Intelligent security alarm method and device based on Internet of Things, and electronic equipment

Abstract

The invention relates to the technical field of intelligent security and protection, in particular to an intelligent security and protection alarm method and device based on the Internet of Things and electronic equipment, and the method comprises the following steps: S1, collecting multi-dimensional environment data in a monitoring area; s2, analyzing the received data in real time, and identifying a potential threat scene; s3, according to the identified threat scene, dynamically adjusting the acquisition frequency and sensitivity of various sensors so as to confirm the authenticity of the potential threat scene; s4, after the threat scene is confirmed, the cloud server triggers a multi-level alarm mechanism; s5, the user communicates with the cloud server through the mobile device; and S6, after each alarm event is ended, automatically generating a threat response report. According to the invention, by realizing multi-modal data fusion and enhancing user interaction, the response speed and accuracy of the security and protection system are remarkably improved, the false alarm rate is reduced, and the convenience of user operation and the overall reliability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent security, and particularly to an intelligent security alarm method, device and electronic device based on the Internet of Things. Background Art

[0002] With the wide application of Internet of Things technology, the demand for intelligent security systems in the fields of residential, enterprise and public security is increasing day by day. Existing security systems mostly use traditional monitoring devices and single data sources for threat detection and response, such as using a single type of camera or simple door and window sensors. Although these systems can achieve basic monitoring and alarm functions, they usually have problems of data isolation and limited analysis capabilities, and it is difficult to effectively process and analyze a large amount of data from multiple sensors, resulting in insufficiently sensitive security responses or a high false alarm rate.

[0003] In the face of the limitations of the existing technology, the present invention is committed to solving the problems of data isolation and low processing efficiency in intelligent security systems. Currently, most systems lack an effective data fusion mechanism and cannot make full use of the rich data resources provided by various sensors, such as multi-dimensional information such as temperature, humidity, gas concentration, etc. This limits the early warning ability and accuracy of the system. In addition, existing security systems also show certain deficiencies in user interaction and remote management, such as the lack of real-time remote monitoring functions and an instant feedback mechanism for alarm information.

[0004] Therefore, developing an intelligent security alarm system integrating multi-modal data processing and efficient user interaction functions has become a necessary step to improve security efficiency and user experience. Summary of the Invention

[0005] Based on the above purposes, the present invention provides an intelligent security alarm method, device and electronic device based on the Internet of Things.

[0006] The intelligent security alarm method based on the Internet of Things includes the following steps: S1: Deploy a multi-modal sensor array inside the monitoring area for collecting multi-dimensional environmental data in the monitoring area, including temperature, humidity, gas concentration, light intensity, noise level and vibration frequency; the sensor array is connected to the cloud server through an Internet of Things gateway; S2: The cloud server receives the environmental data from the multi-modal sensor array and performs real-time analysis on the received data using a preset threat scenario recognition model. The threat scenario recognition model is constructed based on a multi-modal data fusion algorithm and is used to identify potential threat scenarios. The potential threat scenarios include fire, gas leakage, forced intrusion and noise pollution; S3: When the threat scenario recognition model identifies a specific potential threat scenario, it will send instructions to each sensor through the IoT gateway according to the recognized threat scenario, dynamically adjusting the acquisition frequency and sensitivity of various sensors to capture the dynamic data of the recognized threat scenario to confirm the authenticity of the potential threat scenario; S4: After the threat scenario is confirmed, the cloud server will trigger a multi-level alarm mechanism, and the multi-level alarm mechanism includes a first-level alarm, a second-level alarm, and a third-level alarm; The first-level alarm is used to send an alarm signal through the local audible and visual alarm device to prompt the personnel in the monitoring area; The second-level alarm is used to send the alarm information to the user's mobile device through the IoT gateway, and the alarm information includes the threat scenario, the threat occurrence location, and relevant sensor data; The third-level alarm is used to automatically send a remote alarm signal to the preset security personnel, fire, or law enforcement agency when it is determined that the threat level reaches a predetermined severe threshold; S5: The user will communicate with the cloud server through the mobile device, be able to remotely view the real-time monitoring video and the environmental data of relevant sensors, and at the same time the user can confirm or cancel the sent alarm information through the mobile device; S6: After each alarm event ends, all sensor data during the threat occurrence will be automatically stored in the cloud server for generating a threat response report.

[0007] Optionally, the specific content of S1 includes: S11: Layout a multi-modal sensor array in the monitoring area, and the sensor array includes a temperature sensor, a humidity sensor, a gas concentration sensor, a light intensity sensor, a noise sensor, and a vibration sensor; The temperature sensor is used to detect the temperature in the monitoring area in real time; the humidity sensor is used to continuously monitor the environmental humidity; the gas concentration sensor is responsible for monitoring the concentration of harmful gases in the air; the light intensity sensor is used to record the light intensity in the environment; the noise sensor detects the noise level in the environment; the vibration sensor is used to detect the vibration frequency and intensity in the environment; The multi-dimensional environmental data collected by the multi-modal sensors is transmitted to the IoT gateway through the wireless communication module, and the IoT gateway establishes a data connection with the cloud server through Wi-Fi or 4G network. After receiving the data transmitted by the sensor, the gateway will immediately encrypt the data and send it to the cloud server through the secure transmission protocol; S14: After receiving the environmental data from the IoT gateway, the cloud server will maintain real-time communication with the IoT gateway through a preset interface protocol to ensure that the sensor array can continuously send data to the cloud server, and the server will mark and distinguish the data streams of each sensor.

