A target detection-based body temperature monitoring system and method

Abstract

The application discloses a body temperature monitoring system and method based on target detection, which collects information such as body temperature, air temperature and forehead distance of a target object by a single-chip microcomputer, and obtains the actual forehead temperature of the target object through a correction curve by the single-chip microcomputer. The single-chip microcomputer sends the forehead temperature and air temperature and other information to a computer. A fever alarm threshold is set through air temperature and other parameters. The computer runs target detection software, detects a face position through an image acquisition device, and sends position, alarm state and other information to the single-chip microcomputer through a PID algorithm to drive a cloud platform to track the target object to be measured. Through server forwarding and linkage with terminal equipment, information sharing and control are realized. The application can quickly and accurately measure body temperature, accurately obtain an alarm temperature, and greatly improve the detection rate of fever personnel. Linkage with terminal equipment basically realizes unattended operation, and improves personnel scheduling efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things communication, in particular to the fields of target identification tracking and target temperature monitoring. BACKGROUND

[0002] Research shows that there is a large temperature difference between different parts of the human body, and it is related to the air temperature. The traditional non-contact body temperature detection method is to directly obtain body temperature data through infrared temperature measurement. The temperature measurement door and other devices can only detect the highest temperature of the detected person's body surface and cannot accurately obtain the specific body surface temperature of the target. There is no linkage with the total control center, applet, etc. Unmanned value cannot be realized.

[0003] Thanks to the development of Internet of Things technology, communication technology, artificial intelligence technology and microelectronic chip technology, the present remote transmission control system can be realized. SUMMARY

[0004] The present application aims to provide a target detection-based body temperature monitoring system and method, which introduces the influence factor of environmental temperature on human body temperature, realizes accurate detection of the body temperature of the target object, and links the detection result with the video monitoring center and terminal device to realize unmanned value.

[0005] To achieve the above functions, the present application designs a target detection-based body temperature monitoring system, which comprises an image acquisition device, a temperature module, a body temperature monitoring module, a distance measurement module, an alarm module, a pan-tilt module, a single-chip microcomputer, a display module, a computer, a video monitoring center, a server and a terminal device.

[0006] The temperature module, the body temperature monitoring module, the distance measurement module, the alarm module, the pan-tilt module, the display module and the computer are respectively connected to the single-chip microcomputer, the image acquisition device, the video monitoring center and the server are respectively connected to the computer, and the server is in communication connection with the terminal device. The pan-tilt module comprises a two-axis pan-tilt driven by a motor, the image acquisition device, the body temperature monitoring module and the distance measurement module are installed on the two-axis pan-tilt, the image acquisition device is coaxial with the body temperature monitoring module, and the distance measurement direction of the distance measurement module is kept at a preset angle with the axis direction of the image acquisition device and the body temperature monitoring module.

[0007] The alarm module comprises a red-green-yellow three-color LED lamp and a buzzer.

[0008] The present application also designs a target detection-based body temperature monitoring method based on the target detection-based body temperature monitoring system, which is aimed at the target human body and performs the following steps S1-S7 to complete the body temperature detection of the target object:

[0009] Step S1: input the image of the target object collected by the image acquisition device and the current angle of the holder motor into the computer, determine the position of the measured part of the target object in the image based on a face detection algorithm, and adjust the angle of the holder motor based on a PID network according to the current angle of the holder motor, so that the measured part of the target object is at a preset position in the image;

[0010] Step S2: input the body temperature of the measured part of the target object measured by the body temperature monitoring module, the distance between the target object and the distance measuring module measured by the distance measuring module, and the current air temperature measured by the air temperature module into the single-chip microcomputer, calculate the actual body temperature of the measured part of the target object according to the constant temperature water bath method, output the calculation result and the current air temperature measured by the air temperature module to the computer by the single-chip microcomputer, and output the calculation result to the display module for display;

[0011] Step S3: the computer obtains a fever alarm threshold corresponding to the current air temperature for judging whether the target object has a fever according to a curve fitting method based on the current air temperature;

