Intelligent monitoring response method, system and equipment based on Internet of Things

Abstract

The present invention discloses an intelligent monitoring response method, system and device based on the Internet of Things. The method belongs to the field of data processing technology, and includes the following steps: using a face recognition temperature measurement camera to obtain the body temperature data and facial image of the person being detected in real time; preprocessing the real-time body temperature data and facial image; using a multi-source data fusion algorithm to comprehensively analyze the preprocessed data; by analyzing the slight changes in the body temperature data, identifying and marking abnormal temperature areas; using a Bayesian network algorithm to conduct risk assessment on abnormal temperature areas; when a high-risk abnormal temperature area is detected, an automatic alarm is triggered. The present invention uses a multi-source data fusion algorithm to comprehensively analyze the body temperature data and facial image data, and uses a face recognition temperature measurement camera to obtain the body temperature data and facial image in real time to achieve accurate detection of regional temperature, solving the problem of low accuracy of regional temperature detection in the prior art.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent monitoring response method, system and device based on the Internet of Things. Background Art

[0002] With the increase in global population mobility and the frequent occurrence of public health emergencies, the demand for real-time health monitoring and emergency response systems is becoming increasingly urgent. The Internet of Things is a network that connects physical devices and objects to the Internet through sensors, software and other technologies to enable data exchange and communication between devices.

[0003] The existing intelligent monitoring system uses a handheld temperature gun to detect the body temperature of each person being tested, record the temperature data, and conduct further manual screening on people with abnormal temperatures. The recorded data is then managed in the form of paper or electronic spreadsheets.

[0004] For example, the invention patent with announcement number CN116187831B announces an intelligent management and control system and method for a tunnel boring machine based on big data, including: a tunnel boring data acquisition module, an environmental similarity index analysis module, a target inspection coal lane extraction module, an early warning priority analysis module and an early warning response module; the tunnel boring data acquisition module is used to acquire sample data and historical monitoring data; the environmental similarity index analysis module is used to analyze the first sample data and the first monitoring data of the coal lane and calculate the environmental similarity index of the corresponding coal lane; the target inspection coal lane extraction module is used to extract the coal lane corresponding to the maximum value of the environmental similarity index as the target inspection coal lane; the early warning priority analysis module is used to analyze the early warning priority of the second sample data and the second monitoring data; the early warning response module is used to respond to the early warning based on the early warning priority.

[0005] For example, the invention patent with announcement number: CN117094563B announces an intelligent liquid waste leakage monitoring system and method based on big data, including: a monitoring system construction module, a monitoring scene classification module, an optimal parameter configuration analysis module, a new scene matching module and a parameter warning module; the monitoring system construction module is used to construct a liquid waste leakage monitoring system with leakage monitoring, equipment configuration, algorithm management and historical query as the functional main body; the monitoring scene classification module is used to classify the monitoring scenes; the optimal parameter configuration analysis module is used to analyze the algorithm parameter configuration corresponding to the monitoring alarm data recorded in each monitoring scene, and output the optimal parameter configuration under each monitoring scene; the new scene matching module is used to traverse the historical monitoring scenes with the highest matching similarity; the parameter warning module is used to make a warning response to the algorithm parameter configuration corresponding to the new monitoring scene based on the training deviation value.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the prior art, staff members are required to use handheld temperature guns to detect the body temperature of the detected persons one by one, record the temperature data, and conduct further manual screening on persons with abnormal body temperature. The recorded data are then managed in the form of paper or electronic spreadsheets, resulting in the problem of low accuracy of regional temperature detection. Summary of the invention

[0008] The embodiments of the present application solve the problem of low accuracy of regional temperature detection in the prior art by providing an intelligent monitoring and response method, system and device based on the Internet of Things, and achieve accurate detection of regional temperature.

[0009] The embodiment of the present application provides an intelligent monitoring and response method based on the Internet of Things, comprising the following steps: using a face recognition temperature measurement camera to obtain the body temperature data and facial image of the detected person in real time; performing data preprocessing on the real-time body temperature data and facial image; using a multi-source data fusion algorithm to perform a comprehensive analysis on the preprocessed data; by analyzing slight changes in body temperature data, identifying and marking abnormal temperature areas; using a Bayesian network algorithm to perform risk assessment on abnormal temperature areas; and triggering an automatic alarm when a high-risk abnormal temperature area is detected.