[0008] Optionally, the specific steps of S2 are as follows: S21: The cloud server constructs a threat scenario recognition model based on a multi-modal data fusion algorithm, which is composed of a multi-layer neural network combined with a Bayesian inference model. The neural network is used to extract feature data from different sensors and perform preliminary feature extraction; the Bayesian inference model performs probability estimation through historical data to identify the probabilities of various potential threat scenarios. S22: The threat scenario recognition model processes the data of various sensors in terms of time series and assigns weights by inputting the data from the multi-modal sensors. The time series processing is used to analyze the change trends of different data sources at different time points, and the weight assignment is dynamically adjusted according to the relevance of specific threat scenarios in historical data. S23: After the cloud server receives the real-time environmental data from the multi-modal sensor array, it transmits the data to the input layer of the threat scenario recognition model, performs feature extraction through the multi-layer neural network structure of the model, and converts the environmental data collected by various sensors into standardized feature vectors to form a multi-dimensional feature set. S24: The threat scenario recognition model calculates based on the multi-dimensional feature set and combines the probability estimation of the Bayesian inference model to determine whether there are potential threat scenarios related to fire, gas leakage, forced intrusion, or noise pollution in the current data. S25: When the calculation result of the threat scenario recognition model shows that the probability of a certain threat scenario exceeds the preset alarm threshold, the cloud server immediately issues a warning signal and records the current analysis result and corresponding data features; if the data does not reach the threat threshold, continue to monitor and update the real-time data input.

[0009] Optionally, the specific steps of S21 are as follows: S211: The cloud server standardizes the data from the multi-modal sensors: S212: Based on the standardized data, a multi-layer neural network model including an input layer and a hidden layer is constructed for preliminary feature extraction. S213: On the basis of feature extraction, a Bayesian inference model is used to estimate the probability of threat scenarios. The formula of Bayes' theorem is: , where is the given data when the threat scenario is the posterior probability, For the probability of data occurring in the threat scenario, for the prior probability of the threat scenario, and for the total probability of the data; S214: When fusing multi-modal data, the weighted average method is used to perform weighted calculations on the data of different sensors, and the fusion result formula is: , where is the total output after fusion, is the weight of the th sensor, is the th sensor's data, is the number of sensors.

[0010] Optionally, the S3 specifically includes: S31: When the threat scenario recognition model detects a potential threat scenario, the cloud server will transmit the recognition result to each sensor control module through the IoT gateway and generate a control signal containing instructions. The control signal includes preset parameters for adjusting the acquisition frequency and sensitivity of relevant sensors; S32: After receiving the control signal from the cloud server, the IoT gateway will transmit the control signal to the relevant sensors according to the threat scenario requirements. Let the frequency adjustment parameter in the control signal be and the sensitivity adjustment parameter be ; S33: The adjusted sensors start to work at the new acquisition frequency and sensitivity , focusing on capturing dynamic data related to the threat scenario, and monitoring the fluctuations of sensor data in the short term through time series analysis to ensure the accuracy of the identified threat scenario. The formula for time series analysis is: , where is the total sum of dynamic data at the monitoring time point , is the th sensor's acquisition data at the time point , is the number of sensors; S34: The cloud server continuously receives the adjusted sensor data stream and compares it with the threat scenario recognition model to confirm the change trend of the threat scenario; once the data fluctuation exceeds the preset range, the potential threat scenario is confirmed as real.

[0011] Optionally, the S4 specifically includes: S41: After the threat scenario is confirmed, the cloud server triggers a corresponding multi-level alarm mechanism according to the severity of the threat. Specifically, the threat level is divided into a first-level alarm , a second-level alarm and a third-level alarm . The threat level is calculated by the following formula: , where represents the posterior probability of the threat scenario calculated according to the Bayesian inference model , indicating the probability of the threat scenario occurring, represents the total dynamic fluctuation of the sensor data at time point , and are the adjustment coefficients of the posterior probability and the dynamic data fluctuation respectively; S42: According to the calculated threat level , when the threat level meets the following conditions, the cloud server triggers a corresponding multi-level alarm; When , the cloud server triggers a first-level alarm; When , the cloud server triggers a second-level alarm; When , the cloud server triggers a third-level alarm.

[0012] Optionally, the S5 specifically includes: S51: The user communicates with the cloud server through a mobile device. A dedicated application program for docking with the cloud server is installed on the mobile device. The application program establishes a secure connection with the cloud server through the HTTPS protocol. After the user logs in to the application program, the cloud server transmits real-time monitoring data to the mobile device. The data includes the camera video stream in the monitoring area and the environmental data of various sensors; S52: The real-time monitoring screen is compressed through video stream encoding technology to reduce bandwidth consumption. After the video data is transmitted to the cloud server through the Internet of Things gateway, it is forwarded in real time to the user's mobile device. The user can view the high-definition monitoring screen in the application program and switch different camera perspectives as needed; S53: When the cloud server triggers an alarm, the application program will receive an alarm notification. The notification content includes the threat type, the location of occurrence, the threat level, and the real-time data of relevant sensors. The user can view the alarm information through the application program interface; S54: The user can confirm or cancel the alarm information sent in the application program.

