Edge computing control method based on AI abnormal event alarm linkage, gateway system, medium and product

Abstract

The invention discloses an edge computing control method based on AI abnormal event alarm linkage, a gateway system, a medium and a product, and relates to the technical field of intelligent monitoring and edge computing, and the method comprises the steps: obtaining video image data at a current moment and signal data of each Internet of Things sensor at the current moment; judging whether an abnormal event exists or not according to the video picture data at the current moment and the signal data of each Internet of Things sensor at the current moment; if the abnormal event exists, abnormal event information is generated; generating an alarm signal instruction according to the abnormal event information; corresponding alarm equipment in the corresponding alarm area is controlled to give an alarm according to the alarm signal instruction, and multi-channel emergency alarm is achieved; and alarm information including the abnormal event type, the occurrence position and the occurrence time is generated. According to the invention, the detection precision and the response speed of the abnormal event are improved, and the false alarm rate and the missing report rate are effectively reduced.

Description

Technical Field

[0001] The present application relates to the field of intelligent monitoring and edge computing technology, and in particular to an edge computing control method, gateway system, medium and product based on AI abnormal event alarm linkage. Background Art

[0002] With the rapid development of science and technology, intelligent and automated technologies are increasingly used in all walks of life. These technologies not only improve production efficiency, but also greatly improve people's quality of life. Especially in the fields of security monitoring and edge computing, people's demand for real-time monitoring and rapid response to abnormal events is growing.

[0003] With the acceleration of urbanization and the increase of various safety hazards, traditional security monitoring systems can no longer meet current needs. The introduction of edge computing technology enables data to be preliminarily processed and analyzed locally, greatly improving the response speed and accuracy of the system.

[0004] In the relevant security monitoring systems, although a large amount of data can be collected through video images and IoT sensors, there are still deficiencies in the detection and alarm of abnormal events. In particular, when faced with complex and changing environments and diverse abnormal events, existing systems often find it difficult to accurately and promptly identify and respond. For example, when processing video images and sensor data according to traditional algorithms, the accuracy and efficiency are both lacking when detecting certain types of abnormal events. Summary of the invention

[0005] The purpose of this application is to provide a control method, gateway system, medium and product based on AI abnormal event alarm linkage, which can realize accurate, real-time detection and rapid linkage response to abnormal events through intelligent technology.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In a first aspect, the present application provides a control method based on AI abnormal event alarm linkage, comprising:

[0008] Obtain the current video image data and the current signal data of each IoT sensor;

[0009] Determine whether there is a video abnormality event based on the video image data at the current moment, and if there is a video abnormality event, generate video abnormality event information;

[0010] Determine whether there is a sensor abnormality event based on the signal data of each IoT sensor at the current moment; if there is a sensor abnormality event, generate sensor abnormality event information;

[0011] Generate an alarm signal instruction according to the video abnormal event information and the sensor abnormal event information;

[0012] According to the alarm signal instruction, the corresponding alarm device in the corresponding alarm area is controlled to alarm; and alarm information including the abnormal event type, occurrence location and occurrence time is generated.

[0013] Optionally, judging whether there is a video abnormality event based on the video picture data at the current moment, and if there is a video abnormality event, generating video abnormality event information, specifically including:

[0014] Identify the target object in the video image data at the current moment and determine the type, location and action of the target object;

[0015] Determine whether the category of the target object is a preset category; if it is a preset category, the position, action, category and first abnormal time of the target object constitute the target object information;

[0016] Determine whether there is a video abnormality event based on the target object information; if there is a video abnormality event, generate video abnormality event information including the abnormal video type and the target object information.