[0012] Step S4: the computer is in communication connection with the terminal device through the server, and the computer sends a data packet A containing the actual body temperature of the measured part of the target object, the current air temperature, and a detection flag bit to the terminal device through the server at regular intervals, the terminal device sends a data packet B containing a forced alarm and a release alarm to the computer through the server at regular intervals, and the computer receives the data packet B and caches it;

[0013] Step S5: the computer judges whether the target object has a fever according to the actual body temperature of the measured part of the target object and the fever alarm threshold, and marks the judgment result and the actual body temperature of the measured part of the target object in the image collected by the image acquisition device, and saves the image;

[0014] Step S6: the computer sends a data packet containing the data packet A, the data packet B, the angle of the holder motor, and the detection flag bit to the single-chip microcomputer at the same time, and sends the image obtained in step S5 to the video monitoring center;

[0015] Step S7: the single-chip microcomputer receives the data packet sent by the computer, outputs the angle of the holder motor in the data packet to the holder motor of the holder module, adjusts the holder motor to rotate to a preset angle, and outputs the detection flag bit in the data packet to the alarm module, and the alarm module makes corresponding warning according to the detection flag bit.

[0016] As a preferred technical solution of the present application, the specific steps of step S1 are as follows:

[0017] Step S11: input the image of the target object collected by the image collection device into the computer, and obtain the center point coordinates (x, y), width w, and height h of each target frame based on a face detection algorithm, and take the target frame with the largest area as the detection frame of the target object through a non-maximum suppression method;

[0018] Step S12: take the forehead as the measured part of the target object, and the forehead center point coordinates as Subtract the forehead center point coordinates from the image center point coordinates to obtain error coordinates (x1, y1);

[0019] Step S13: input the error coordinates (x1, y1) and the current angle (degree_x, degree_y) of the gimbal motor into the PID network to adjust the angle of the gimbal motor, and the adjusted angle of the gimbal motor is (degree_x1, degree_y1).

[0020] As a preferred technical solution of the present application, the specific steps of step S2 are as follows:

[0021] Step S21: the single-chip microcomputer triggers the body temperature monitoring module, the distance measuring module, and the air temperature module to measure the body temperature of the measured part of the target object, the distance between the target object and the distance measuring module, and the current air temperature for a preset number of times within a preset period;

[0022] Step S22: calculate the actual body temperature of the measured part of the target object according to the constant temperature water bath method, wherein the compensation coefficient fitted by the constant temperature water bath method is as follows:

[0023] y b =10 -6 x b 3 -0.0003x b 2 +0.0381

[0024] In the formula, x b is the distance between the target object and the distance measuring module, in cm, and y b is the compensation temperature, in ℃;

[0025] Step S23: the single-chip microcomputer calculates the average of the actual body temperature of the measured part of the target object, the distance between the target object and the distance measuring module, and the current air temperature within the preset period, and sends the data report in the form of (actual body temperature of the measured part of the target object, current air temperature, distance enable flag) to the computer, wherein the distance enable flag is high when the distance between the target object and the distance measuring module is less than a preset value, and is low otherwise.

[0026] As a preferred technical solution of the present application, the method for obtaining the fever alarm threshold according to the curve fitting method in step S3 is as follows:

[0027] y a = -0.02717x a 2 + 1.3453x a + 15.388

[0028] wherein x a is the current air temperature, and y a is the fever alarm threshold.

[0029] As a preferred technical solution of the present application, the specific steps of step S4 are as follows:

[0030] Step S41: the computer sends a datagram A to the server at a regular time, the datagram A being in the form of (actual body temperature of the measured part of the target object, current air temperature, detection flag bit);

[0031] Step S42: the terminal device acquires the datagram A from the server at a regular time and displays it;

[0032] Step S43: the terminal device sends a datagram B to the server at a regular time, the datagram B being in the form of (forced alarm, release alarm), the states of the forced alarm and the release alarm including 0 and 1, and being valid when the state is 1;

[0033] Step S44: the computer acquires the datagram B from the server at a regular time and buffers it.