[0010] Furthermore, the specific steps of the data preprocessing are: gray-scaling the collected facial image to convert the color image into a gray-scale image; smoothing the facial image using a Gaussian filtering algorithm to remove high-frequency noise; smoothing the body temperature data using a moving average filtering method to remove noise in the body temperature data; performing image enhancement processing on the facial image using a histogram equalization technique; calculating the grayscale histogram of the image to obtain the number of pixels at each grayscale level; calculating the cumulative distribution function to map the grayscale value of the original image to a new grayscale value; reconstructing the facial image according to the new grayscale value to enhance the contrast and details of the image; and extracting features from the reconstructed facial image using a convolutional neural network to extract the positions of facial key points and the corresponding temperature values ​​to form a temperature distribution map.

[0011] Furthermore, the specific steps of using the multi-source data fusion algorithm to conduct a comprehensive analysis of the pre-processed data are: defining the data sources of body temperature data and facial image data; standardizing the data from different data sources; assigning weights to each data source and calculating the comprehensive score of each data source; and weighted averaging the comprehensive scores of all data sources to obtain the final fusion score.

[0012] Furthermore, the calculation formula of the final fusion score is:

[0013] S represents the final fusion score, ω bThe weight of the b-th data source, b represents the weight number, b=1,2,...z, z represents the total number, X b represents the data value of the bth data source collected in real time, μ b represents the average value of the historical data of the bth data source, σ b Indicates the standard deviation of the historical data of the bth data source.

[0014] Furthermore, the steps of analyzing slight changes in body temperature data and identifying and marking abnormal temperature areas are as follows: selecting a time window of fixed size, performing window processing on the body temperature data, and recording the body temperature value at each moment in the window; calculating the temperature change rate in the current time window; setting a threshold for the temperature change rate, and if the temperature change rate exceeds the threshold, marking the temperature in the current time window as an abnormal temperature area; extracting key features from the marked abnormal temperature area and marking them.

[0015] Furthermore, the step of using the Bayesian network algorithm to conduct risk assessment on abnormal temperature areas includes: defining the prior probability of regional temperature risk corresponding to the abnormal temperature area based on historical health data and statistical information; calculating the conditional probability based on the abnormal temperature area data; calculating the total probability of the current observation data; calculating the posterior probability of the regional temperature risk based on the conditional probability and the total probability of the current observation data using the Bayesian formula; setting multiple risk level thresholds and dividing the risk levels; and obtaining the risk assessment level based on the numerical value of the posterior probability.

[0016] Furthermore, the calculation formula of the posterior probability is:

[0017]

[0018] In the formula, P represents the posterior probability, A represents a specific temperature state, and A i represents all possible events with all features, b i represents the feature of the current observation data, i represents the feature number of the current observation data, i = 1, 2, ..., n, n represents the total number of features, P(A) represents the prior probability, P(b i |A) represents the conditional probability, P(A i ) represents the prior probability of all possible events, P(A i ) represents the conditional probability of all possible events, Represents the total probability.

[0019] Furthermore, the step of triggering automatic alarm includes: automatically alarming according to the risk assessment level and displaying the corresponding risk level; generating alarm information according to the alarm level; pushing the generated alarm information to the user in real time; and recording each alarm information.

[0020] The embodiment of the present application provides an intelligent monitoring response system based on the Internet of Things, wherein the data acquisition module is used to use a face recognition temperature measurement camera to obtain the body temperature data and facial image of the detected person in real time; the data preprocessing module is used to perform data preprocessing on the real-time body temperature data and facial image; the comprehensive analysis module is used to use a multi-source data fusion algorithm to perform a comprehensive analysis on the preprocessed data; the temperature marking module is used to identify and mark abnormal temperature areas by analyzing slight changes in body temperature data; the risk assessment module is used to perform risk assessment on abnormal temperature areas using a Bayesian network algorithm. The automatic alarm module is used to trigger an automatic alarm when a high-risk abnormal temperature area is detected.

[0021] An embodiment of the present application provides an electronic device, comprising: a processor, and a memory for storing instructions executable by the processor; when the processor is configured to execute the instructions, the electronic device implements an intelligent monitoring response method based on the Internet of Things.

