Inspection method and system based on AI, storage medium and electronic equipment

By performing AI abnormality detection on the Grafana monitoring board, the problem of easy missed abnormalities in manual inspection is solved, the inspection efficiency and accuracy are improved, human resources are saved, and abnormalities are detected and handled in a timely manner.

CN119007102APending Publication Date: 2024-11-22SHENZHEN YISHIHUOLALA TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411021291.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

In the prior art, the inspection method of the Grafana monitoring board manually checking by the on-duty personnel is prone to miss abnormal information, and the labor consumption is large, resulting in failures that may be ignored.

Method used

Using AI-based inspection methods, we use screenshots of Grafana monitoring boards and use target detection models to perform abnormal detection, generate alarm information, and execute preset alarm processes to avoid omissions of abnormalities and save manpower and material resources.

Benefits of technology

It improves the work efficiency of inspection personnel and the accuracy of inspection results, realizes timely detection of problems and intervenes in handling them early, reducing the risk of failures.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119007102A_ABST
    Figure CN119007102A_ABST
Patent Text Reader

Abstract

The invention provides an AI-based inspection method and system, a storage medium and electronic equipment, and the method comprises the steps: carrying out the screenshot of a monitoring billboard indicated by an inspection task when any preset inspection task is executed, and obtaining a billboard image; inputting the billboard image into a target detection model, and detecting the billboard image by using a target detection technology to obtain a detection result output by the target detection model; the target detection model is obtained by training an abnormal billboard image containing abnormal information in advance; and if the abnormal information exists in the detection result, generating alarm information based on the abnormal information, and executing a preset alarm process. According to the scheme, the monitoring billboard is subjected to screenshot to obtain the billboard image, and the AI model with the target detection function is utilized to perform target detection on the billboard image to obtain the detection result, so that the working efficiency of inspection personnel and the accuracy of the inspection result can be greatly improved, and the purposes of avoiding omission and abnormity and saving manpower and material resources are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an AI-based inspection method and system, a storage medium, and an electronic device. Background Art

[0002] Grafana is a powerful and easy-to-use open source indicator analysis and monitoring visualization platform. It can query indicators from time series databases, display key indicators in the form of charts, and issue alarms, which is suitable for building monitoring systems. It provides powerful and beautiful charts, rich dashboard customization configurations, and its functions can be infinitely expanded through plug-ins. At present, most Internet companies require on-duty personnel to complete daily inspection tasks on the "Grafana-like monitoring dashboard" on time during daily duty hours.

[0003] In the prior art, the on-duty personnel of each service on that day manually check the monitoring indicators of all relevant system services on the Grafana monitoring dashboard at fixed time nodes, and complete the inspection based on their personal experience to judge the health status of the system services.

[0004] Due to the different experiences of the on-duty personnel, abnormal information is easily missed, which eventually leads to failures, and the manual inspection method consumes a lot of manpower. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides an AI-based inspection method and system, a storage medium, and an electronic device to avoid missing abnormalities and save manpower and material resources.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] A first aspect of an embodiment of the present invention discloses an AI-based inspection method, the method comprising:

[0008] When any preset inspection task is executed, a screenshot is taken of the monitoring board indicated by the inspection task to obtain a board image;

[0009] Inputting the billboard image into a target detection model and performing detection using a target detection technology to obtain a detection result output by the target detection model; the target detection model is pre-trained using an abnormal billboard image containing abnormal information;

[0010] If the abnormal information exists in the detection result, alarm information is generated based on the abnormal information, and a preset alarm process is executed.

[0011] Preferably, the training process of the anomaly detection model includes:

[0012] Acquire multiple abnormal billboard images containing abnormal information;

[0013] Annotate each of the abnormal billboard images to obtain a corresponding annotation file containing annotation information; the annotation information includes: abnormality type and coordinate information;

[0014] Based on each of the abnormal billboard images and each of the annotation files, construct a training set;

[0015] Create a target detection model to be trained;

[0016] Based on the training set and a pre-written training script, the target detection model to be trained is trained to obtain a trained target detection model.

[0017] Preferably, if the abnormal information exists in the detection result, generating alarm information based on the abnormal information and executing a preset alarm process include:

[0018] If the abnormal information exists in the detection result, the abnormal information is marked in the billboard image to obtain an alarm image;

[0019] Alarm information is generated based on the abnormal information and the alarm image, and a preset alarm process is executed.

[0020] Preferably, the execution of the preset alarm process includes:

[0021] Obtain the push interface of the communication tool set in the alarm strategy of the inspection task configuration; each of the push interfaces corresponds to a duty officer;

[0022] The warning information is sent through the push interface.