[0034] As a preferred technical solution of the present application, the specific method of step S5 is as follows:

[0035] The fever flag bit is set, and the computer buffers the datagram based on step S4, wherein the fever flag bit is 1 when the forced alarm is 1, the fever flag bit is 0 when the release alarm is 1, and the priority of the release alarm is lower than that of the forced alarm;

[0036] The image collected by the image collection device is normalized, the center point coordinate of the measured part of the target object is located in a preset interval of the normalized image, and the distance enabling flag bit is high, so that the detection flag bit is 3, otherwise the detection flag bit is 0;

[0037] The computer sorts the actual body temperatures of the measured part of the target object from high to low and buffers them, if the detection time is greater than two seconds and the highest actual body temperature is not greater than the fever alarm threshold, it is determined that the body temperature of the target object is normal, the detection flag bit is 2, the actual body temperature is marked in the image, and the image is saved;

[0038] If the highest actual body temperature is greater than the fever alarm threshold, the detection flag bit is 1, the fever flag bit is 1, the fever is marked in the image, the actual body temperature is marked, and the image is saved.

[0039] As a preferred technical scheme of the present application: in step S7, after the single-chip microcomputer receives the datagram sent by the computer, the degree (degree_x1, degree_y1) of the holder motor is output to the holder motor of the holder module in the form of a PWM signal, so as to adjust the holder motor to rotate to the preset angle.

[0040] The single-chip microcomputer outputs the detection flag bit in the datagram to the alarm module, and the alarm module makes corresponding warning according to the detection flag bit, wherein when the detection flag bit is 1, the alarm module executes the sound-light alarm and the red light flashes; when the detection flag bit is 2, the alarm module prompts the green light; when the detection flag bit is 3, the alarm module prompts the yellow light; and when the detection flag bit is 0, the red, green and yellow lights of the alarm module are all not lit.

[0041] As a preferred technical scheme of the present application: in step S5, the computer sets the cycle flag bit and the ranking identifier, and when the detection flag bit is set to 3 for the first time, the cycle flag bit is set to 1, and the ranking identifier is increased by 1; starting from the moment when the detection flag bit is set to 3 for the first time, if the detection flag bit is not 0, the cycle flag bit and the ranking identifier remain unchanged until the detection flag bit is 0.

[0042] If the cycle flag bit is 1, the image collected by the image collection device is intercepted, the actual body temperature of the measured part of the highest target object is obtained, and the actual body temperature is marked in the image and saved in the corresponding folder in the computer with the ranking identifier as the name, so as to complete the update of the actual body temperature of the measured part of the highest target object in real time.

[0043] If the detection flag bit is 0, it indicates that the current target object has completed detection, and the cycle flag bit is 0.

[0044] Advantages: Compared with the prior art, the present application has the following advantages:

[0045] (1) The target detection technology is adopted to make the detection position accurate and controllable.

[0046] (2) The body temperature is quickly and accurately measured through laser ranging, air temperature and other data, and the fever alarm threshold is lowered.

[0047] (3) The detection data is sent to the video monitoring center and the terminal device in real time to realize unmanned value keeping, and the personnel scheduling efficiency is improved.

[0048] (4) The target detection runs on the computer, which can save resources and realize low-cost operation.

[0049] (5) The ranging module has a certain downward angle, which reduces the influence of the hair of the target object on the ranging result, so that the temperature measurement and ranging result is more accurate.

[0050] (6) The single-chip microcomputer can operate independently from the computer and the network to accurately measure the temperature. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 is a target detection-based body temperature monitoring system structure diagram provided according to an embodiment of the present application;

[0052] Figure 2 is a structure diagram of an image acquisition device and a body temperature monitoring module provided according to an embodiment of the present application;

[0053] Figure 3 is an ESP32 pin diagram provided according to an embodiment of the present application;

[0054] Figure 4 is an I2C bus connection schematic diagram provided according to an embodiment of the present application

[0055] Figure 5 is a WeChat applet effect diagram provided according to an embodiment of the present application;

[0056] Figure 6 is a flowchart of a flag detection processing procedure in step S5 provided according to an embodiment of the present application;

[0057] Figure 7 is a repeated temperature measurement experiment result waveform diagram provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0058] The present application will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and cannot be used to limit the protection scope of the present application.