[0022] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0023] 1. By adopting a multi-source data fusion algorithm, the body temperature data and facial image data are comprehensively analyzed to integrate information from different data sources, improve the accuracy and reliability of data analysis, reduce false alarms and missed alarms, and thus achieve accurate detection of regional temperature, effectively solving the problem of low accuracy of regional temperature detection in the existing technology.

[0024] 2. By using the face recognition temperature measurement camera to obtain body temperature data and facial images in real time, and using the Bayesian network algorithm to conduct risk assessment on abnormal temperature areas, rapid data processing and risk assessment can be achieved, thereby improving the efficiency of regional temperature detection.

[0025] 3. By preprocessing real-time body temperature data and facial images, data quality and processing efficiency are improved, data accuracy and consistency are ensured, and intelligent data processing and analysis are realized, which can provide timely warnings for abnormal temperature areas and improve the system's automation and intelligence level. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flow chart of an intelligent monitoring and response method based on the Internet of Things provided in an embodiment of the present application;

[0027] Figure 2 A posterior probability data graph provided for an embodiment of the present application;

[0028] Figure 3 A schematic diagram of the structure of an intelligent monitoring and response system based on the Internet of Things provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] The embodiments of the present application solve the problem of low accuracy of regional temperature detection in the prior art by providing an intelligent monitoring and response method, system and equipment based on the Internet of Things. By utilizing a face recognition temperature measurement camera to obtain body temperature data and facial images in real time, using a Bayesian network algorithm to conduct risk assessment on abnormal temperature areas, and using a multi-source data fusion algorithm to conduct comprehensive analysis of body temperature data and facial image data, accurate detection of regional temperature is achieved.

[0030] The technical solution in the embodiment of the present application is to solve the problem of low temperature detection accuracy in the above-mentioned area, and the overall idea is as follows:

[0031] By using a face recognition temperature measurement camera, the body temperature data and facial images of the detected person are obtained in real time; the real-time body temperature data and facial images are preprocessed; a multi-source data fusion algorithm is used to conduct a comprehensive analysis of the preprocessed data; by analyzing small changes in body temperature data, abnormal temperature areas are identified and marked; the Bayesian network algorithm is used to conduct risk assessment of abnormal temperature areas; when a high-risk abnormal temperature area is detected, an automatic alarm is triggered, achieving accurate detection of regional temperature.

[0032] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0033] like Figure 1 As shown, it is a flow chart of the intelligent monitoring and response method based on the Internet of Things provided in an embodiment of the present application, and the method includes the following steps: using a face recognition temperature measurement camera to obtain the body temperature data and facial image of the detected person in real time; performing data preprocessing on the real-time body temperature data and facial image; using a multi-source data fusion algorithm to perform a comprehensive analysis on the preprocessed data; by analyzing the slight changes in the body temperature data, identifying and marking abnormal temperature areas; using a Bayesian network algorithm to perform risk assessment on abnormal temperature areas; when a high-risk abnormal temperature area is detected, triggering an automatic alarm.

[0034] In this embodiment, body temperature data and facial images are acquired and processed in real time to ensure the real-time and high quality of the data, thereby providing more complete and accurate temperature data. Through data preprocessing and comprehensive analysis technology, abnormal temperature areas can be accurately identified and marked, and abnormal temperature areas can be warned in time. The application of Bayesian network algorithm realizes scientific and accurate regional temperature risk assessment, helping decision makers to better understand and respond to potential regional temperature risks. The automatic alarm function further improves the speed and efficiency of emergency response, ensuring that when high-risk abnormal temperature areas are discovered, necessary measures can be taken quickly to ensure the health and safety of public places. Through these technical means, this embodiment provides an intelligent, real-time, and precise health monitoring solution, which effectively improves the health management level and safety of public places.

[0035] Furthermore, the specific steps of data preprocessing are as follows: grayscale the collected facial image to convert the color image into a grayscale image; use the Gaussian filter algorithm to smooth the facial image and remove high-frequency noise; use the moving average filter method to smooth the body temperature data and remove the noise in the body temperature data; use the histogram equalization technology to enhance the facial image; calculate the grayscale histogram of the image to obtain the number of pixels at each grayscale level; calculate the cumulative distribution function to map the grayscale value of the original image to a new grayscale value; reconstruct the facial image according to the new grayscale value to enhance the contrast and details of the image; use the convolutional neural network to extract features from the reconstructed facial image, extract the positions of facial key points and the corresponding temperature values, and form a temperature distribution map.

