Infrared temperature-sensing based student dormitory health monitoring and early warning system and method

Abstract

The present application relates to the technical field of dormitory health early warning, and discloses a student dormitory health monitoring and early warning system and method based on infrared temperature sensing, which comprises a sensing layer, a benchmark self-correction layer, a network layer, a platform layer and an application layer, and combines a mobile temperature calibration device, an array calibration control module and a data linkage processing module to realize unmanned automatic online calibration of an infrared thermal imaging sensor array; the method realizes deep linkage of calibration results, body temperature monitoring data and early warning algorithm parameters through a closed-loop process of calibration triggering, calibration execution, calibration curve generation, data linkage correction and monitoring and early warning, and solves the problems of long-term operation precision reduction of the sensor and monitoring interruption caused by manual intervention during calibration on the basis of retaining the core advantages of original system non-contact continuous monitoring and CUSUM algorithm trend early warning.

Description

Technical Field

[0001] This invention relates to the field of dormitory health early warning technology, specifically to a student dormitory health monitoring and early warning system and method based on infrared temperature sensing. Background Technology

[0002] Traditional student dormitory health management practices primarily rely on manual temperature monitoring methods, such as using forehead or ear thermometers for regular or random checks. This method suffers from significant operational inefficiency, as it requires administrators to measure each student individually, which is not only time-consuming and labor-intensive but also difficult to cover all students, especially in high-density living environments like eight-person rooms, where missed checks are easily caused by staff movement or scheduling conflicts. Furthermore, manual monitoring lacks continuity, typically limited to specific time slots, such as morning or evening checks, making 24 / 7 monitoring impossible. This makes it difficult to detect sudden health abnormalities at night or in a timely manner. More importantly, close-contact measurement increases the risk of cross-infection, especially during periods of high infectious disease prevalence, potentially exacerbating the spread of outbreaks and contradicting the safety goals of campus public health management.

[0003] While existing infrared temperature measurement technology has achieved non-contact detection to some extent, most applications are limited to single-point sensors, such as handheld infrared thermometers or fixed single-point thermal probes. These devices cannot meet the complex needs of multi-person scenarios. In dormitory environments, students frequently enter and exit, and multiple people move around simultaneously. Single-point detection struggles to accurately distinguish individual targets and is easily affected by environmental interference, leading to measurement errors or false alarms. More importantly, existing technologies focus on instantaneous threshold alarms, such as triggering an alarm when body temperature exceeds 37.3℃. They lack the ability to track and analyze body temperature trends over a long period and cannot identify subtle but continuous abnormal changes, such as low-grade fever or a slowly rising temperature curve. This results in a delayed early warning mechanism, hindering early intervention. Overall, existing technologies have significant shortcomings in multi-target adaptability, continuous monitoring, and intelligent early warning, failing to meet the urgent needs of vocational college dormitories for efficient, safe, and intelligent health management.

[0004] In view of this, the applicant conducted in-depth research on the above-mentioned issues, which led to this case. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a health monitoring and early warning system and method for student dormitories based on infrared temperature sensing, which can effectively solve the aforementioned technical problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A student dormitory health monitoring and early warning system based on infrared temperature sensing includes a sensing layer, a reference self-correction layer, a network layer, a platform layer, and an application layer. The perception layer includes multiple infrared thermal imaging sensor arrays, environmental temperature and humidity sensors, wind speed sensors, edge computing units, and distributed calibration points deployed in student dormitories. The distributed calibration points are set within the effective detection range of the infrared thermal imaging sensor arrays. The reference self-correction layer includes a mobile temperature calibration device, a calibration data acquisition module, and an array calibration control module. The mobile temperature calibration device includes a standard temperature source, a data measurement module, a movement module, and a power supply module. The human body simulation module of the standard temperature source is equipped with a heat spreader and bionic skin. The movement module uses a lidar positioning system and achieves autonomous navigation through the SLAM algorithm. The network layer uses Wi-Fi, Ethernet, or LoRa communication protocols and is based on the lightweight MQTT data protocol to achieve bidirectional low-latency transmission of monitoring data, calibration commands, and calibration data. The platform layer includes a cloud server, a health monitoring algorithm module, a calibration algorithm module, and a data linkage processing module. The health monitoring algorithm module is equipped with the CUSUM trend detection algorithm, and the data linkage processing module establishes a linkage feedback mechanism between calibration data, monitoring data, and early warning parameters. The application layer includes a web management dashboard, mobile push notification functionality, and a calibration management module, supporting visualized monitoring and early warning, calibration status monitoring, and remote manual calibration command issuance.