[0059] An embodiment of the present application provides a target detection-based body temperature monitoring system, referring to Figure 1 , comprising an image acquisition device, an air temperature module, a body temperature monitoring module, a distance measurement module, an alarm module, a pan-tilt module, a single-chip microcomputer, a display module, a computer, a video monitoring center, a server, and a terminal device.

[0060] The air temperature module, the body temperature monitoring module, the distance measurement module, the alarm module, the pan-tilt module, the display module, and the computer are respectively connected to the single-chip microcomputer, the image acquisition device, the video monitoring center, and the server are respectively connected to the computer, and the server is in communication connection with the terminal device. The pan-tilt module comprises a two-axis pan-tilt driven by a motor, the image acquisition device, the body temperature monitoring module, and the distance measurement module are all installed on the two-axis pan-tilt, the image acquisition device is coaxial with the body temperature monitoring module, the distance measurement direction of the distance measurement module maintains a preset angle with the axis direction of the image acquisition device and the body temperature monitoring module, referring to Figure 2 , the preset angle is 5.7 degrees of downward inclination.

[0061] The alarm module comprises red, green, and yellow three-color LED lights and a buzzer.

[0062] In one embodiment, the single-chip microcomputer model is ESP32, the image acquisition device is a USB high-speed camera, the air temperature module and the body temperature monitoring module adopt MLX90614-DCI, the distance measuring module adopts VL53L0X, and the display module adopts a 0.91-inch OLED display screen.

[0063] With reference to Figure 3 , the gimbal PWM signal is output by GPIO16 and GPIO17, the frequency is 50HZ, and the resolution is 12bit. The red, green and yellow LED lamps are output by GPIO12, GPIO13 and GPIO14 respectively, the buzzer and the red LED lamp are connected in parallel, and a triode is used as a switching amplification current. The I2C bus is connected to GPIO22 (SCL) and GPIO21 (SDA).

[0064] With reference to Figure 4 , the MLX90614-DCI, XL53LOX and OLED display screen are connected with the ESP32 through the I2C bus, the three modules are powered by 3.3v, and the SDA and SCL buses need two 4.7KΩ pull-up resistors.

[0065] The body temperature monitoring method based on target detection provided by the embodiment of the application, the body temperature monitoring system based on the target detection, for the target human body, executes the following steps S1 to S7, and completes the body temperature detection of the target object:

[0066] Step S1: input the image of the target object collected by the image acquisition device and the current angle of the gimbal motor into the computer, determine the position of the measured part of the target object in the image based on the face detection algorithm, and adjust the angle of the gimbal motor based on the PID network according to the current angle of the gimbal motor, so that the measured part of the target object is in the preset position in the image.

[0067] The specific steps of step S1 are as follows:

[0068] Step S11: input the image of the target object collected by the image acquisition device into the computer, and obtain the center point coordinates (x, y), width w and height h of each target frame based on the YOLOv5 face detection algorithm of the computer, and take the target frame with the largest area as the detection frame of the target object by the non-maximum suppression method;

[0069] Step S12: take the forehead as the measured part of the target object, and the forehead center point coordinates are Subtract the forehead center point coordinates from the image center point coordinates to obtain the error coordinates (x1, y1);

[0070] Step S13: input the error coordinates (x1, y1) and the current angles of the gimbal motors (degree_x, degree_y) into the PID network, adjust the angles of the gimbal motors, and the adjusted angles of the gimbal motors are (degree_x1, degree_y1).

[0071] Step S2: input the body temperature of the measured part of the target object measured by the body temperature monitoring module, the distance between the target object and the distance measuring module measured by the distance measuring module, and the current air temperature measured by the air temperature module into the single-chip microcomputer, calculate the actual body temperature of the measured part of the target object according to the constant-temperature water bath method, output the calculation result and the current air temperature measured by the air temperature module to the computer by the single-chip microcomputer, and output the calculation result to the display module for display.

[0072] The specific steps of step S2 are as follows:

[0073] Step S21: the single-chip microcomputer triggers the body temperature monitoring module, the distance measuring module, and the air temperature module to measure the body temperature of the measured part of the target object, the distance between the target object and the distance measuring module, and the current air temperature for a preset number of times within a preset period.