[0036] In this embodiment, the facial image is reconstructed according to the new grayscale value to enhance the contrast and details of the image, such as the facial edge and light intensity. The detailed steps of data preprocessing are: grayscale the collected facial image, convert the color image into a grayscale image, simplify the image data structure, reduce the computational complexity, and facilitate subsequent image processing, and use the Gaussian filter algorithm to smooth the facial image and remove high-frequency noise. The Gaussian filter algorithm is specifically implemented by applying the Gaussian kernel function to perform convolution operations, and the formula is as follows:

[0037]

[0038] In the formula, G(x,y) represents the Gaussian kernel function, which is used to remove high-frequency noise in the facial image and enhance the smoothness and clarity of the facial image. x, y represent the horizontal and vertical coordinates of the facial image, σ represents the standard deviation parameter, which is used to control the smoothing effect, π represents the circumference of the circle, and e represents the natural constant. The specific steps of the moving average filter are: windowing the body temperature data, selecting a window of a fixed size (such as 5 data points), and calculating the average value of the data points in the window as the smoothed body temperature value at the current moment. The purpose is to remove noise from the body temperature data and ensure the smoothness and continuity of the data. The specific method of histogram equalization is: calculate the grayscale histogram, count the number of pixels at each grayscale level, obtain the grayscale histogram, and calculate the cumulative distribution function formula of the grayscale histogram as follows: CDF(p) is the cumulative distribution function of the grayscale value, p is the total number of pixels, obtained by calculating the number of pixels, and h(j) is the number of pixels of grayscale level j. Then, the original grayscale value is mapped to a new grayscale value according to the cumulative distribution function, and the facial image is reconstructed according to the new grayscale value to enhance the contrast and details of the facial image. The specific method of using a convolutional neural network to extract features of the reconstructed facial image, extract the positions of facial key points and the corresponding temperature values, and form a temperature distribution map is as follows: the preprocessed facial image is input into a pre-trained convolutional neural network (such as VGG16), and the facial image is extracted through multiple convolutional layers. The positions of facial key points and the corresponding temperature values ​​are extracted, and a feature map is generated, which contains facial key points and temperature value information to form a temperature distribution map. The input layer of the Gaussian filter algorithm: a three-dimensional tensor (height, width, channels), where height and width are the height and width of the image, channels is the number of channels (such as RGB images have 3 channels). For an RGB image of size 224x224, the input data format is (224,224,3). Convolution layer: convolution kernel (filter): size is 7x7, number is K, stride: the step length of the convolution kernel sliding on the input image is 2, pooling layer: the pooling window is 2x2, and the output layer outputs the feature map.

[0039] Furthermore, the specific steps of using the multi-source data fusion algorithm to conduct a comprehensive analysis of the pre-processed data are as follows: define the data sources of body temperature data and facial image data; standardize the data from different data sources; assign weights to each data source and calculate the comprehensive score of each data source; and weighted average the comprehensive scores of all data sources to obtain the final fusion score.