[0017] Optionally, it is determined whether there is a sensor abnormality event according to the signal data of each IoT sensor at the current moment. If there is a sensor abnormality event, the sensor abnormality event information is generated, specifically including:

[0018] According to the signal data of each IoT sensor at the current moment, the materialized indicator data corresponding to the signal data is obtained;

[0019] Compare each physical and chemical indicator data with the respective preset indicators to obtain multiple comparison results; if at least one of each comparison result reaches or exceeds the preset alarm value, an abnormal IoT data event occurs;

[0020] The generated comparison result is the corresponding sensor abnormal event information; the sensor abnormal event information includes: abnormal sensor type, physical and chemical indicator data, second abnormal time and abnormal sensor position.

[0021] Optionally, controlling a corresponding alarm device in a corresponding alarm area to sound an alarm according to the alarm signal instruction specifically includes:

[0022] Control the smart power switch to turn on or off;

[0023] Control the opening or closing of the sound and light alarm;

[0024] Control the camera to record videos and capture pictures, and send the recorded videos and captured pictures to the user in real time.

[0025] Optionally, the alarm information further includes: voice broadcast;

[0026] Voice broadcast includes broadcasting of the user's pre-recorded voice or the voice automatically converted from the user's input text; it also includes specific voice broadcast generated based on the alarm information.

[0027] In the second aspect, the present application provides an edge computing gateway system based on AI abnormal event alarm linkage, including:

[0028] Video screen access module, used to obtain the video screen data at the current moment;

[0029] IoT sensor access module, used to obtain the signal data of each IoT sensor at the current moment;

[0030] The video abnormal event detection module is used to determine whether there is a video abnormal event based on the video image data at the current moment, and if there is a video abnormal event, generate video abnormal event information;

[0031] The sensor abnormal event detection module is used to determine whether there is a sensor abnormal event based on the signal data of each IoT sensor at the current moment; if there is a sensor abnormal event, the sensor abnormal event information is generated;

[0032] An alarm signal instruction generation module is used to generate an alarm signal instruction according to the video abnormal event information and the sensor abnormal event information;

[0033] The alarm linkage module is used to control the corresponding alarm device in the corresponding alarm area to alarm according to the alarm signal instruction, and generate alarm information including the type of abnormal event, the location and the time of occurrence.

[0034] Optionally, the video abnormal event detection module specifically includes: a video decoding and frame extraction unit, an AI reasoning algorithm application unit, a video abnormal event determination unit, and an alarm information generation and sending unit;

[0035] A video decoding and frame extraction unit, connected to the AI ​​reasoning algorithm application unit, is used to decompose the video picture data into a continuous multi-frame static picture sequence upon receiving a video processing instruction from the user, and send the static picture sequence to the AI ​​reasoning algorithm application unit;

[0036] An AI reasoning algorithm application unit, connected to the abnormal event determination unit, is used to identify a target object in a static image sequence, generate target object information, and send the target object information to the abnormal event determination unit;

[0037] An abnormal event determination unit is connected to the alarm information generation and sending unit, and is used to determine whether there is a video abnormal event according to whether the action of the target object in the target object information reaches a preset action matching threshold, and generate video abnormal event information and send it to the alarm information generation and sending unit;

[0038] The video decoding and frame extraction unit is also connected to the alarm linkage module and is used to send a static picture containing video abnormal event information to the alarm linkage module;

[0039] The alarm information generating and sending unit is connected to the alarm linkage module, and is used to generate alarm information according to the video abnormal event information and the corresponding static image, and send it to the alarm linkage module.

[0040] Optionally, the AI ​​reasoning algorithm application unit specifically includes: a target detection subunit, a target classification subunit, a target tracking subunit, and a picture OCR subunit;

[0041] The target detection subunit is connected to the target classification subunit, and is used to identify the target object in the static picture and send the target object to the target classification subunit;

[0042] The target classification subunit is used to classify the target object, determine the category, position and action of the target object; and is also used to detect the specific category attributes of the target object according to the category of the target object, obtain the attribute information of the target object, and send the attribute information to the target tracking subunit;

[0043] The target tracking subunit is used to track the target object according to the attribute information of the target object to obtain the moving trajectory of the target object; and determine the moving speed and direction of the target object according to the moving trajectory of the target object;

[0044] The image OCR subunit is used to extract text information from images.