[0074] In one embodiment, a 10HZ trigger signal is obtained by using the hardware timer of the ESP32 single-chip microcomputer, and the body temperature of the measured part of the target object, the distance between the target object and the distance measuring module, and the current air temperature are sampled 6 times in each period to obtain the average value.

[0075] Step S22: calculate the actual body temperature (ObjectTemp) of the measured part of the target object according to the constant-temperature water bath method, wherein the compensation coefficient fitted by the constant-temperature water bath method is as follows:

[0076] y b =10 -6 x b 3 -0.0003x b 2 +0.0381

[0077] In the formula, x b is the distance between the target object and the distance measuring module, in cm, and y b is the compensation temperature, in ℃.

[0078] Step S23: the single-chip microcomputer calculates the average values of the actual body temperature of the measured part of the target object, the distance between the target object and the distance measuring module, and the current air temperature within the preset period, and sends them to the computer in the form of (actual body temperature of the measured part of the target object, current air temperature, distance enable flag) data report at a frequency of 10HZ, wherein the distance enable flag is high when the distance between the target object and the distance measuring module is less than 120cm, and is low otherwise.

[0079] Step S3: The computer obtains a fever alarm threshold corresponding to the current air temperature for judging whether the target object has a fever according to the curve fitting method, wherein the curve can be loaded by software later.

[0080] The method for obtaining the fever alarm threshold according to the curve fitting method is as follows:

[0081] y a = -0.02717x a 2 + 1.3453x a + 15.388

[0082] In the formula, x a is the current air temperature, and y a is the fever alarm threshold.

[0083] Step S4: The computer and the terminal device communicate through the MQTT (Message Queue Telemetry Transport) server forwarding. The computer sends a data packet A containing the actual body temperature of the measured part of the target object, the current air temperature, and the detection flag bit to the terminal device through the server at regular intervals. The terminal device sends a data packet B containing the forced alarm and the alarm cancellation to the computer through the server at regular intervals. The computer receives and caches the data packet B.

[0084] The specific steps of step S4 are as follows:

[0085] Step S41: The computer sends a data packet A to the server topic A at regular intervals. The data packet A is in the form of (actual body temperature of the measured part of the target object, current air temperature, detection flag bit);

[0086] Step S42: The terminal device obtains the data packet A from the server subscription topic A at regular intervals and displays it.

[0087] Step S43: The terminal device sends a data packet B to the server topic B at regular intervals. The data packet B is in the form of (forced alarm, alarm cancellation). The states of the forced alarm and the alarm cancellation include 0 and 1, and are valid when the state is 1. The forced alarm and the alarm cancellation are in the form of buttons on the terminal device for the staff to operate.

[0088] Step S44: The computer obtains the data packet B from the server at regular intervals and caches it.

[0089] In one embodiment, with reference to Figure 5 , the terminal device is a WeChat mini-program that can obtain the device state alarm state, air temperature and body temperature information, and manually control the switch of the alarm. The specific implementation principle is to use the publish-subscribe mode to realize the synchronization of the mini-program data and the detection center data.

[0090] Subscription is to get information of a given topic / channel. Publication is to send information to a given topic / channel. Only devices that have subscribed to the topic can receive messages sent to the topic.

[0091] Mini-program side process: first set the mqtt server address of the client side, then subscribe to the corresponding topic through client. subscribe, get the temperature and humidity data of the mqtt server through the data. client. on() function, and display the data on the mini-program according to the specific delimiter; upload the alarm switch information to the server through the data. client. publish() function, and wait for the python side to read it to realize information synchronization.

[0092] Python side MQTT process: similarly, subscribe to the corresponding topic, read the alarm switch information through msg. topic function, and pass it to the lower computer operation. Since the python side also needs to complete the operation of detecting objects, real-time transmission will affect the efficiency of the program, and in severe cases, it will cause the program to crash, so when getting data, the client. loop_start() function is used to operate in multiple threads.

[0093] Step S5: The computer judges whether the target object has a fever according to the actual temperature of the measured part of the target object and the fever alarm threshold. The computer labels the judgment result and the actual temperature of the measured part of the target object in the image collected by the image acquisition device and saves the image.