[0040] In this embodiment, the step of assigning weights to each data source is as follows: each weight is obtained from a database, and the database is used to store regional temperature information and weights. The weights in the database can be obtained by multivariate regression analysis to obtain a weight mapping table, and the corresponding weights are obtained according to the weight mapping table and the regional temperature data. The types of data sources include body temperature data and facial image data. These data sources come from face recognition temperature measurement cameras, which can provide real-time regional temperature information. Clarifying the data source helps to standardize the subsequent data processing steps, ensure that the data of each data source can be fully utilized, and improve the comprehensiveness of the analysis. Data source definition and standardization ensure that the data of different data sources are comparable, eliminate dimensional differences, make the fusion analysis more scientific and reasonable, ensure that the key data sources have a greater impact on the final results, improve the accuracy of the analysis, and the weighted average comprehensively reflects the regional temperature information by fusing the comprehensive scores of each data source, providing a reliable basis for health monitoring and risk assessment. The specific use steps of the multi-source data fusion algorithm are: define different data sources, including body temperature data and facial image data, and standardize these data to ensure data consistency. For body temperature data, it can be normalized to a range of 0 to 1, and for facial image data, the position of the image feature points can be normalized to a range of 0 to 1. Next, a weight is assigned to each data source, and the comprehensive score of each data source is calculated. The comprehensive score is calculated by multiplying the standardized data of each data source by the corresponding weight, where the sum of all weights is 1. Then, the comprehensive scores of all data sources are weighted averaged to obtain the final fusion score. The input of the multi-source data fusion algorithm includes body temperature data, facial image data and weight information, and the output is the comprehensive score of each data source and the final fusion score. In this way, the comprehensive analysis module can effectively fuse information from different data sources, improve the accuracy and robustness of data analysis, and provide a reliable data basis for the subsequent risk assessment module. For example, suppose there are two data sources: body temperature data and facial image data, the normalized value of body temperature data is 0.7, the normalized value of facial image data is 0.5, and the weights are 0.6 and 0.4 respectively. After being processed by the multi-source data fusion algorithm, the final fusion score is 0.62. The multi-source data fusion algorithm achieves the effective fusion of body temperature data and facial image data by standardizing different data sources and calculating the comprehensive score, improving the accuracy and robustness of data analysis, and providing strong technical support for the real-time monitoring and response of the system. The input of the multi-source fusion algorithm is: body temperature data, facial image data, weight information, and the output is: comprehensive score, fusion score.

[0041] Furthermore, the calculation formula for the final fusion score is:

[0042]

[0043] S represents the final fusion score, ωb The weight of the b-th data source, b represents the weight number, b=1,2,...z, z represents the total number, X b represents the data value of the bth data source collected in real time, μ b represents the average value of the historical data of the bth data source, σ b Indicates the standard deviation of the historical data of the bth data source.

[0044] In this embodiment, X b It can be obtained through expert experimental data. Through reasonable weight distribution, it ensures that the key data source has a greater impact on the final result, and improves the accuracy and reliability of the comprehensive analysis. Standardization eliminates the dimensional differences between different data sources, making the fusion analysis more reasonable and accurate. The weight distribution is adjusted according to the importance of the data source, ensuring that the key data source has a greater impact on the final result, thereby improving the accuracy of the analysis. The weighted summation comprehensively reflects the regional temperature information by combining the weighted scores of each data source, providing a reliable basis for health monitoring and risk assessment.

[0045] Furthermore, the steps for analyzing small changes in body temperature data and identifying and marking abnormal temperature areas are as follows: select a time window of fixed size, perform window processing on the body temperature data, and record the body temperature value at each moment in the window; calculate the temperature change rate in the current time window; set a threshold for the temperature change rate, and if the temperature change rate exceeds the threshold, mark the temperature in the current time window as an abnormal temperature area; extract key features from the marked abnormal temperature area and mark them.

[0046] In this embodiment, slight changes in body temperature data are analyzed, and slight changes refer to temperature changes with a temperature difference of 0.1 degrees Celsius. If the temperature change rate does not exceed the threshold, it is a normal temperature. By analyzing slight changes in body temperature data, abnormal temperature areas are identified and marked, specifically including the following steps: First, a time window of fixed size is selected, the body temperature data is windowed, and the body temperature value at each moment in the window is recorded. This step ensures the continuity and time correlation of the data. Next, the temperature change rate in the current time window is calculated, and then the threshold of the temperature change rate is set. When the temperature change rate exceeds the preset threshold, the system marks the temperature in the current time window as an abnormal temperature area. Finally, key features are extracted from the marked abnormal temperature area and marked. The key features include the size, position, temperature difference, etc. of the abnormal temperature area. Through these steps, areas with abnormal body temperature can be accurately identified and potential regional temperature risks can be discovered in a timely manner.

[0047] Furthermore, the steps of using the Bayesian network algorithm to conduct risk assessment on abnormal temperature areas include: defining the prior probability of regional temperature risk corresponding to the abnormal temperature area based on historical health data and statistical information; calculating the conditional probability based on the abnormal temperature area data; calculating the total probability of the current observation data; calculating the posterior probability of the regional temperature risk based on the conditional probability and the total probability of the current observation data using the Bayesian formula; setting a risk level threshold and dividing the risk level; and obtaining the risk assessment level based on the value of the posterior probability.