[0013] Optionally, the threat response report generated in S6 includes the basic information of the threat event, the specific response of the sensors, the preliminary analysis results of the threat event, and the operation records of the user during the event; the basic information of the threat event includes the time, location, threat type and level of the event; the specific response of the sensors includes the time series data charts of various sensors; the preliminary analysis results of the threat event include the duration, fluctuation range and response of the threat; the operation records of the user during the event include the user's confirmation or cancellation operations of the alarm and the corresponding time points.

[0014] An intelligent security alarm device based on the Internet of Things is used to implement the above-mentioned intelligent security alarm method based on the Internet of Things, and includes the following modules: Multimodal sensor module: Deployed in the monitoring area, it is used to collect multi-dimensional environmental data, including data on temperature, humidity, gas concentration, light intensity, noise level and vibration frequency; Internet of Things gateway module: It is used to receive real-time data from the multimodal sensor module and connect to the cloud server through an encrypted wireless communication protocol to ensure the security and real-time of data transmission; Cloud server module: It is used to receive sensor data from the Internet of Things gateway module and perform real-time analysis on the received data using a threat scenario recognition model. The threat scenario recognition model is constructed based on a multimodal data fusion algorithm and can identify potential threat scenarios of fire, gas leakage, and intrusion, and trigger an alarm mechanism according to the recognition results; Alarm module: It includes a local alarm device and a remote alarm device. The local alarm device receives an alarm instruction from the cloud server module through the Internet of Things gateway module and triggers an audible and visual alarm signal; the remote alarm device is used to send alarm information to the user's mobile device through the Internet of Things gateway; User terminal module: By communicating with the cloud server module, it enables the user to access the security alarm device through a mobile device or a web browser, remotely view real-time monitoring images and sensor data, and confirm or cancel the alarm information sent by the security alarm device.

[0015] An electronic device includes a processor and a storage device. A computer program is stored on the storage device. When the computer program is run by the processor, it is used to execute the steps of the above-mentioned intelligent security alarm method based on the Internet of Things.

[0016] The beneficial effects of the present invention: In the present invention, by adopting a multi-modal data fusion algorithm, data from various sensors such as temperature, humidity, gas concentration, etc. are effectively integrated, thereby significantly enhancing the monitoring ability and recognition accuracy of the security system for various environmental changes. This algorithm utilizes advanced data processing technologies to be able to analyze and interpret complex environmental data in real time, ensuring a prompt response when potential threats are detected. In addition, the application of this technology reduces false alarms. Through precise data analysis, alarms are triggered only when a real threat is confirmed, thereby improving the reliability of the system and the trust of users.

[0017] In the present invention, through a real-time monitoring and instant feedback mechanism, users can directly view the monitoring images through a mobile device and manage the alarm information generated by the system. This interactive method not only enables users to control the security system more effectively but also allows for a quick response in case of an emergency, thereby enhancing security. Brief Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only those of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a schematic flowchart of the intelligent security alarm method according to an embodiment of the present invention; Figure 2 It is a schematic diagram of the composition of the intelligent security alarm device according to an embodiment of the present invention. Detailed Embodiments

[0020] The present invention will be described in detail below in conjunction with the drawings and specific embodiments. At the same time, it should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative methods for implementation; moreover, the drawings are only for a more specific description of the embodiments and are not intended to specifically limit the present invention.

[0021] It should be pointed out that in the specification, when referring to "an embodiment", "embodiments", "exemplary embodiments", "some embodiments", etc., it indicates that the described embodiments may include specific features, structures or characteristics, but not necessarily every embodiment includes such specific features, structures or characteristics. Additionally, when combining embodiments to describe specific features, structures or characteristics, implementing such features, structures or characteristics in combination with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the relevant art.

[0022] Generally, terms can be understood at least in part from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or can be used to describe a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey a set of exclusive factors, but rather can alternatively, depending at least in part on the context, allow for the existence of other factors that are not necessarily explicitly described.