[0094] Reference Figure 6 The specific method of step S5 is as follows:

[0095] Set the fever flag bit. When the forced alarm is 1, the testing flag bit is set to 1 based on the data report cached in step S4. When the alarm is released, the fever flag bit is set to 0, and the priority of the released alarm is lower than that of the forced alarm.

[0096] Normalize the image collected by the image acquisition device. The center point coordinate of the measured part of the target object is located in the 0.4-0.6 interval of the normalized image, and the distance enabling flag bit is high, then the testing flag bit is set to 3, indicating that the image acquisition device has been aligned with the measured part of the target object, otherwise the testing flag bit is set to 0.

[0097] The computer sorts and caches the actual temperatures of the measured part of the target object from high to low. If the detection time is greater than two seconds and the highest actual temperature is not greater than the fever alarm threshold, it is determined that the temperature of the target object is normal, the testing flag bit is set to 2, the actual temperature is labeled in the image and the image is saved.

[0098] If the highest actual body temperature is greater than the fever alarm threshold, the detection flag position is 1, the fever flag position is 1, the fever is marked in the image, the actual body temperature is marked, and the image is saved.

[0099] In step S5, the computer sets a cycle flag (first) and a rank identifier (rank). When the detection flag is set to 3 for the first time, the cycle flag is 1, and the rank identifier is incremented by 1. Starting from the moment when the detection flag is set to 3 for the first time, if the detection flag is not 0, the cycle flag and the rank identifier remain unchanged until the detection flag is 0.

[0100] If the cycle flag is 1, the image acquisition device is set to 10HZ frequency to intercept the image, the highest actual body temperature of the target object is obtained, and the image is marked with the rank identifier to save the corresponding folder in the computer, completing the real-time update of the highest actual body temperature of the target object.

[0101] If the detection flag is 0, it means that the current target object has completed the detection, and the cycle flag is 0, waiting for the next person to restart the cycle.

[0102] Step S6: The computer sends the data packet containing data packet A, data packet B, the angle of the pan-tilt motor, and the detection flag to the single-chip microcomputer, and sends the image obtained in step S5 to the video monitoring center.

[0103] In one embodiment, the image acquisition device is a multi-camera, and the computer sends data packets to the single-chip microcomputer in the format of (“A”+degree_x1+“B”+degree_y1+“C”+testing) at a frequency of up to 100HZ. The computer pulls the stream through OpenCV, performs frame detection, and splices the images captured by the multi-camera. The spliced image is pushed into the pipeline of the rtmp streaming media server built by nginx, realizing rtmp streaming.

[0104] Step S7: The single-chip microcomputer receives the data packet sent by the computer, outputs the angle of the pan-tilt motor in the data packet to the pan-tilt motor of the pan-tilt module, adjusts the pan-tilt motor to rotate to the preset angle, and outputs the detection flag in the data packet to the alarm module, which makes corresponding warning according to the detection flag.

[0105] In step S7, after the single-chip microcomputer receives the data packet sent by the computer, the angle of the pan-tilt motor (degree_x1, degree_y1) is output to the pan-tilt motor of the pan-tilt module as a PWM signal, adjusting the pan-tilt motor to rotate to the preset angle.

[0106] The single-chip microcomputer outputs a detection flag in the data packet to an alarm module, and the alarm module gives corresponding warning according to the detection flag, wherein when the detection flag is 1, the alarm module executes sound-light alarm and red light flashes; when the detection flag is 2, the alarm module prompts green light; when the detection flag is 3, the alarm module prompts yellow light; and when the detection flag is 0, it indicates that there is no detection object, and the red, green and yellow lights of the alarm module are all not bright.

[0107] Referring to Figure 7 Experiments show that the present application can quickly and accurately measure the body temperature, accurately obtain the alarm temperature, and greatly improve the detection rate of fever personnel. The non-contact temperature measurement detection distance can reach 100 CM, the accuracy can reach 0.2 DEG C in the body temperature range, the detection time only needs two seconds. It is good in interaction with the video monitoring center and the small program. It can be applied to various target tracking temperature measurement occasions.