[0048] In this embodiment, the total probability of the current observation data refers to the total probability of the temperature zone data. Historical health data and statistical information are obtained according to the National Statistics Network. The prior probability is calculated by collecting and analyzing a large amount of historical health data, counting the frequency of occurrence of different temperature states (such as no risk, low risk, medium risk, and high risk), and calculating the prior probability of each temperature state. The conditional probability is calculated by counting the observation data under different temperature states, and calculating the frequency of occurrence of each eigenvalue under a specific temperature state. For example, under the temperature state, the frequency of a certain eigenvalue of the observation data is 0.6, then the conditional probability is 0.6. The total probability of the observation data can be obtained by using the Monte Carlo simulation method, through a large number of random sampling and simulation calculations of the probability distribution of the observation data. For example: generate a large number of random samples, and obtain an estimated value of the total probability by calculating the frequency of occurrence of the observation data in the sample. According to the different severity of regional temperature risks, the regional temperature risks are divided into multiple levels, and the corresponding posterior probability threshold is set for each level. When the posterior probability value is between 0 and 0.3, it is a low risk, when the posterior probability is greater than 0.3 and less than or equal to 0.8, it is a medium risk, and when the posterior probability is greater than 0.8, it is a high risk. Specific use of the Bayesian algorithm: Based on historical health data and statistical information, define the prior probability of regional temperature risk corresponding to the abnormal temperature area, calculate the conditional probability of observing these data under a specific temperature state based on the observed abnormal temperature data, calculate the total probability of the observed data, and use the Bayesian formula to calculate the posterior probability of regional temperature risk. The input layer of the Bayesian algorithm includes the following data: real-time collected body temperature data, such as: 36.5℃, 37.2℃, etc., feature point information extracted from facial image data, prior probability, conditional probability, and output layer: posterior probability.

[0049] Furthermore, the calculation formula of the posterior probability is:

[0050]

[0051] In the formula, P represents the posterior probability, A represents a specific temperature state, and A i represents all possible events with all features, b irepresents the feature of the current observation data, i represents the feature number of the current observation data, i=1,2,...,n, n represents the total number of features, P(A) represents the prior probability, P(b i |A) represents the conditional probability, P(A i ) represents the prior probability of all possible events, P(A i ) represents the conditional probability of all possible events, Represents the total probability.

[0052] In this embodiment, the temperature state can be obtained through historical data, and the posterior probability can also be obtained through frequency statistics of large-scale actual observation data, and the frequency of event occurrence can be directly calculated as the posterior probability. As shown in Table 1, the posterior probability data table:

[0053] Table 1 Posterior probability data table

[0054]

[0055]

[0056] Figure 2 As shown in the figure, it is a posterior probability data diagram, which is obtained from Table 1 and Figure 2 It can be seen that the posterior probability of most groups is high. The posterior probability of groups 4, 6 and 9 is 1, indicating that the body temperature data of these groups fully meet the high-risk criteria and require special attention. The posterior probability will increase with the increase of the prior probability and conditional probability, and decrease with the increase of the total probability of the observed data. The prior probability reflects the possibility of a certain temperature state in the absence of observed data. When the prior probability increases, the posterior probability will also increase accordingly when other conditions remain unchanged. Similarly, the conditional probability indicates the possibility of observing specific data under a specific temperature state. If the conditional probability increases, the posterior probability will also increase when other conditions remain unchanged. The total probability of the observed data is the joint probability of all possible events. When the total probability increases, the posterior probability will decrease. If more historical data shows that a certain temperature state (such as high risk) occurs more frequently, or if the observed body temperature data is more frequent under a certain temperature state, then the corresponding prior probability and conditional probability will increase, resulting in an increase in the posterior probability. Conversely, if the overall frequency of observed body temperature data increases, then the total probability will also increase, resulting in a decrease in the posterior probability. Such analysis helps to more accurately assess regional temperature risks and optimize the effectiveness of monitoring and response systems.

[0057] Furthermore, the step of triggering automatic alarm includes: automatically alarming according to the risk assessment level and displaying the corresponding risk level; generating alarm information according to the alarm level; pushing the generated alarm information to the user in real time; and recording each alarm information.