[0045] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-mentioned control methods based on AI abnormal event alarm linkage.

[0046] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned control methods based on AI abnormal event alarm linkage.

[0047] According to the specific embodiments provided in this application, this application has the following technical effects:

[0048] The present application provides an edge computing control method, gateway system, medium and product based on AI abnormal event alarm linkage. By acquiring the video picture data and signal data of the IoT sensor at the current moment in real time, it can accurately determine whether there are abnormal events in the video and sensor, and generate corresponding abnormal event information accordingly. It solves the problem of insufficient detection of complex and changeable environments and diversified abnormal events in traditional security monitoring systems, and realizes rapid identification and response of abnormal events. At the same time, the method further generates an alarm signal instruction based on the generated abnormal event information, and controls the alarm equipment in the corresponding alarm area to alarm, and generates detailed alarm information including the type of abnormal event, location and time of occurrence. This linkage mechanism solves the shortcomings of the existing system in abnormal event alarm linkage, realizes instant alarm and precise positioning of abnormal events, and greatly improves the intelligence level and response speed of the security monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A flowchart of an edge computing control method based on AI abnormal event alarm linkage provided in one embodiment of the present application;

[0051] Figure 2 for Figure 1 A detailed flowchart of the abnormal event detection steps in FIG.

[0052] Figure 3 for Figure 1 Detailed flow chart of the alarm linkage steps;

[0053] Figure 4 for Figure 1 A detailed flow chart of the linkage devices in the alarm linkage step;

[0054] Figure 5 for Figure 1 A detailed flow chart of the voice broadcast in the alarm linkage step;

[0055] Figure 6 A functional module diagram of an edge computing gateway system based on AI abnormal event alarm linkage provided in one embodiment of the present application;

[0056] Figure 7 A schematic diagram of a detailed functional module of a video abnormal event detection module provided in an embodiment of the present application;

[0057] Figure 8 A schematic diagram of functional modules of an edge intelligent device command control system provided in one embodiment of the present application;

[0058] Fig. 9 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0060] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0061] In an exemplary embodiment, Figure 1 As shown, an edge computing control method based on AI abnormal event alarm linkage is provided, including the following steps 101 to 105. Among them:

[0062] Step 101, obtaining the video picture data at the current moment and the signal data of each IoT sensor at the current moment.

[0063] Step 102, judging whether there is a video abnormality event based on the video image data at the current moment, and if there is a video abnormality event, generating video abnormality event information.

[0064] Step 103, judging whether there is a sensor abnormality event according to the signal data of each IoT sensor at the current moment; if there is a sensor abnormality event, generating sensor abnormality event information.

[0065] Step 104: Generate an alarm signal instruction according to the video abnormal event information and the sensor abnormal event information.

[0066] Step 105, according to the alarm signal instruction, the corresponding alarm device in the corresponding alarm area is controlled to sound an alarm; and alarm information including the abnormal event type, occurrence location and occurrence time is generated.

[0067] Implementing the above steps 101 to 105 can significantly improve the detection efficiency and response speed of abnormal events, and reduce the losses and risks caused by abnormal events. In addition, the method is highly flexible and scalable, and can be customized and optimized according to actual needs to meet the needs of abnormal detection and alarm linkage in different scenarios.

[0068] In another exemplary embodiment of the present application, Figure 2 As shown, steps 101 to 103 can also receive video images and IoT sensor signals through the video image access module and the IoT sensor access module; the abnormal event detection module has a built-in abnormal event recognition model, and the recognition model includes two categories: video image AI detection and IoT data indicator abnormality detection. When it is detected that the received video image or IoT sensor signal matches the abnormal event recognition model, it is determined to be an abnormal event.

[0069] In another exemplary embodiment of the present application, Figure 3 As shown in the figure, when an abnormal event is detected, the alarm linkage rules pre-defined in the alarm linkage module will be automatically triggered. This design is to ensure that the system can respond quickly and accurately when an abnormal situation is detected.