[0023] As Figure 1 shown, an Internet of Things-based intelligent security alarm method includes the following steps: S1: Deploy a multi-modal sensor array within the monitoring area for collecting multi-dimensional environmental data in the monitoring area, including temperature, humidity, gas concentration, light intensity, noise level, and vibration frequency; the sensor array is connected to a cloud server through an Internet of Things gateway; S2: The cloud server receives the environmental data from the multi-modal sensor array and performs real-time analysis on the received data using a preset threat scenario recognition model. The threat scenario recognition model is constructed based on a multi-modal data fusion algorithm for identifying potential threat scenarios, and the potential threat scenarios include fire, gas leakage, forced intrusion, and noise pollution; S3: When the threat scenario recognition model identifies a specific potential threat scenario, it will send instructions to each sensor through the Internet of Things gateway according to the identified threat scenario to dynamically adjust the collection frequency and sensitivity of various sensors for capturing dynamic data of the identified threat scenario to confirm the authenticity of the potential threat scenario; S4: After the threat scenario is confirmed, the cloud server will trigger a multi-level alarm mechanism, and the multi-level alarm mechanism includes a first-level alarm, a second-level alarm, and a third-level alarm; The first-level alarm is used to emit an alarm signal through a local audible and visual alarm device to prompt the personnel within the monitoring area; The second-level alarm is used to send the alarm information to the user's mobile device through the Internet of Things gateway, and the alarm information includes the threat scenario, the location where the threat occurred, and relevant sensor data; The third-level alarm is used to automatically send a remote alarm signal to a preset security personnel, fire department, or law enforcement agency when it is determined that the threat level reaches a predetermined severe threshold; S5: The user will communicate with the cloud server through the mobile device to remotely view the real-time monitoring video and the environmental data of relevant sensors. At the same time, the user can confirm or cancel the sent alarm information through the mobile device; S6: After each alarm event ends, all sensor data at the time of the threat occurrence will be automatically stored in the cloud server for generating a threat response report and providing data support for the optimization of the subsequent threat scenario recognition model.

[0024] S1 specifically includes: S11: Layout a multi-modal sensor array within the monitoring area. The sensor array includes a temperature sensor, a humidity sensor, a gas concentration sensor, a light intensity sensor, a noise sensor, and a vibration sensor. The sensors are reasonably distributed according to the environmental characteristics and potential threat points in the monitoring area to ensure no blind spots in the monitoring range, and the position of each sensor is fixedly installed to ensure the stable operation of the sensor; S12: The temperature sensor is used to detect the temperature in the monitoring area in real time; the humidity sensor is used to continuously monitor the environmental humidity; the gas concentration sensor is responsible for monitoring the concentration of harmful gases in the air; the light intensity sensor is used to record the light intensity in the environment; the noise sensor detects the noise level in the environment; the vibration sensor is used to detect the vibration frequency and intensity in the environment; all sensors are powered by the built-in power module and continuously collect data; S13: The multi-dimensional environmental data collected by the multi-modal sensors is transmitted to the Internet of Things gateway through the wireless communication module. The Internet of Things gateway establishes a data connection with the cloud server through Wi-Fi or 4G network. After receiving the data transmitted by the sensors, the gateway will immediately encrypt the data and send it to the cloud server through the secure transmission protocol to ensure the security and real-time of the data transmission process; S14: After receiving the environmental data from the Internet of Things gateway, the cloud server will maintain real-time communication with the Internet of Things gateway through the preset interface protocol to ensure that the sensor array can continuously send data to the cloud server, and the server will mark and distinguish the data stream of each sensor to ensure the accuracy and stability of data processing; by reasonably deploying the multi-modal sensor array, all potential threat points in the monitoring area can be effectively covered to ensure the comprehensiveness of data collection. The specific functions of each type of sensor are clearly defined, and through the stable connection between the Internet of Things gateway and the cloud server, real-time and stable data transmission is achieved, which helps to accurately analyze abnormal situations in the environment and improve the accuracy and reaction speed of threat recognition.

[0025] S2 specifically includes: S21: The cloud server constructs a threat scenario recognition model based on the multi-modal data fusion algorithm. The multi-modal data fusion algorithm is composed of the combination of a multi-layer neural network and a Bayesian inference model. The neural network is used to extract the feature data from different sensors and perform preliminary feature extraction; the Bayesian inference model performs probability estimation through historical data to identify the probabilities of various potential threat scenarios; the training process of this model is based on a large-scale historical data set, covering various threat scenarios such as fire, gas leakage, forced intrusion, and noise pollution; S22: The threat scenario recognition model processes the data from multi-modal sensors in time series and assigns weights to the data of various sensors by using a data fusion algorithm. The time series processing is used to analyze the change trends of different data sources at different time points, and the weight assignment is dynamically adjusted according to the relevance of specific threat scenarios in historical data to ensure the accuracy of the model under multi-dimensional data; S23: After receiving the real-time environmental data from the multi-modal sensor array, the cloud server transmits the data to the input layer of the threat scenario recognition model, extracts features through the multi-layer neural network structure of the model, and converts the environmental data collected by various sensors into standardized feature vectors to form a multi-dimensional feature set; S24: The threat scenario recognition model calculates based on the multi-dimensional feature set and combines the probability estimation of the Bayesian inference model to determine whether there are potential threat scenarios related to fire, gas leakage, forced intrusion or noise pollution in the current data; S25: When the calculation result of the threat scenario recognition model shows that the probability of a certain threat scenario exceeds the preset alarm threshold, the cloud server immediately issues a warning signal and records the current analysis result and corresponding data features; if the data does not reach the threat threshold, continue to monitor and update the real-time data input to ensure the continuity and accuracy of the recognition result.