[0007] Furthermore, an infrared thermal imaging sensor array is deployed inside the dormitory door, 2.5–3 meters above the ground. The edge computing unit performs region segmentation, target extraction, and local audible and visual alarm operations. When the compensated temperature exceeds 37.3°C, the local audible and visual alarm is triggered.

[0008] Furthermore, the first temperature sensor of the mobile temperature calibration device is located at the center of the heat spreader, the data measurement module includes a second temperature sensor and a wind speed sensor, and the array calibration control module can set the temperature calibration benchmark, wind speed calibration benchmark and corresponding deviation threshold, and supports three calibration modes: threshold trigger, timed trigger and manual trigger.

[0009] Furthermore, the CUSUM algorithm formula for the health monitoring algorithm module is: ; in This is the current temperature reading. The baseline mean is based on individual temperature data over the past 14 days. To allow for fluctuations, As a threshold, when Individual trend warnings are triggered in real time; at the same time, the module can calculate the average temperature index and abnormality rate of the dormitory building and floors. When the abnormality rate rises by more than 10% within 2 hours, a group warning is triggered.

[0010] Furthermore, the data linkage processing module's feedback includes: applying the calibration curves and array calibration models generated by the calibration algorithm module to accurately correct the raw body temperature data; and dynamically adjusting the CUSUM algorithm based on the calibrated temperature measurement accuracy. , , Abnormal proportion thresholds for parameters and population health analysis.

[0011] A method for a student dormitory health monitoring and early warning system based on the above-mentioned infrared temperature sensing includes the following steps: S1: Calibration trigger: The array calibration control module sets the temperature calibration benchmark, wind speed calibration benchmark, temperature deviation threshold and wind speed deviation threshold. The ambient temperature and humidity sensor and wind speed sensor collect environmental parameters in real time. When the deviation reaches the threshold, the preset timed calibration time is reached or a manual calibration command is received, the calibration process is triggered and a calibration notification is sent to the platform layer. S2: Calibration execution. The array calibration control module plans the movement path and controls the mobile temperature calibration device to move autonomously to the distributed calibration measurement points. The temperature control module sets multiple temperature control points and adjusts the output temperature of the standard temperature source. The calibration data acquisition module synchronously collects sensor measurement data, standard temperature data and environmental parameters. The edge computing unit cleans the data and uploads it to the platform layer. S3: Calibration curve generation. The platform-level calibration algorithm module performs correlation analysis on the collected data, generates single-sensor calibration curves, and fuses them to establish an array calibration model. The calibration results are stored on the cloud server. S4: Data linkage correction. The data linkage processing module applies the calibration curve and array calibration model to the raw body temperature data to complete accurate calibration, while dynamically adjusting the core parameters of the health monitoring algorithm module. S5: Monitoring and early warning. Based on calibrated effective body temperature data and corrected early warning parameters, the system performs individual body temperature trend detection and group health analysis. When abnormalities occur, it pushes graded early warning information through the application layer, and the edge computing unit triggers local audible and visual alarms when the temperature exceeds the threshold.

[0012] Furthermore, in step S1, the temperature calibration benchmark is set to 25℃ by default, the wind speed calibration benchmark is set to 0.2m / s by default, the temperature deviation threshold is set to ±2℃, and the wind speed deviation threshold is set to ±0.3m / s; in step S2, the temperature fluctuation of the standard temperature source is controlled within ±0.1℃.

[0013] Furthermore, in step S2, the multiple temperature control points adopt a 3-point calibration method or a 5-point calibration method. The control points for the 3-point calibration method are 35℃, 37.5℃, and 42℃, and the control points for the 5-point calibration method are 35℃, 36.5℃, 37.5℃, 39℃, and 41℃. The temperature control points cover the range of human body temperature measurement.