[0070] Once the alarm linkage rule is triggered, the system will first construct the corresponding alarm signal instruction according to the specific content of the current alarm information, including the alarm category (such as intrusion detection, equipment failure, etc.), occurrence time, related screenshots or videos, etc., to ensure that the alarm information can be accurately conveyed to the relevant processing units. Then, according to the preset rules, the alarm signal is output using appropriate protocols and interfaces. This covers the use of specific communication protocols (to control the switch of sound and light alarm equipment, so as to quickly attract the attention of personnel in an emergency. At the same time, the system can also send alarm event notifications to other systems or platforms through interface technology to achieve cross-system information sharing and collaborative processing.

[0071] In addition, the system also has the ability to control the switching value of sound and light alarms through protocol output, and send alarm data such as sound clips, text content, images, etc. through the interface, so that the receiving end can have a more comprehensive understanding of the situation of abnormal events, and thus make more accurate judgments and decisions.

[0072] In another exemplary embodiment of the present application, step 102 specifically includes:

[0073] Identify the target object in the video image data at the current moment and determine the type, location and action of the target object.

[0074] It is determined whether the category of the target object is a preset category; if it is a preset category, the position, action, category and first abnormal time of the target object constitute the target object information.

[0075] Determine whether there is a video abnormality event based on the target object information; if there is a video abnormality event, generate video abnormality event information including the abnormal video type and the target object information.

[0076] In another exemplary embodiment of the present application, step 103 specifically includes:

[0077] According to the signal data of each IoT sensor at the current moment, the physical and chemical index data corresponding to the signal data are obtained; the physical and chemical index data include excessive concentration of harmful gases, excessive temperature, excessive humidity, excessive smoke, excessive PM2.5, etc.

[0078] Compare each physical and chemical indicator data with their respective preset indicators to obtain multiple comparison results; if at least one of each comparison result reaches or exceeds the preset alarm value, there is an abnormal IoT data event; specifically, physical and chemical indicators are compared in the form of quantitative comparison, trend comparison, etc., and exceeding a certain numerical range is a trigger. According to a certain accuracy rate and spatial area and time range, if the AI ​​alarm event is within or outside this range, it is considered to trigger an alarm.

[0079] The generated comparison result is the corresponding sensor abnormal event information; the sensor abnormal event information includes: abnormal sensor type, physical and chemical indicator data, second abnormal time and abnormal sensor position.

[0080] After the edge computing gateway device detects an abnormal event, it can trigger the IoT control device, flash device, voice broadcast and event screen push.

[0081] In another exemplary embodiment of the present application, Figure 4 As shown, according to the alarm signal instruction, the corresponding alarm device in the corresponding alarm area is controlled to alarm, specifically including:

[0082] Controls the smart power switch to be on or off.

[0083] Control the opening or closing of the sound and light alarm.

[0084] Control the camera to record videos and capture pictures, and send the recorded videos and captured pictures to the user in real time.

[0085] In another exemplary embodiment of the present application, Figure 5 As shown, the alarm information also includes: voice broadcast.

[0086] Voice broadcast includes broadcasting of the user's pre-recorded voice or the voice automatically converted from the user's input text; it also includes specific voice broadcast generated based on the alarm information.