[0108] The embodiments of the present application are described in detail above in combination with the drawings, but the present application is not limited to the above-mentioned embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the purpose of the present application.

Claims

1. A target detection based body temperature monitoring system, characterized in that, The system comprises an image acquisition device, an air temperature module, a body temperature monitoring module, a distance measuring module, an alarm module, a cloud module, a single-chip microcomputer, a display module, a computer, a video monitoring center, a server, and a terminal device. The air temperature module, the body temperature monitoring module, the distance measuring module, the alarm module, the cloud module, the display module, and the computer are connected to the single-chip microcomputer, the image acquisition device, the video monitoring center, and the server are connected to the computer, and the server is connected to the terminal device. The cloud module comprises a two-axis cloud module driven by a motor, the image acquisition device, the body temperature monitoring module, and the distance measuring module are installed on the two-axis cloud module, the image acquisition device is coaxial with the body temperature monitoring module, and the distance measuring direction of the distance measuring module is at a preset angle with the axis direction of the image acquisition device and the body temperature monitoring module. The alarm module comprises a red-green-yellow three-color LED lamp and a buzzer. Based on the target detection-based body temperature monitoring system, the following steps S1-S7 are performed for a target human body to complete body temperature detection of the target object: Step S1: input the image of the target object acquired by the image acquisition device and the current angle of the cloud motor into the computer, determine the position of the measured part of the target object in the image based on a face detection algorithm, and adjust the angle of the cloud motor based on a PID network according to the current angle of the cloud motor, so that the measured part of the target object is at a preset position in the image. Step S2: input the body temperature of the measured part of the target object measured by the body temperature monitoring module, the distance between the target object and the distance measuring module measured by the distance measuring module, and the current air temperature measured by the air temperature module into the single-chip microcomputer, calculate the actual body temperature of the measured part of the target object according to the constant-temperature water bath method, output the calculation result and the current air temperature measured by the air temperature module to the computer by the single-chip microcomputer, and output the calculation result to the display module for display. The specific steps of step S2 are as follows: Step S21: the single-chip microcomputer triggers the body temperature monitoring module, the distance measuring module, and the air temperature module to measure the body temperature of the measured part of the target object, the distance between the target object and the distance measuring module, and the current air temperature for a preset number of times within a preset period; In the formula, is the distance between the target object and the ranging module, in cm, is the compensation temperature, in °C; Step S22: calculate the actual body temperature of the measured part of the target object according to the constant-temperature water bath method, wherein the compensation coefficient fitted by the constant-temperature water bath method is as follows: Step S23: the single-chip microcomputer calculates the average of the actual body temperature of the measured part of the target object, the distance between the target object and the distance measuring module, and the current air temperature within the preset period, and sends the data report in the form of (actual body temperature of the measured part of the target object, current air temperature, distance enable flag) to the computer, wherein the distance enable flag is high when the distance between the target object and the distance measuring module is less than a preset value, and low otherwise; Step S3: the computer obtains a fever alarm threshold corresponding to the current air temperature according to the curve fitting method based on the current air temperature, which is used to determine whether the target object has a fever; The method for obtaining the fever alarm threshold according to the curve fitting method in step S3 is as follows: wherein is the current air temperature, is a fever alert threshold; Step S4: The computer is connected with the terminal device through the server, and the computer sends the data packet A containing the actual body temperature of the measured part of the target object, the current air temperature and the detection flag bit to the terminal device through the server at regular time intervals, the terminal device sends the data packet B containing the forced alarm and the alarm cancellation to the computer through the server at regular time intervals, and the computer receives the data packet B and buffers; Step S5: The computer judges whether the target object has a fever according to the actual body temperature of the measured part of the target object and the fever alarm threshold, marks and saves the image in the image acquisition device according to the judgment result and the actual body temperature of the measured part of the target object. Step S6: The computer sends the data packet A, the data packet B, the angle of the pan-tilt motor and the detection flag bit to the single-chip microcomputer at the same time, and sends the image obtained in step S5 to the video monitoring center; Step S7: The single-chip microcomputer receives the data packet sent by the computer, outputs the angle of the pan-tilt motor in the data packet to the pan-tilt motor of the pan-tilt module, adjusts the pan-tilt motor to rotate to the preset angle, and outputs the detection flag bit in the data packet to the alarm module, and the alarm module makes corresponding warning according to the detection flag bit.