[0026] The construction of the threat scenario recognition model based on the multi-modal data fusion algorithm in S21 specifically includes: S211: The cloud server performs standardization processing on the data from multi-modal sensors. The specific formula is: , where represents the original data collected by the sensor, represents the mean of the original data, represents the standard deviation of the original data, represents the standardized data. Through this standardization process, it is ensured that the data from different sensors have the same scale during model processing: S212: Based on the standardized data, a multi-layer neural network model including an input layer and a hidden layer is constructed for preliminary feature extraction. The input layer is used to receive the data from multi-modal sensors, and the input feature vector is expressed as: , where represents the input data vector, is the standardized data of various sensors. The hidden layer processes the input data through the ReLU activation function. The expression of the ReLU activation function is: , where represents the input data, represents the activated output data; the output feature vector after being processed by the hidden layer is: , where is the extracted feature vector, is the intermediate feature after processing; S213: Based on feature extraction, use the Bayesian inference model to estimate the probability of threat scenarios. The formula of Bayes' theorem is: , where is the given data when the threat scenario the posterior probability of is the threat scenario the probability of the occurrence of data under is the threat scenario the prior probability of is the total probability of data ; This inference model is used to estimate the probabilities of different threat scenarios; S214: When fusing multi-modal data, use the weighted average method to perform weighted calculations on the data of different sensors. The fusion result formula is: , where is the total output after fusion, is the weight of the th sensor, is the th sensor data, is the number of sensors, and the weight is dynamically adjusted according to the relevance of each sensor to the threat scenario to ensure accurate judgment of key threats; Through standardization processing, multi-layer neural network feature extraction, probability estimation of the Bayesian inference model, and data fusion of the weighted average method, the threat scenario recognition model can effectively fuse multi-modal data and has the ability to dynamically adjust weights under different threat scenarios, improving the recognition accuracy and real-time performance of the model, and ensuring efficient and accurate threat judgment in complex scenarios.

[0027] S3 specifically includes: S31: When the threat scenario recognition model detects a potential threat scenario, the cloud server will transmit the recognition result to each sensor control module through the IoT gateway and generate a control signal containing instructions. The control signal includes preset parameters for adjusting the acquisition frequency and sensitivity of relevant sensors; S32: After receiving the control signal from the cloud server, the IoT gateway will transmit the control signal to the relevant sensors according to the threat scenario requirements. Let the frequency adjustment parameter in the control signal be and the sensitivity adjustment parameter be ; Among them, the acquisition frequency adjustment parameter Indicates the sampling frequency of each sensor, which is dynamically adjusted according to the urgency of the threat scenario through the following formula: , where is the initial sampling frequency, is the adjustment coefficient, which is determined according to the severity of the threat scenario. The higher the threat level, the higher the sampling frequency; the sensitivity adjustment parameter varies dynamically according to the type of threat scenario. The sensitivity adjustment formula is: , where is the initial sensitivity, is the coefficient adjusted according to the threat scenario. The coefficient is determined according to the degree of association between the sensor and the threat scenario; S33: The adjusted sensor starts to work at the new acquisition frequency and sensitivity , focusing on capturing dynamic data related to the threat scenario, monitoring the fluctuations of sensor data in the short term through time series analysis to ensure the accuracy of the identified threat scenario. The formula for time series analysis is: , where is the total sum of dynamic data at the monitoring time point , is the acquisition data of the nd sensor at the time point , is the number of sensors; S34: The cloud server continuously receives the adjusted sensor data stream and compares it with the threat scenario recognition model to confirm the change trend of the threat scenario; once the data fluctuation exceeds the preset range, the potential threat scenario is confirmed as real; by dynamically adjusting the acquisition frequency and sensitivity of the sensor, it can quickly respond to potential threats, enhance the real-time monitoring ability of the threat scenario, and time series analysis is used to capture the dynamic changes of the threat scenario to ensure that during the occurrence of the threat, it can continuously monitor and adjust the response strategy, thereby effectively improving the accuracy and timeliness of threat scenario confirmation.

[0028] S4 specifically includes: S41: When the threat scenario is confirmed, the cloud server triggers the corresponding multi-level alarm mechanism according to the severity of the threat. Specifically, the threat level is divided into level 1 alarm , level 2 alarm and level 3 alarm , and the threat level is calculated through the following formula: , where represents the posterior probability of the threat scenario calculated according to the Bayesian inference model, indicating the probability of the threat scenario occurring, Indicates a time point The total dynamic fluctuation of sensor data and are the adjustment coefficients of the posterior probability and the dynamic data fluctuation respectively, which are used to balance the weights of the two and ensure that the calculated threat level can accurately reflect the severity of the threat; S42: According to the calculated threat level When the threat level meets the following conditions, the cloud server triggers corresponding multi-level alarms; When , the cloud server triggers a first-level alarm; the first-level alarm is activated through a local audible and visual alarm device to notify the personnel in the monitoring area with visual and auditory signals. The audible and visual alarm device is connected to the cloud server wirelessly or wiredly and starts immediately after receiving the signal; When , the cloud server triggers a second-level alarm; the second-level alarm sends the alarm information to the user's mobile device through the Internet of Things gateway. The alarm information includes the threat type, the location where the threat occurred, the real-time data of relevant sensors, and the calculated threat level. The user's mobile device receives the alarm information through a dedicated application connected to the server and can view the detailed information and perform remote operations; When , the cloud server triggers a third-level alarm; the third-level alarm sends detailed alarm information to the preset security personnel, fire or law enforcement agencies through a remote communication channel. The information of the third-level alarm includes a detailed description of the threat scenario, the real-time dynamic data changes, and relevant historical records; the above steps can accurately determine the severity of the threat through a clear threat level calculation formula and a preset threshold range, and trigger corresponding alarm mechanisms according to the threat level. Different levels of alarms can effectively distinguish the urgency of the threat, thus ensuring the timely response and effective handling of potential threats.