[0087] The present application also provides an application scenario, which applies the above-mentioned control method based on AI abnormal event alarm linkage. Specifically: The control method based on AI abnormal event alarm linkage provided in this embodiment can be applied in the security monitoring scenario of smart cities. The security monitoring scenario of smart cities includes data collection links, abnormal detection and alarm linkage response links, and alarm response and disposal links. Various types of data in urban operation are captured in real time through various monitoring devices, and these data are transmitted to the data center through the communication network for storage and preprocessing. Through the abnormal detection and alarm linkage response link, the abnormal time is identified, the abnormal information is promptly conveyed to the relevant departments and personnel, and the alarm response and disposal link is entered to take corresponding measures to respond to and dispose of the abnormal events. Specifically, it can be judged whether there is a video abnormal event based on the video picture data at the current moment, such as abnormal gathering of people, illegal driving of vehicles, etc.; at the same time, it will also be judged whether there is an environmental abnormal event based on the signal data of the environmental monitoring sensor, such as excessive air quality, fire hazards, etc. Once an abnormal event is detected, the system will immediately generate an alarm signal instruction and extract detailed information about the abnormal event, such as the event type, location, and time of occurrence, which not only provides a key basis for subsequent alarm response and disposal, but also triggers the Internet of Things control equipment, flash equipment, voice broadcast system, etc. through the linkage mechanism, achieving instant response and warning to abnormal events. At the same time, the system will also push the screen of the abnormal event to relevant departments and personnel in real time so that they can quickly understand the on-site situation and make decisions.

[0088] Based on the same inventive concept, the embodiment of the present application also provides an edge computing gateway system for implementing the above-mentioned AI abnormal event alarm linkage. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more edge computing gateway system embodiments based on AI abnormal event alarm linkage provided below can be found in the above limitations on the edge computing control method based on AI abnormal event alarm linkage, which will not be repeated here.

[0089] In an exemplary embodiment, Figure 6 As shown, an edge computing gateway system based on AI abnormal event alarm linkage is provided, including:

[0090] The video picture access module 201 is used to obtain the video picture data at the current moment.

[0091] The IoT sensor access module 202 is used to obtain the signal data of each IoT sensor at the current moment.

[0092] The video abnormal event detection module 203 is used to determine whether there is a video abnormal event based on the video image data at the current moment, and if there is a video abnormal event, generate video abnormal event information.

[0093] The sensor abnormal event detection module 204 is used to determine whether there is a sensor abnormal event based on the signal data of each IoT sensor at the current moment; if there is a sensor abnormal event, generate sensor abnormal event information.

[0094] The alarm signal instruction generating module 205 is used to generate an alarm signal instruction according to the video abnormal event information and the sensor abnormal event information.

[0095] The alarm linkage module 206 is used to control the corresponding alarm device in the corresponding alarm area to alarm according to the alarm signal instruction, and generate alarm information including the abnormal event type, occurrence location and occurrence time.

[0096] As an optional implementation, the video abnormal event detection module specifically includes: a video decoding and frame extraction unit, an AI reasoning algorithm application unit, a video abnormal event determination unit and an alarm information generation and sending unit.

[0097] The video decoding and frame extraction unit is connected to the AI ​​reasoning algorithm application unit and is used to decompose the video picture data into a continuous multi-frame static image sequence when receiving a user's video processing instruction, and send the static image sequence to the AI ​​reasoning algorithm application unit.

[0098] The AI ​​reasoning algorithm application unit is connected to the abnormal event determination unit, and is used to identify the target object in the static image sequence, generate the target object information, and send the target object information to the abnormal event determination unit.

[0099] The abnormal event determination unit is connected to the alarm information generation and sending unit, and is used to determine whether there is a video abnormal event based on whether the action of the target object in the target object information reaches a preset action matching threshold, and generate video abnormal event information and send it to the alarm information generation and sending unit.

[0100] The video decoding and frame extraction unit is also connected to the alarm linkage module and is used to send a static picture containing video abnormal event information to the alarm linkage module.

[0101] The alarm information generating and sending unit is connected to the alarm linkage module, and is used to generate alarm information according to the video abnormal event information and the corresponding static image, and send it to the alarm linkage module.

[0102] Among them, in implementing this implementation method, the AI ​​reasoning algorithm application unit specifically includes: a target detection subunit, a target classification subunit, a target tracking subunit, and a picture OCR subunit.

[0103] The target detection subunit is connected to the target classification subunit, and is used to identify the target object in the static picture and send the target object to the target classification subunit.

