from the cloud platform through the Internet, and displays the temperature measurement data to the user in a digital display form. The user can view the temperature measurement data at any time, build an alarm mechanism, and set a temperature alarm threshold. When the measured temperature exceeds the temperature alarm threshold, the user interaction terminal sends an alarm message to remind relevant personnel to take solutions. The user feeds back the temperature measurement result through the user interaction terminal, and the user feedback information is transmitted to the cloud platform to further optimize the model.

7. According to claim 1, a method for optimizing temperature measurement accuracy of an AI-based intelligent forehead thermometer is characterized in that: The calibration and maintenance terminal uses high-precision calibration equipment to regularly calibrate the forehead thermometer, remotely monitors the temperature measurement accuracy of the intelligent forehead thermometer through the network, optimizes the operating status of the system, obtains the performance indicators of the system in real time, and diagnoses the system when abnormal performance indicators are found to find out the cause of the fault, and performs software upgrades and parameter adjustments on the system according to the cause of the fault.

8. An AI-based intelligent forehead thermometer temperature measurement accuracy optimization system, characterized in that: An AI-based intelligent forehead thermometer temperature measurement accuracy optimization method applied to any one of claims 1-7, comprising a sensor acquisition terminal, an edge computing node, a cloud platform, a user interaction terminal, and a calibration maintenance terminal; The sensor acquisition terminal is used to collect temperature data, environmental data and image data in real time, and transmit the collected data to the edge computing node; The edge computing node is used to pre-process the collected data, perform dynamic compensation of temperature data, and perform local caching, and upload the pre-processed data to the cloud platform; The cloud platform is used for data storage, in-depth analysis and model optimization, and feeds back the optimized model to the edge computing node; The user interaction terminal is used to obtain temperature measurement results from the cloud platform, display temperature measurement data to users, and provide alarm functions and feedback mechanisms; The calibration and maintenance terminal is used to regularly calibrate the intelligent forehead thermometer, remotely monitor the operating status of the system, perform fault diagnosis, and perform software upgrades and parameter adjustments on the system.

9. According to claim 1, the AI-based intelligent forehead thermometer temperature measurement accuracy optimization system is characterized by: User interaction terminals specifically include smart phones, tablet computers, personal computers and medical display interfaces, which realize functions of temperature data display, alarm setting and user feedback through applications or network interfaces.

Images