[0029] S5 specifically includes: S51: The user communicates with the cloud server through a mobile device. A dedicated application for docking with the cloud server is installed on the mobile device. The application establishes a secure connection with the cloud server through the HTTPS (HyperText Transfer Protocol Secure) protocol. After the user logs in to the application, the cloud server transmits real-time monitoring data to the mobile device. The data includes the camera video stream in the monitoring area and the environmental data of various sensors; S52: The real-time monitoring screen is compressed through video stream encoding technology (such as H.264 or H.265) to reduce bandwidth consumption. After the video data is transmitted to the cloud server through the IoT gateway, it is forwarded in real-time to the user's mobile device. The user can view the high-definition monitoring screen within the application and switch different camera perspectives as needed. In addition, the data of relevant sensors (such as temperature, humidity, gas concentration, etc.) are synchronously displayed in the form of charts or numbers on the application interface, and the user can monitor the changes of various environmental data in real-time; S53: When the cloud server triggers an alarm, the application will receive an alarm notification. The notification content includes the threat type, occurrence location, threat level, and real-time data of relevant sensors. The user can view the alarm information through the application interface; S54: The user can confirm or cancel the issued alarm information in the application: The confirmation operation is used when the user confirms that the alarm is valid. The application will generate a confirmation signal. The confirmation signal is transmitted to the cloud server through the mobile device. The cloud server records the user's confirmation operation and marks the alarm information as the confirmed status. The confirmed alarm information will automatically generate a threat handling record for subsequent reference. Cancellation operation: If the user determines that the alarm is a false alarm, the alarm can be cancelled. The application generates a cancellation signal. The cancellation signal is transmitted to the cloud server through the mobile device. After receiving the cancellation signal, the cloud server immediately aborts the current alarm mechanism, turns off the audible and visual alarm device, and notifies the relevant parties to stop further response measures. The cancellation operation will also be recorded in the log for subsequent analysis.

[0030] The threat response report generated in S6 includes the basic information of the threat event, the specific response situation of the sensors, the preliminary analysis results of the threat event, and the operation records of the user in the event; The basic information of the threat event includes the time, location, threat type, and level of the event; The specific response situation of the sensors includes the time series data charts of various sensors, and the key change points of the threat event are marked on the charts; For the preliminary analysis results of the threat event, the cloud server gives an assessment of the threat based on the dynamic change trend of the sensor data, including the duration, fluctuation range, and response situation of the threat; The operation records of the user in the event include the user's confirmation or cancellation operation of the alarm and the corresponding time points.

[0031] As Figure 2 shown, the intelligent security alarm device based on the Internet of Things is used to implement the above-mentioned intelligent security alarm method based on the Internet of Things, and includes the following modules: Multimodal sensor module: Deployed in the monitoring area, it is used to collect multi-dimensional environmental data, including data on temperature, humidity, gas concentration, light intensity, noise level, and vibration frequency; Internet of Things gateway module: It is used to receive real-time data from the multimodal sensor module and connect to the cloud server through an encrypted wireless communication protocol to ensure the security and real-time nature of data transmission; Cloud server module: It is used to receive sensor data from the Internet of Things gateway module and perform real-time analysis on the received data using a threat scenario recognition model. The threat scenario recognition model is built based on a multimodal data fusion algorithm, which can identify potential threat scenarios such as fire, gas leakage, and intrusion, and trigger an alarm mechanism according to the recognition results; Alarm module: It includes a local alarm device and a remote alarm device. The local alarm device receives an alarm instruction from the cloud server module through the Internet of Things gateway module and triggers an audible and visual alarm signal; the remote alarm device is used to send alarm information to the user's mobile device through the Internet of Things gateway; User terminal module: By communicating with the cloud server module, it enables users to access the security alarm device through a mobile device or a web browser, remotely view real-time monitoring images and sensor data, and confirm or cancel the alarm information sent by the security alarm device.

[0032] An electronic device includes a processor and a storage device. A computer program is stored on the storage device. When the computer program is run by the processor, it is used to execute the steps of the above-mentioned Internet of Things-based intelligent security alarm method.