[0104] The target classification subunit is used to classify the target object and determine the category, position and action of the target object; it is also used to detect the specific category attributes of the target object according to the category of the target object, obtain the attribute information of the target object, and send the attribute information to the target tracking subunit.

[0105] The target tracking subunit is used to track the target object according to the attribute information of the target object to obtain the moving trajectory of the target object; and determine the moving speed and direction of the target object according to the moving trajectory of the target object.

[0106] The image OCR subunit is used to extract text information from images.

[0107] As an optional implementation, Figure 7 As shown in the figure, the video surveillance platform integrates multiple key modules that work together to achieve efficient video surveillance and intelligent analysis. Among them, the surveillance camera, as the front-end device, is responsible for capturing real-time video streams and transmitting these video data to the video surveillance platform through video protocols / standards. Inside the video surveillance platform, the video access module first receives these video streams and sends them to the video decoding module. The video decoding module (relying on hardware decoding computing power) has powerful processing capabilities and can decompose continuous video streams into a series of clear and coherent multi-frame static image sequences. These image sequences are then sent to the AI ​​reasoning algorithm model module, which embeds multiple parallel and partially interdependent AI algorithm models, including but not limited to target detection models, target classification models, target tracking models, and image OCR models. The target detection model focuses on identifying key targets in video frames, such as human bodies, vehicles, and flames. The target classification model further classifies the attributes of the detected targets in detail, such as judging the clothing characteristics of the human body, the items carried, and the model and color of the vehicle, providing accurate information for subsequent tracking and disposal. The target tracking model is responsible for continuously tracking the detected target and obtaining key information such as its movement trajectory, speed and direction in real time to ensure dynamic monitoring of security events without omissions. The image OCR model has the ability to extract text from images, such as road signs, license plate numbers, etc., to provide data support for text analysis.

[0108] When a model in the AI ​​reasoning algorithm model (such as a target detection model) detects a specific target or target action and finds that its match with the preset model reaches the set threshold, a detection event will be triggered immediately. This event is then sent to the AI ​​algorithm comprehensive application module (i.e., the event alarm distribution module), which will conduct a comprehensive analysis based on time, area and other rules to accurately determine whether the event is an abnormal event, such as human intrusion, vehicle intrusion or animal intrusion. Once confirmed as an abnormal event, the AI ​​abnormal event alarm and disposal interaction module will respond quickly, establish a connection with the application system, and automatically trigger the corresponding alarm and disposal process. At the same time, the video encoding module (relying on hardware encoding computing power) will send real-time video (video stream) to the AI ​​abnormal event alarm and disposal interaction module, so as to intuitively display the on-site situation of the abnormal event on the interactive interface, and provide intuitive visual reference for relevant personnel.

[0109] In addition, in order to improve the accuracy and reliability of the system, the system will also send key frames or screenshots to the AI ​​abnormal event alarm and handling interaction module. These screenshots are not only used for UI interface display, but also provide important data support for the subsequent review of possible false alarms. Through this mechanism, the system can form a closed loop that can be retrained to continuously optimize and improve its own performance.

[0110] In addition, it also includes a list database, which is specifically used to store comparison information related to face recognition. When the system detects a face, it will immediately compare it with the information in the list database to quickly determine whether it is a known person or a potential suspicious person.

[0111] Finally, the image processing module performs enhancement operations on the video frames. Through a series of processes such as noise reduction, defogging, and exposure correction, this module can significantly improve the video quality and provide clearer and more accurate video data for subsequent AI algorithm analysis and processing. These processing processes may rely on hardware AI reasoning computing power and hardware image processing computing power to ensure processing efficiency and accuracy.

[0112] In summary, this optimized implementation method achieves efficient processing and intelligent analysis of video data by integrating multiple key modules and advanced technologies of the video surveillance platform.