[0058] In this embodiment, an alarm is automatically issued and the corresponding risk level is displayed according to the risk assessment level: the system uses the Bayesian network algorithm to conduct risk assessment on the abnormal temperature area and determine the risk assessment level (such as low, medium, and high risk). When a high risk or medium risk is detected, the system will automatically trigger the alarm mechanism and display the corresponding risk level on the monitoring interface. Through automatic alarm, the system can timely identify and respond to potential regional temperature risks and improve the efficiency and accuracy of emergency treatment. Generate alarm information according to the alarm level: the system generates detailed alarm information according to the detected risk assessment level. The alarm information includes the location of the abnormal temperature area, the specific risk level, the temperature data, the timestamp, and the recommended response measures. Push the generated alarm information to the user in real time: the system pushes the alarm information to relevant users and decision makers in real time through various channels (such as mobile phone notifications, emails, monitoring platforms, etc.) to ensure the timeliness and effectiveness of information transmission. Record each alarm information: the system records the alarm information generated each time in detail in the system log, and the log content includes the alarm time, alarm content, processing results, etc. These records can be used for subsequent query and analysis.

[0059] like Figure 3 As shown, it is a structural schematic diagram of the intelligent monitoring and response system based on the Internet of Things provided in an embodiment of the present application. The intelligent monitoring and response system based on the Internet of Things provided in an embodiment of the present application includes: a data acquisition module, a data preprocessing module, a comprehensive analysis module, a temperature marking module, a risk assessment module, and an automatic alarm module; wherein the data acquisition module is used to use a face recognition temperature measurement camera to obtain the body temperature data and facial image of the detected person in real time; the data preprocessing module is used to perform data preprocessing on the real-time body temperature data and facial image; the comprehensive analysis module is used to use a multi-source data fusion algorithm to perform a comprehensive analysis on the preprocessed data; the temperature marking module is used to identify and mark abnormal temperature areas by analyzing slight changes in body temperature data; the risk assessment module is used to perform risk assessment on abnormal temperature areas using a Bayesian network algorithm; the automatic alarm module is used to trigger an automatic alarm when a high-risk abnormal temperature area is detected.

[0060] In this embodiment, the data acquisition module is associated with the data preprocessing module: the body temperature data and facial images collected in real time by the data acquisition module are directly transmitted to the data preprocessing module. The preprocessing module performs preprocessing such as denoising and enhancement on these data to ensure the quality and reliability of the data. The data preprocessing module is associated with the comprehensive analysis module: the preprocessed data is transmitted to the comprehensive analysis module. The comprehensive analysis module adopts a multi-source data fusion algorithm to perform a comprehensive analysis on the preprocessed data and calculate a comprehensive score. The comprehensive analysis module is associated with the risk assessment module: the comprehensive score output by the comprehensive analysis module is transmitted to the risk assessment module. The risk assessment module uses a Bayesian network algorithm to calculate the posterior probability based on the comprehensive score, and performs risk assessment and level classification. The risk assessment module is associated with the automatic alarm module: when the risk assessment module detects a high-risk abnormal temperature area, the risk level information is transmitted to the automatic alarm module. The automatic alarm module generates and pushes alarm information to ensure that relevant personnel obtain and handle it in a timely manner.

[0061] An embodiment of the present application also provides an electronic device, including: a processor, and a memory for storing processor-executable instructions; when the processor is configured to execute the instructions, the electronic device implements an intelligent monitoring and response method based on the Internet of Things.

[0062] In summary, this embodiment adopts a multi-source data fusion algorithm to comprehensively analyze body temperature data and facial image data, thereby integrating information from different data sources, improving the accuracy and reliability of data analysis, reducing false alarms and missed alarms, and thus achieving accurate detection of regional temperature.