[0033] The present invention covers any alternatives, modifications, equivalent methods, and solutions made within the essence and scope of the present invention. To enable the public to have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention. However, those skilled in the art can fully understand the present invention without these detailed descriptions. Additionally, to avoid unnecessary confusion to the essence of the present invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0034] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent security alarm method based on the Internet of Things, characterized in that, Including the following steps: S1: Deploy a multi-modal sensor array inside the monitoring area to collect multi-dimensional environmental data in the monitoring area, including temperature, humidity, gas concentration, light intensity, noise level, and vibration frequency; the sensor array is connected to the cloud server through an IoT gateway; S2: The cloud server receives the environmental data from the multi-modal sensor array and uses a preset threat scenario recognition model to perform real-time analysis on the received data. The threat scenario recognition model is constructed based on a multi-modal data fusion algorithm and is used to identify potential threat scenarios. The potential threat scenarios include fire, gas leakage, forced intrusion, and noise pollution; S3: When the threat scenario recognition model identifies a specific potential threat scenario, it will send instructions to each sensor through the IoT gateway according to the identified threat scenario to dynamically adjust the acquisition frequency and sensitivity of various sensors to capture the dynamic data of the identified threat scenario to confirm the authenticity of the potential threat scenario; S4: After the threat scenario is confirmed, the cloud server will trigger a multi-level alarm mechanism. The multi-level alarm mechanism includes a first-level alarm, a second-level alarm, and a third-level alarm; The first-level alarm is used to send an alarm signal through a local acoustic-optic alarm device to alert the personnel in the monitoring area; The second-level alarm is used to send the alarm information to the user's mobile device through the IoT gateway. The alarm information includes the threat scenario, the location where the threat occurred, and relevant sensor data; The third-level alarm is used to automatically send a remote alarm signal to a preset security personnel, fire department, or law enforcement agency when it is determined that the threat level reaches a predetermined severe threshold; S5: The user will communicate with the cloud server through the mobile device to remotely view the real-time monitoring video and the environmental data of relevant sensors. At the same time, the user can confirm or cancel the sent alarm information through the mobile device; S6: After each alarm event ends, all sensor data at the time of the threat occurrence will be automatically stored in the cloud server for generating a threat response report.

2. The intelligent security alarm method based on the Internet of Things according to claim 1, wherein The specific content of S1 includes: S11: Layout a multi-modal sensor array in the monitoring area. The sensor array includes a temperature sensor, a humidity sensor, a gas concentration sensor, a light intensity sensor, a noise sensor, and a vibration sensor; S12: The temperature sensor is used to detect the temperature in the monitoring area in real time; the humidity sensor is used to continuously monitor the environmental humidity; the gas concentration sensor is responsible for monitoring the concentration of harmful gases in the air; the light intensity sensor is used to record the light intensity in the environment; the noise sensor detects the noise level in the environment; the vibration sensor is used to detect the vibration frequency and intensity in the environment; S13: The multi-dimensional environmental data collected by the multi-modal sensors is transmitted to the IoT gateway through a wireless communication module. The IoT gateway establishes a data connection with the cloud server through Wi-Fi or 4G network. After receiving the data transmitted by the sensors, the gateway will immediately encrypt the data and send it to the cloud server through a secure transmission protocol; S14: After receiving the environmental data from the IoT gateway, the cloud server will maintain real-time communication with the IoT gateway through a preset interface protocol to ensure that the sensor array can continuously send data to the cloud server, and the server will mark and distinguish the data streams of each sensor.

3. The intelligent security alarm method based on the Internet of Things according to claim 1, characterized in that, The specific steps of S2 are as follows: S21: The cloud server constructs a threat scenario recognition model based on a multi-modal data fusion algorithm. The multi-modal data fusion algorithm is composed of a multi-layer neural network combined with a Bayesian inference model. The neural network is used to extract feature data from different sensors and perform preliminary feature extraction. The Bayesian inference model estimates probabilities through historical data to identify the probabilities of various potential threat scenarios. S22: The threat scenario recognition model processes the data of various sensors in terms of time series and assigns weights by inputting data from multi-modal sensors using the data fusion algorithm. The time series processing is used to analyze the change trends of different data sources at different time points, and the weight assignment is dynamically adjusted according to the relevance of threat scenarios in historical data. S23: After the cloud server receives the real-time environmental data from the multi-modal sensor array, it transmits the data to the input layer of the threat scenario recognition model, performs feature extraction through the multi-layer neural network structure of the model, and converts the environmental data collected by various sensors into standardized feature vectors to form a multi-dimensional feature set. S24: The threat scenario recognition model calculates based on the multi-dimensional feature set and combines the probability estimation of the Bayesian inference model to determine whether there are potential threat scenarios related to fire, gas leakage, forced intrusion, or noise pollution in the current data. S25: When the calculation result of the threat scenario recognition model shows that the probability of a certain threat scenario exceeds the preset alarm threshold, the cloud server immediately issues a warning signal and records the current analysis result and corresponding data features. If the data does not reach the threat threshold, continue to monitor and update the real-time data input.

4. The intelligent security alarm method based on the Internet of Things according to claim 3, characterized in that, The specific steps of S21 are as follows: S211: The cloud server standardizes the data from multi-modal sensors. S212: Based on the standardized data, a multi-layer neural network model including an input layer and a hidden layer is constructed for preliminary feature extraction. S213: Based on feature extraction, use the Bayesian inference model to estimate the probability of threat scenarios. The formula for Bayes' theorem is: , where is the posterior probability of threat scenario given data . is the probability of data occurring under threat scenario . is the prior probability of threat scenario . is the total probability of data ; S214: When fusing multi-modal data, the weighted average method is used to perform weighted calculations on the data of different sensors. The fusion result formula is: , where is the total output after fusion, is the weight of the -th sensor, is the data of the -th sensor, is the number of sensors.