[0113] As an optional implementation, the present application also includes an edge intelligent device command control system, where the edge intelligent device accesses the command platform through an encrypted and secure communication channel, ensuring efficient and stable communication between the edge device and the command platform. First, the command channel is not subject to conventional firewall restrictions, has strong penetration capabilities, can easily bypass firewall restrictions, and achieve smooth communication between edge devices and the command platform. Regardless of the network environment in which the edge device is located, the command channel can ensure the timely transmission of information. Secondly, the command channel supports two-way command sending and receiving. This means that the command platform can not only send commands to edge devices, but also receive status reports, data uploads and other information from edge devices. This two-way communication mechanism provides great convenience for remote management and control of edge devices. In addition, the command platform also has command group sending and auditing functions. It can send commands to multiple edge devices at the same time, greatly improving management efficiency. At the same time, the system also records the sending, receiving and processing of each command, providing strong support for subsequent tracking and analysis.

[0114] In the specific communication process, such as Figure 8 As shown in the figure, the edge device first sends a request to the command platform to establish a command control channel. After verifying the identity and authority of the edge device, the command platform will establish an encrypted command control channel. Once the channel is successfully established, the command platform can start sending commands to the lower-level edge devices. After receiving the command, the edge device will perform the corresponding operation according to the command content, and upload the processing results, status information and other report information to the upper-level device or command platform through the command control channel.

[0115] Throughout the communication process, the device authentication mechanism always plays an important role. It ensures that only legitimate devices can access the command platform, establish command control channels, and receive and execute commands, thus greatly improving the security and reliability of the system.

[0116] In summary, the edge intelligent device command control system realizes two-way communication, command group sending, auditing and other functions between edge devices and command platforms by building a safe and efficient command channel. This provides strong support for the remote management and control of edge intelligent devices and promotes the rapid development and application of edge intelligent technology.

[0117] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Fig. 9As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store video images and sensor data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an edge computing control method based on AI abnormal event alarm linkage is implemented.

[0118] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0119] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0120] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0121] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0122] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0123] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0124] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0125] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A control method based on AI abnormal event alarm linkage, characterized in that: The control method based on AI abnormal event alarm linkage includes: Obtain the current video image data and the current signal data of each IoT sensor; Determine whether there is a video abnormality event based on the video image data at the current moment, and if there is a video abnormality event, generate video abnormality event information; Determine whether there is a sensor abnormality event based on the signal data of each IoT sensor at the current moment; if there is a sensor abnormality event, generate sensor abnormality event information; Generate an alarm signal instruction according to the video abnormal event information and the sensor abnormal event information; According to the alarm signal instruction, the corresponding alarm device in the corresponding alarm area is controlled to alarm; and alarm information including the abnormal event type, occurrence location and occurrence time is generated.

2. The edge computing gateway system based on AI abnormal event alarm linkage according to claim 1 is characterized in that: Determine whether there is a video abnormality event based on the video image data at the current moment. If there is a video abnormality event, generate video abnormality event information, including: Identify the target object in the video image data at the current moment and determine the type, location and action of the target object; Determine whether the category of the target object is a preset category; if it is a preset category, the position, action, category and first abnormal time of the target object constitute the target object information; Determine whether there is a video abnormality event based on the target object information; if there is a video abnormality event, generate video abnormality event information including the abnormal video type and the target object information.

3. The edge computing gateway system based on AI abnormal event alarm linkage according to claim 1 is characterized in that: According to the signal data of each IoT sensor at the current moment, it is determined whether there is a sensor abnormal event. If there is a sensor abnormal event, the sensor abnormal event information is generated, including: According to the signal data of each IoT sensor at the current moment, the materialized indicator data corresponding to the signal data is obtained; Compare each physical and chemical indicator data with the respective preset indicators to obtain multiple comparison results; if at least one of each comparison result reaches or exceeds the preset alarm value, an abnormal IoT data event occurs; The generated comparison result is the corresponding sensor abnormal event information; the sensor abnormal event information includes: abnormal sensor type, physical and chemical indicator data, second abnormal time and abnormal sensor position.