[0063] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0064] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0065] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0066] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0067] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0068] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent monitoring and response method based on the Internet of Things, characterized in that: The following steps are involved: Use the face recognition temperature measurement camera to obtain the body temperature data and facial image of the person being tested in real time; Perform data preprocessing on real-time body temperature data and facial images; Use multi-source data fusion algorithm to conduct comprehensive analysis on pre-processed data; By analyzing small changes in body temperature data, abnormal temperature areas can be identified and marked; Use Bayesian network algorithm to conduct risk assessment on abnormal temperature areas; When a high-risk abnormal temperature area is detected, an automatic alarm is triggered; The specific steps of the data preprocessing are: Grayscale the collected facial image and convert the color image into a grayscale image; Use Gaussian filtering algorithm to smooth the facial image and remove high-frequency noise; The body temperature data is smoothed using the moving average filtering method to remove the noise in the body temperature data; Use histogram equalization technology to enhance the facial image; Calculate the grayscale histogram of the image and obtain the number of pixels at each grayscale level; Calculate the cumulative distribution function to map the grayscale value of the original image to the new grayscale value; Reconstruct the facial image based on the new grayscale values ​​to enhance the contrast and details of the image; The facial image reconstructed using a convolutional neural network is used for feature extraction, and the locations of key facial points and corresponding temperature values ​​are extracted to form a temperature distribution map; The specific steps of using the multi-source data fusion algorithm to conduct comprehensive analysis on the pre-processed data are: Define the data source for body temperature data and facial image data; Standardize data from different data sources; Assign weights to each data source and calculate a composite score for each data source; The comprehensive scores of all data sources are weighted averaged to obtain the final fusion score; The steps of analyzing small changes in body temperature data and identifying and marking abnormal temperature areas are as follows: Select a fixed-size time window, perform window processing on the body temperature data, and record the body temperature value at each moment in the window; Calculate the temperature change rate within the current time window; Set the threshold of the temperature change rate. If the temperature change rate exceeds the threshold, the temperature in the current time window is marked as an abnormal temperature area. Extract key features from the marked abnormal temperature areas and mark them; The step of using the Bayesian network algorithm to conduct risk assessment on abnormal temperature areas includes: Based on historical health data and statistical information, define the prior probability of regional temperature risk corresponding to the abnormal temperature area; Calculate the conditional probability based on the abnormal temperature area data; Calculate the total probability of the current observation data; The posterior probability of regional temperature risk is calculated using the Bayesian formula based on the conditional probability and the total probability of the current observation data; Set multiple risk level thresholds and divide risk levels; According to the value of the posterior probability, the risk assessment level is obtained.

2. The intelligent monitoring and response method based on the Internet of Things as claimed in claim 1, characterized in that: The calculation formula of the final fusion score is: S represents the final fusion score, ω b The weight of the b-th data source, b represents the weight number, b=1,2,...z, z represents the total number, X b represents the data value of the bth data source collected in real time, μ b represents the average value of the historical data of the bth data source, σ b Indicates the standard deviation of the historical data of the bth data source.

3. The intelligent monitoring and response method based on the Internet of Things as claimed in claim 1, characterized in that: The calculation formula of the posterior probability is: In the formula, P represents the posterior probability, A represents a specific temperature state, and A i represents all possible events with all features, b i represents the feature of the current observation data, i represents the feature number of the current observation data, i=1,2,...,n, n represents the total number of features, P(A) represents the prior probability, P(b i |A) represents the conditional probability, P(A i ) represents the prior probability of all possible events, P(A i ) represents the conditional probability of all possible events, Represents the total probability.

4. The intelligent monitoring and response method based on the Internet of Things as claimed in claim 1, characterized in that: The step of triggering automatic alarm includes: Automatically give an alarm based on the risk assessment level and display the corresponding risk level; Generate alarm information according to the alarm level; Push the generated alarm information to users in real time; Record each alarm information.

5. A system using the intelligent monitoring and response method based on the Internet of Things as described in any one of claims 1 to 4, characterized in that: include: Data acquisition module, data preprocessing module, comprehensive analysis module, temperature marking module, risk assessment module, automatic alarm module; The data acquisition module is used to acquire the body temperature data and facial image of the detected person in real time using a face recognition temperature measurement camera; The data preprocessing module is used to perform data preprocessing on real-time body temperature data and facial images; The comprehensive analysis module is used to perform comprehensive analysis on the pre-processed data using a multi-source data fusion algorithm; The temperature marking module is used to identify and mark abnormal temperature areas by analyzing slight changes in body temperature data; The risk assessment module is used to perform risk assessment on abnormal temperature areas using a Bayesian network algorithm; The automatic alarm module is used to trigger an automatic alarm when a high-risk abnormal temperature area is detected.

6. An electronic device, characterized in that: include: a processor, a memory for storing instructions executable by the processor; When the processor is configured to execute the instructions, the electronic device implements the intelligent monitoring response method based on the Internet of Things as described in any one of claims 1-4.

Images