[0014] Furthermore, it also includes a pre-calibration step for the mobile temperature calibration device: the mobile temperature calibration device is placed in a constant temperature space, a surface temperature sensor is installed on the outside of the bionic skin, and monitoring data is collected by changing the ambient temperature and air flow rate. A first calibration equation between the environmental parameters and the data from the first temperature sensor, and a second calibration equation between the environmental parameters and the temperature of the temperature control module are established to achieve precise temperature control of the standard temperature source.

[0015] Furthermore, step S3 includes a calibration accuracy assessment step. The platform layer assesses the calibration accuracy of each infrared sensor. If the temperature measurement error is greater than ±0.2℃, a sensor fault alert is sent to the management personnel through the application layer. The raw body temperature data collected by the sensing layer is first processed by the environmental temperature and humidity compensation formula. After compensation, precise calibration is then performed using calibration curves and array calibration models. For compensation coefficient, The current ambient temperature. Calibrate the ambient temperature.

[0016] The present invention has the following beneficial effects: 1. This invention solves the problems of time-consuming, labor-intensive, and monitoring interruptions associated with traditional calibration methods. It supports three triggering modes: threshold, timed, and manual, adapting to the management needs of campus residential environments and reducing labor costs. Furthermore, through a dual compensation mechanism of environmental temperature and humidity compensation and automatic online calibration, combined with precise correction of wind speed parameters, the long-term temperature measurement error of the sensor is controlled within ±0.2℃, offsetting the effects of environmental fluctuations and equipment aging.

[0017] 2. This invention establishes a linkage feedback mechanism between calibration data, monitoring data, and early warning parameters, dynamically adjusting the early warning algorithm parameters to match the early warning strategy with the current measurement accuracy of the sensor. This reduces the false alarm rate to below 1% and increases the average early warning time to 2.5 hours. Furthermore, the five-layer architecture retains all functional modules of the original system, and the newly added reference self-correction layer can flexibly adapt to different models of infrared thermal imaging sensor arrays. The system has strong compatibility and can be quickly expanded to other collective living scenarios such as factory dormitories and nursing homes.

[0018] 3. The application layer of this invention adds a calibration management module, which realizes the visualization of calibration status, calibration records and accuracy analysis. Managers can remotely monitor the sensor status and receive fault reminders in a timely manner, realizing the integration of campus public health management and equipment management. Attached Figure Description

[0019] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see the appendix Figure 1 -Appendix Figure 2 This invention provides a student dormitory health monitoring and early warning system based on infrared temperature sensing, comprising a sensing layer, a reference self-correction layer, a network layer, a platform layer, and an application layer.

[0022] The perception layer comprises multiple infrared thermal imaging sensor arrays deployed within student dormitories, environmental temperature and humidity sensors, edge computing units, and distributed calibration points adapted to the sensor arrays. The infrared thermal imaging sensor arrays utilize either the AMG8833 (8×8 pixels) or MLX90640 (32×24 pixels) models, deployed above the dormitory doors (2.5–3 meters above the ground), employing a top-down angle to cover the entrance area and ensure effective scanning of students' foreheads or faces as they pass by. The sampling frequency is set to 1 time / second by default, which can be reduced to 1 time / 5 seconds at night to save energy. The infrared thermal imaging sensor arrays connect to the edge computing units via I2C or SPI interfaces for non-contact, continuous acquisition of student body surface temperature data. The environmental temperature and humidity sensors, using the SHT30 or DHT22 models, collect environmental temperature and humidity data to perform basic compensation calibration on the body surface temperature data; a wind speed sensor is also added to collect real-time wind speed data within the dormitory, providing environmental parameters for automatic online calibration. The edge computing unit uses a Raspberry Pi 4B or Jetson Nano embedded platform to handle local data preprocessing, initial early warning, local execution of calibration commands, and initial acquisition of calibration data. Distributed calibration measurement points are set within the effective detection range of each infrared thermal imaging sensor array, providing preset calibration locations for the temperature calibration device and adapting to the synchronous and step-by-step calibration requirements of multiple sensor arrays.