2. The target detection based body temperature monitoring system of claim 1, wherein, The specific steps of step S1 are as follows: Step S11: Input the image of the target object acquired by the image acquisition device into the computer. The computer obtains the center point coordinates of each target box based on the face detection algorithm. x , y ),width w ,high h By using nonmaximum suppression, the target box with the largest area is used as the detection box of the target object; Step S12: Taking the forehead as the target object, the forehead center point coordinate is The error coordinate is obtained by subtracting the image center point coordinate from the forehead center point coordinate ; Step S13: coordinate of the error current angle of the gimbal motor input the PID network, adjust the angle of the gimbal motor, and the adjusted angle of the gimbal motor is .

3. The target detection based body temperature monitoring system of claim 1, wherein, The specific steps of step S4 are as follows: Step S41: The computer sends the data packet A to the server at regular time intervals, and the data packet A is in the form of (actual body temperature of the measured part of the target object, current air temperature, detection flag bit); Step S42: The terminal device acquires the data packet A from the server at regular time intervals and displays it; Step S43: The terminal device sends the data packet B to the server at regular time intervals, and the data packet B is in the form of (forced alarm, alarm cancellation), the state of the forced alarm and the alarm cancellation includes 0 and 1, and the state is valid when it is 1; Step S44: The computer acquires the data packet B from the server at regular time intervals and buffers it.

4. The target detection based body temperature monitoring system of claim 3, wherein, The specific method of step S5 is as follows: Set the fever flag bit, and the computer buffers the data packet based on step S4, wherein the forced alarm is 1, the detection flag bit is 1, the alarm cancellation is 1, the fever flag bit is 0, and the priority of the alarm cancellation is lower than that of the forced alarm; The image collected by the image acquisition device is normalized, the center point coordinate of the measured part of the target object is located in the normalized image preset interval, and the distance enabling flag bit is high, then the detection flag bit is 3, otherwise the detection flag bit is 0; The computer sorts the actual body temperature of the measured part of the target object from high to low and buffers it, if the detection time is greater than two seconds and the highest actual body temperature is not greater than the fever alarm threshold, then it is determined that the target object has normal body temperature, the detection flag bit is 2, the actual body temperature is marked in the image and the image is saved; If the highest actual body temperature is greater than the fever alarm threshold, then the detection flag bit is 1, the fever flag bit is 1, the fever is marked in the image, the actual body temperature is marked and the image is saved.

5. The target detection based body temperature monitoring system of claim 4, wherein, After the single-chip microcomputer receives the datagram sent by the computer in step S7, the angle of the holder motor is adjusted to the preset angle The holder motor outputs a PWM signal to the holder module, and the holder motor is adjusted to rotate to the preset angle. The single-chip microcomputer outputs the detection flag bit in the data packet to the alarm module, and the alarm module makes corresponding warning according to the detection flag bit, wherein when the detection flag bit is 1, the alarm module executes the sound and light alarm and the red light flashes; When the detection flag bit is 2, the alarm module prompts the green light; When the detection flag bit is 3, the alarm module prompts the yellow light; When the detection flag is 0, the red, green and yellow lights of the alarm module are not lit.

6. The target detection based body temperature monitoring system of claim 5, wherein, In step S5, the computer sets the cycle flag and the ranking identifier, and when the detection flag is set to 3 for the first time, the cycle flag is set to 1, and the ranking identifier is increased by 1; Taking the moment when the detection flag is set to 3 for the first time as the starting point, if the detection flag is not 0, the cycle flag and the ranking identifier remain unchanged until the detection flag is 0; If the cycle flag is 1, the image collected by the image collection device is intercepted, the actual body temperature of the measured part of the current highest target object is obtained, and the image is marked with the ranking identifier to cover the corresponding folder in the computer, thereby completing the update of the actual body temperature of the measured part of the real-time highest target object. If the detection flag is 0, it means that the current target object has completed the detection, and the cycle flag is 0.

Images