4. The edge computing gateway system based on AI abnormal event alarm linkage according to claim 1 is characterized in that: According to the alarm signal instruction, the corresponding alarm device in the corresponding alarm area is controlled to alarm, specifically including: Control the smart power switch to turn on or off; Control the opening or closing of the sound and light alarm; Control the camera to record videos and capture pictures, and send the recorded videos and captured pictures to the user in real time.

5. The edge computing gateway system based on AI abnormal event alarm linkage according to claim 1 is characterized in that: The alarm information also includes: voice broadcast; Voice broadcast includes broadcasting of the user's pre-recorded voice or the voice automatically converted from the user's input text; it also includes specific voice broadcast generated based on the alarm information.

6. An edge computing gateway system based on AI abnormal event alarm linkage, characterized in that: The edge computing gateway system based on AI abnormal event alarm linkage includes: Video screen access module, used to obtain the video screen data at the current moment; IoT sensor access module, used to obtain the signal data of each IoT sensor at the current moment; The video abnormal event detection module is used to determine whether there is a video abnormal event based on the video image data at the current moment, and if there is a video abnormal event, generate video abnormal event information; The sensor abnormal event detection module is used to determine whether there is a sensor abnormal event based on the signal data of each IoT sensor at the current moment; if there is a sensor abnormal event, the sensor abnormal event information is generated; An alarm signal instruction generation module is used to generate an alarm signal instruction according to the video abnormal event information and the sensor abnormal event information; The alarm linkage module is used to control the corresponding alarm device in the corresponding alarm area to alarm according to the alarm signal instruction, and generate alarm information including the type of abnormal event, the location and the time of occurrence.

7. The edge computing gateway system based on AI abnormal event alarm linkage according to claim 6 is characterized in that: The video abnormal event detection module specifically includes: a video decoding and frame extraction unit, an AI reasoning algorithm application unit, a video abnormal event determination unit, and an alarm information generation and sending unit; A video decoding and frame extraction unit, connected to the AI ​​reasoning algorithm application unit, is used to decompose the video picture data into a continuous multi-frame static picture sequence upon receiving a video processing instruction from the user, and send the static picture sequence to the AI ​​reasoning algorithm application unit; An AI reasoning algorithm application unit, connected to the abnormal event determination unit, is used to identify a target object in a static image sequence, generate target object information, and send the target object information to the abnormal event determination unit; An abnormal event determination unit is connected to the alarm information generation and sending unit, and is used to determine whether there is a video abnormal event according to whether the action of the target object in the target object information reaches a preset action matching threshold, and generate video abnormal event information and send it to the alarm information generation and sending unit; The video decoding and frame extraction unit is also connected to the alarm linkage module and is used to send a static picture containing video abnormal event information to the alarm linkage module; The alarm information generating and sending unit is connected to the alarm linkage module, and is used to generate alarm information according to the video abnormal event information and the corresponding static image, and send it to the alarm linkage module.

8. The edge computing gateway system based on AI abnormal event alarm linkage according to claim 7 is characterized in that: AI reasoning algorithm application unit, specifically including: target detection subunit, target classification subunit, target tracking subunit, and image OCR subunit; The target detection subunit is connected to the target classification subunit, and is used to identify the target object in the static picture and send the target object to the target classification subunit; The target classification subunit is used to classify the target object, determine the category, position and action of the target object; and is also used to detect the specific category attributes of the target object according to the category of the target object, obtain the attribute information of the target object, and send the attribute information to the target tracking subunit; The target tracking subunit is used to track the target object according to the attribute information of the target object to obtain the moving trajectory of the target object; and determine the moving speed and direction of the target object according to the moving trajectory of the target object; The image OCR subunit is used to extract text information from images.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the control method based on AI abnormal event alarm linkage described in any one of claims 1 to 5 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the control method based on AI abnormal event alarm linkage described in any one of claims 1 to 5 is implemented.

Images