[0023] The reference self-correction layer includes a mobile temperature calibration device, a calibration data acquisition module, and an array calibration control module, enabling unattended automatic online calibration of the infrared thermal imaging sensor array in the sensing layer. Specifically, the mobile temperature calibration device includes a standard temperature source, a data measurement module, a movement module, and a power supply module. The standard temperature source includes a temperature control module, a first temperature sensor, and a human body simulation module. The data measurement module includes a second temperature sensor and a wind speed sensor. It can autonomously move to various distributed calibration measurement points via the movement module, simulating human body temperature characteristics and outputting a standard temperature. The calibration data acquisition module works in conjunction with the edge computing unit of the sensing layer and the mobile temperature calibration device to collect measurement data from each infrared sensor at different standard temperatures, as well as environmental temperature, humidity, and wind speed data during the calibration process. The array calibration control module, deployed in the edge computing unit, is used to set calibration benchmarks and deviation thresholds, determine calibration trigger conditions, and control the movement and positioning of the mobile temperature calibration device, enabling synchronous or step-by-step calibration of multiple sensor arrays.

[0024] The network layer adopts Wi-Fi, Ethernet or LoRa communication protocols and is based on the lightweight MQTT data protocol. It not only achieves low-latency transmission of monitoring data, but also completes bidirectional transmission of calibration commands and calibration data between the sensing layer, the reference self-correction layer and the platform layer. The transmission latency of the entire process is less than 3 seconds.

[0025] The platform layer, comprising a cloud server, a health monitoring algorithm module, a calibration algorithm module, and a data linkage processing module, is the core processing unit of the system. Specifically, the cloud server receives monitoring and calibration data, enabling centralized storage and management. The health monitoring algorithm module retains the original CUSUM trend detection algorithm, individual body temperature baseline modeling, and group health analysis algorithm, enabling real-time monitoring and trend warnings of student body temperatures. The calibration algorithm module generates calibration curves for each infrared sensor based on calibration data, establishing a calibration model for the multi-sensor array to achieve accurate calibration of sensor measurement data. It also supports dynamic adjustment of calibration parameters to adapt to changes in the dormitory environment. The data linkage processing module establishes a feedback mechanism between calibration and monitoring data, feeding back the sensor calibration results to the health monitoring algorithm module in real time to correct body temperature measurement data and warning parameters, achieving closed-loop management of calibration, monitoring, and warning.

[0026] The application layer includes a web management dashboard and mobile push notification functionality. While retaining the original real-time monitoring map, abnormal alarm list, historical trend statistics, and anonymous query functionality for students, a new calibration management module has been added. This module displays sensor calibration status, calibration records, accuracy analysis, calibration reminders, and other information, and supports real-time heatmaps, warning lists, and calibration status monitoring. The mobile app pushes tiered warnings via WeChat mini-programs, including alerts, warnings, and critical warnings. The student mini-program supports anonymous querying of individual body temperature trends.

[0027] In this embodiment, the health monitoring algorithm module The algorithm formula is: ;in This is the current temperature reading. The baseline mean is based on individual temperature data over the past 14 days. To allow for fluctuations, As a threshold, when The module triggers individual trend warnings; simultaneously, it calculates the average temperature index and abnormality rate for the dormitory building and floors, triggering a group warning when the abnormality rate rises by more than 10% within 2 hours. The linkage feedback from the linkage processing module includes: applying the calibration curve and array calibration model generated by the calibration algorithm module to accurately correct the original body temperature data; and dynamically adjusting the measurement based on the calibrated temperature accuracy. Algorithm , , Abnormal proportion thresholds for parameters and population health analysis.

[0028] In this embodiment, the human body simulation module of the mobile temperature calibration device is located inside the insulation layer, and bionic skin covers the surface of the heat spreader to simulate the thermal properties of human skin. The first temperature sensor is located at the center of the heat spreader to accurately measure the output temperature of the standard temperature source. The mobile module uses a lidar positioning system. The algorithm enables precise positioning and autonomous navigation, allowing it to move autonomously between dormitory corridors and various distributed calibration measurement points.