5. The intelligent security alarm method based on the Internet of Things according to claim 1, characterized in that, The specific steps of S3 are as follows: S31: When the threat scenario recognition model detects a potential threat scenario, the cloud server will transmit the recognition result to each sensor control module through the IoT gateway and generate a control signal containing instructions. The control signal includes preset parameters for adjusting the acquisition frequency and sensitivity of relevant sensors. S32: After receiving the control signal from the cloud server, the IoT gateway will transmit the control signal to the relevant sensors according to the requirements of the threat scenario. Let the frequency adjustment parameter in the control signal be , and the sensitivity adjustment parameter be ; S33: The adjusted sensor starts to work at the new acquisition frequency and sensitivity to capture dynamic data related to threat scenarios. Monitor the fluctuations of sensor data in the short term through time series analysis to ensure the accuracy of the identified threat scenarios. The formula for time series analysis is: , where is the total sum of dynamic data at the monitoring time point , is the acquisition data of the th sensor at the time point , is the number of sensors; S34: The cloud server continuously receives the adjusted sensor data stream and compares it with the threat scenario recognition model to confirm the change trend of the threat scenario. Once the data fluctuation exceeds the preset range, it is confirmed that the potential threat scenario is real.

6. The intelligent security alarm method based on the Internet of Things according to claim 1, characterized in that The specific steps of S4 are as follows: S41: After the threat scenario is confirmed, the cloud server triggers a corresponding multi-level alarm mechanism according to the severity of the threat. Specifically, the threat level is divided into level 1 alarm , level 2 alarm and level 3 alarm . The threat level is calculated by the following formula: , where represents the posterior probability of the threat scenario calculated according to the Bayesian inference model , indicating the probability of the threat scenario occurring, represents the total dynamic fluctuation of the sensor data at time point , and are the adjustment coefficients of the posterior probability and the dynamic data fluctuation respectively;​ S42: According to the calculated threat level , when the threat level meets the following conditions, the cloud server triggers corresponding multi-level alarms; When occurs, the cloud server triggers a first-level alarm; When occurs, the cloud server triggers a secondary alarm; When occurs, the cloud server triggers a level-three alarm.

7. The intelligent security alarm method based on the Internet of Things according to claim 1, characterized in that The specific steps of S5 are as follows: S51: The user communicates with the cloud server via a mobile device. A dedicated application for docking with the cloud server is installed on the mobile device. The application establishes a secure connection with the cloud server through the HTTPS protocol. After the user logs in to the application, the cloud server transmits real-time monitoring data to the mobile device, including the camera video stream in the monitored area and the environmental data of various sensors. S52: The real-time monitoring screen is compressed through video stream encoding technology to reduce bandwidth consumption. After the video data is transmitted to the cloud server through the Internet of Things gateway, it is forwarded in real-time to the user's mobile device. The user can view high-definition monitoring screens within the application and switch different camera perspectives as needed. S53: When the cloud server triggers an alarm, the application will receive an alarm notification. The notification content includes the threat type, occurrence location, threat level, and real-time data of relevant sensors. The user can view the alarm information through the application interface. S54: The user can confirm or cancel the issued alarm information in the application.

8. The intelligent security alarm method based on the Internet of Things according to claim 1, wherein The threat response report generated in S6 includes the basic information of the threat event, the specific response of the sensors, the preliminary analysis results of the threat event, and the operation records of the user during the event. The basic information of the threat event includes the time, location, threat type, and level of the event. The specific response of the sensors includes the time series data charts of various sensors. The preliminary analysis results of the threat event include the duration, fluctuation range, and response of the threat. The operation records of the user during the event include the user's confirmation or cancellation operation of the alarm and the corresponding time points.

9. An intelligent security alarm device based on the Internet of Things is used to implement the intelligent security alarm method based on the Internet of Things as described in any one of claims 1-8, characterized in that, It includes the following modules: Multimodal sensor module: Deployed in the monitored area, it is used to collect multi-dimensional environmental data, including data on temperature, humidity, gas concentration, light intensity, noise level, and vibration frequency. Internet of Things gateway module: It is used to receive real-time data from the multimodal sensor module and connect to the cloud server through an encrypted wireless communication protocol to ensure the security and real-time nature of data transmission. Cloud server module: It is used to receive sensor data from the Internet of Things gateway module and perform real-time analysis on the received data using a threat scenario recognition model. The threat scenario recognition model is built based on a multimodal data fusion algorithm and can identify potential threat scenarios such as fires, gas leaks, and intrusions, and trigger an alarm mechanism according to the recognition results. Alarm module: It includes a local alarm device and a remote alarm device. The local alarm device receives an alarm instruction from the cloud server module through the Internet of Things gateway module and triggers an audible and visual alarm signal. The remote alarm device is used to send alarm information to the user's mobile device through the Internet of Things gateway. User terminal module: By communicating with the cloud server module, it enables the user to access the security alarm device through a mobile device or a web browser, remotely view real-time monitoring screens and sensor data, and confirm or cancel the alarm information issued by the security alarm device.

10. An electronic device, characterized in that, It includes a processor and a storage device, and a computer program is stored on the storage device. When the computer program is run by the processor, it executes the Internet of Things-based intelligent security alarm method according to any one of claims 1-8.

Images