[0029] In this embodiment, the array calibration control module presets temperature deviation thresholds and wind speed deviation thresholds. When the deviation between the real-time environmental parameters collected by the ambient temperature and humidity sensors and the calibration benchmark reaches the threshold, the calibration process is automatically triggered. It also supports timed calibration according to preset cycles, such as 7 days or 15 days, and manual calibration initiated by management personnel.

[0030] A self-calibration method for a student dormitory health monitoring and early warning system based on infrared temperature sensing, specifically including the following steps: S1: Calibration Trigger The array calibration control module sets the temperature calibration benchmark (default 25℃), wind speed calibration benchmark (default 0.2m / s), temperature deviation threshold (±2℃), and wind speed deviation threshold (±0.3m / s). The ambient temperature and humidity sensors and wind speed sensors in the sensing layer collect environmental parameters in the dormitory in real time. The array calibration control module calculates the deviation between the real-time parameters and the calibration benchmark. If the deviation reaches the threshold, or the preset timed calibration time is reached, or a manual calibration instruction is received from the administrator, the calibration process is automatically triggered, and a calibration notification is sent to the platform layer through the network layer.

[0031] S2 calibration execution: S21: The array calibration control module plans the optimal movement path of the mobile temperature calibration device based on the distribution of the infrared sensor array in the sensing layer. It controls the device to move autonomously from the standby charging position to the first distributed calibration measurement point through the mobile module, thereby achieving precise positioning.

[0032] S22: The temperature control module of the mobile temperature calibration device is set with multiple temperature control points to cover the range of human body temperature. It preferably uses a three-point calibration method of 35℃, 37.5℃, and 42℃ or a five-point calibration method of 35℃, 36.5℃, 37.5℃, 39℃, and 41℃. The output temperature of the standard temperature source is adjusted sequentially, and the temperature of the heat spreader is monitored by the first temperature sensor to ensure that the temperature fluctuation does not exceed ±0.1℃ within the preset time.

[0033] S23: The calibration data acquisition module synchronously acquires the measurement data of each infrared sensor at the current standard temperature; the standard temperature data monitored by the first temperature sensor; and the calibration environment temperature, humidity, and wind speed data acquired by the second temperature sensor and the wind speed sensor.

[0034] S24: Move the mobile temperature calibration device to the remaining distributed calibration measurement points in sequence, repeat steps S22-S23, complete the calibration data acquisition for all infrared thermal imaging sensor arrays, and return to the standby charging position after the acquisition is completed.

[0035] S25: The edge computing unit performs preliminary cleaning of the calibration data, removes outliers, and then uploads it to the calibration algorithm module of the platform layer through the network layer.

[0036] S3: Calibration curve generation: The calibration algorithm module of the S31 platform layer receives calibration data, performs correlation analysis on the measurement data of each infrared sensor and the standard temperature data, and combines the ambient temperature, humidity and wind speed data during the calibration process to generate a single sensor calibration curve through mathematical fitting, establishing the correspondence between the standard temperature and the measured value.

[0037] S32: For infrared thermal imaging sensor arrays, the calibration algorithm module integrates the calibration curves of each individual sensor to establish an array calibration model, adapting to the needs of multi-sensor collaborative temperature measurement scenarios.

[0038] S33: The calibration algorithm module stores the calibration curve and array calibration model to the cloud server and sends a calibration completion notification to the application layer, displaying the calibration accuracy and status of each sensor.

[0039] S4: Data linkage correction. The platform-level data linkage processing module feeds back the calibration results to the health monitoring algorithm module in real time, achieving two core corrections: Temperature data correction: The raw body temperature data collected by the sensing layer is first compensated for by ambient temperature and humidity, and then precisely calibrated by the corresponding calibration curve / array calibration model to obtain the final effective body temperature data. Warning parameter correction: Dynamically adjusted based on the calibrated temperature measurement accuracy. The algorithm's parameters and the abnormal proportion threshold for population health analysis ensure that the warning parameters match the current sensor measurement accuracy.

[0040] S5: Monitoring and Early Warning. Based on calibrated valid body temperature data and corrected early warning parameters, the system executes the original health monitoring and early warning process: S51: The health monitoring algorithm module performs individual body temperature baseline modeling. Trend detection: If an abnormal body temperature trend is detected, an individual alert is triggered.

[0041] S52: Calculate the average temperature index and abnormality rate of the dormitory building and floors. If the abnormality rate rises significantly in a short period of time, trigger a group warning.

[0042] S53: Application layer through The management dashboard displays real-time monitoring data and early warning information after calibration, and pushes tiered early warnings to counselors via WeChat mini-program or SMS. Students can anonymously query their personal body temperature trends after calibration.

[0043] S54: If the edge computing unit detects that the calibrated body temperature data exceeds the instantaneous threshold, such as 37.3℃, it will immediately trigger a local audible and visual alarm.

[0044] In this embodiment, a pre-calibration step for the mobile temperature calibration device is also included. Before system deployment, the mobile temperature calibration device is placed in a constant temperature space, and the surface temperature of the bionic skin is measured by a surface temperature sensor. A calibration equation is established based on the ambient temperature and humidity, wind speed and the data from the first temperature sensor, so as to achieve precise temperature control of the standard temperature source and ensure calibration accuracy.

[0045] In this embodiment, after calibration, the platform layer evaluates the calibration accuracy of each infrared sensor. If the calibration accuracy of a sensor fails to meet the preset standard, such as a temperature measurement error > ±0.2℃, the application layer sends a sensor fault alert to the management personnel, prompting timely repair or replacement. The raw body temperature data collected by the sensing layer is first processed using the environmental temperature and humidity compensation formula. After compensation, precise calibration is then performed using calibration curves and array calibration models. The compensation coefficient was calibrated to be 0.1–0.3. The current ambient temperature. The calibration ambient temperature is set to 25℃ by default. After compensation, the error is reduced to below ±0.2℃.

[0046] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A student dormitory health monitoring and early warning system based on infrared temperature sensing, characterized in that, It includes the perception layer, the baseline self-correction layer, the network layer, the platform layer, and the application layer; The perception layer includes multiple infrared thermal imaging sensor arrays, environmental temperature and humidity sensors, wind speed sensors, edge computing units, and distributed calibration points deployed in student dormitories. The distributed calibration points are set within the effective detection range of the infrared thermal imaging sensor arrays. The reference self-correction layer includes a mobile temperature calibration device, a calibration data acquisition module, and an array calibration control module. The mobile temperature calibration device includes a standard temperature source, a data measurement module, a movement module, and a power supply module. The human body simulation module of the standard temperature source is equipped with a heat spreader and bionic skin. The movement module uses a lidar positioning system and achieves autonomous navigation through the SLAM algorithm. The network layer uses Wi-Fi, Ethernet or LoRa communication protocols, and implements bidirectional low-latency transmission of monitoring data, calibration commands and calibration data based on the lightweight MQTT data protocol; The platform layer includes a cloud server, a health monitoring algorithm module, a calibration algorithm module, and a data linkage processing module. The health monitoring algorithm module is equipped with the CUSUM trend detection algorithm, and the data linkage processing module establishes a linkage feedback mechanism between calibration data, monitoring data, and early warning parameters. The application layer includes a web management dashboard, mobile push notification functionality, and a calibration management module, supporting visualized monitoring and early warning, calibration status monitoring, and remote manual calibration command issuance.

2. The student dormitory health monitoring and early warning system based on infrared temperature sensing according to claim 1, characterized in that, An infrared thermal imaging sensor array is deployed inside the dormitory door, 2.5–3 meters above the ground. The edge computing unit performs region segmentation, target extraction, and local audible and visual alarm operations. When the compensated temperature exceeds 37.3°C, the local audible and visual alarm is triggered.

3. The student dormitory health monitoring and early warning system based on infrared temperature sensing according to claim 1, characterized in that, The first temperature sensor of the mobile temperature calibration device is located at the center of the heat spreader. The data measurement module includes a second temperature sensor and a wind speed sensor. The array calibration control module can set the temperature calibration benchmark, wind speed calibration benchmark and corresponding deviation threshold, and supports three calibration modes: threshold trigger, timed trigger and manual trigger.

4. The student dormitory health monitoring and early warning system based on infrared temperature sensing according to claim 1, characterized in that, The CUSUM algorithm formula for the health monitoring algorithm module is: ； in This is the current temperature reading. The baseline mean is based on individual temperature data over the past 14 days. To allow for fluctuations, As a threshold, when The module can trigger individual trend warnings; at the same time, it can calculate the average temperature index and abnormality rate of the dormitory building and floors. When the abnormality rate rises by more than 10% within 2 hours, a group warning is triggered.

5. A student dormitory health monitoring and early warning system based on infrared temperature sensing according to claim 4, characterized in that, The data linkage processing module's feedback includes: applying the calibration curves and array calibration models generated by the calibration algorithm module to accurately correct the raw body temperature data; and dynamically adjusting the CUSUM algorithm based on the calibrated temperature measurement accuracy. , , Abnormal proportion thresholds for parameters and population health analysis.

6. A method for a student dormitory health monitoring and early warning system based on the infrared temperature sensing system according to any one of claims 1-5, characterized in that, Includes the following steps: S1: Calibration trigger: The array calibration control module sets the temperature calibration benchmark, wind speed calibration benchmark, temperature deviation threshold and wind speed deviation threshold. The ambient temperature and humidity sensor and wind speed sensor collect environmental parameters in real time. When the deviation reaches the threshold, the preset timed calibration time is reached or a manual calibration command is received, the calibration process is triggered and a calibration notification is sent to the platform layer. S2: Calibration execution. The array calibration control module plans the movement path and controls the mobile temperature calibration device to move autonomously to the distributed calibration measurement points. The temperature control module sets multiple temperature control points and adjusts the output temperature of the standard temperature source. The calibration data acquisition module synchronously collects sensor measurement data, standard temperature data and environmental parameters. The edge computing unit cleans the data and uploads it to the platform layer. S3: Calibration curve generation. The platform-level calibration algorithm module performs correlation analysis on the collected data, generates single-sensor calibration curves, and fuses them to establish an array calibration model. The calibration results are stored on the cloud server. S4: Data linkage correction. The data linkage processing module applies the calibration curve and array calibration model to the raw body temperature data to complete accurate calibration, while dynamically adjusting the core parameters of the health monitoring algorithm module. S5: Monitoring and early warning. Based on calibrated effective body temperature data and corrected early warning parameters, the system performs individual body temperature trend detection and group health analysis. When abnormalities occur, it pushes graded early warning information through the application layer, and the edge computing unit triggers local audible and visual alarms when the temperature exceeds the threshold.

7. The method according to claim 6, characterized in that, In step S1, the temperature calibration benchmark is set to 25℃ by default, the wind speed calibration benchmark is set to 0.2m / s by default, the temperature deviation threshold is set to ±2℃, and the wind speed deviation threshold is set to ±0.3m / s; in step S2, the temperature fluctuation of the standard temperature source is controlled within ±0.1℃.

8. The method according to claim 6, characterized in that, In step S2, the multiple temperature control points adopt either a 3-point calibration method or a 5-point calibration method. The control points for the 3-point calibration method are 35℃, 37.5℃, and 42℃, while the control points for the 5-point calibration method are 35℃, 36.5℃, 37.5℃, 39℃, and 41℃. The temperature control points cover the range of human body temperature measurement.

9. The method according to claim 6, characterized in that, It also includes a pre-calibration step for the mobile temperature calibration device: the mobile temperature calibration device is placed in a constant temperature space, a surface temperature sensor is installed on the outside of the bionic skin, and monitoring data is collected by changing the ambient temperature and air flow rate. A first calibration equation between the environmental parameters and the data of the first temperature sensor, and a second calibration equation between the environmental parameters and the temperature of the temperature control module are established to achieve precise temperature control of the standard temperature source.

10. The method according to claim 6, characterized in that, Step S3 is followed by a calibration accuracy assessment step. The platform layer assesses the calibration accuracy of each infrared sensor. If the temperature measurement error is greater than ±0.2℃, a sensor fault alert is sent to the management personnel through the application layer. The raw body temperature data collected by the sensing layer is first processed by the environmental temperature and humidity compensation formula. After compensation, precise calibration is then performed using calibration curves and array calibration models. For compensation coefficient, The current ambient temperature. Calibrate the ambient temperature.

Images