[0023] Preferably, the method further comprises:

[0024] Adding a life cycle status mark of an initial state to the alarm information; the life cycle status mark is used to indicate the processing progress of the alarm information;

[0025] In response to a follow-up instruction input by an on-duty personnel, the life cycle status mark is modified based on the follow-up instruction.

[0026] Preferably, the method further comprises:

[0027] Poll the preset configuration information database to obtain the inspection configuration information corresponding to the inspection task;

[0028] Acquire a cron expression from the inspection configuration information; the cron expression is used to represent the execution time of the inspection task;

[0029] The inspection task corresponding to the inspection configuration information is executed based on the cron expression.

[0030] Preferably, after obtaining the detection result output by the target detection model, the method further includes:

[0031] The detection results are stored in a preset detection result database.

[0032] A second aspect of an embodiment of the present invention discloses an AI-based inspection system, the system comprising:

[0033] A screenshot module is used to take a screenshot of the monitoring dashboard indicated by any preset inspection task when the inspection task is executed, so as to obtain a dashboard image;

[0034] An AI detection module is used to input the billboard image into a target detection model and perform detection using target detection technology to obtain a detection result output by the target detection model; the target detection model is pre-trained using an abnormal billboard image containing abnormal information;

[0035] The alarm module is used to generate alarm information based on the abnormal information if the abnormal information exists in the detection result, and execute a preset alarm process.

[0036] A third aspect of an embodiment of the present invention discloses a storage medium for storing a computer program. When the computer program is executed, it is specifically used to implement any AI-based inspection method described in the first aspect of the embodiment of the present invention.

[0037] A fourth aspect of an embodiment of the present invention discloses an electronic device, including: a memory and a processor;

[0038] The memory is used to store computer programs;

[0039] The processor is used to execute the computer program, specifically to implement the AI-based inspection method described in any one of the first aspects of the embodiments of the present invention.

[0040] Based on the above-mentioned embodiments of the present invention, an AI-based inspection method and system, storage medium, and electronic device are provided. When any preset inspection task is executed, a screenshot of the monitoring board indicated by the inspection task is taken to obtain a board image; the board image is input into the target detection model and detected using target detection technology to obtain a detection result output by the target detection model; the target detection model is pre-trained using an abnormal board image containing abnormal information; if the abnormal information exists in the detection result, an alarm message is generated based on the abnormal information, and a preset alarm process is executed. In this solution, a screenshot of the monitoring board is taken to obtain a board image, and an AI model with target detection function is used to perform target detection on the board image to obtain a detection result, which can greatly improve the work efficiency of the inspection personnel and the accuracy of the inspection results, thereby avoiding missing abnormalities and saving manpower and material resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0042] Figure 1 A flowchart of an AI-based inspection method disclosed in an embodiment of the present invention;

[0043] Figure 2 A flowchart of a trigger inspection task disclosed in an embodiment of the present invention;

[0044] Figure 3 A flowchart of a target detection model training disclosed in an embodiment of the present invention;

[0045] Figure 4 A structural diagram of an AI-based inspection system disclosed in an embodiment of the present invention;

[0046] Figure 5 The present invention is a structural diagram of an electronic device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0049] As can be seen from the background technology, in the prior art, the on-duty personnel of each service on that day manually check the monitoring indicators of all relevant system services on the Grafana monitoring dashboard at a fixed time node, and complete the inspection based on their personal experience to judge the health status of the system services. Due to the different experience of the on-duty personnel, it is easy to miss abnormal information and eventually cause failures, and the manual inspection method consumes a lot of manpower.

[0050] Therefore, the embodiments of the present invention disclose an AI-based inspection method and system, storage medium, and electronic device. In this solution, a screenshot of the monitoring board is taken to obtain a board image, and an AI model with a target detection function is used to perform target detection on the board image to obtain a detection result, which can greatly improve the work efficiency of the inspection personnel and the accuracy of the inspection results, thereby avoiding missing abnormalities and saving manpower and material resources.

[0051] like Figure 1 As shown, it is a flow chart of an AI-based inspection method disclosed in an embodiment of the present invention, which includes the following steps:

[0052] Step S101: when any preset inspection task is executed, a screenshot is taken of the monitoring board indicated by the inspection task to obtain a board image.

[0053] In step S101, each monitoring dashboard is used to display multiple indicator information, and each inspection task is responsible for inspecting the indicator information displayed on a certain monitoring dashboard. The monitoring dashboard can be a Grafana monitoring dashboard.

[0054] like Figure 2 FIG. 1 is a flowchart of a trigger inspection task disclosed in an embodiment of the present invention, comprising the following steps:

[0055] Step S11: polling a preset configuration information database to obtain inspection configuration information corresponding to the inspection task.

[0056] It should be noted that the configuration information includes the appid that needs to be inspected, indicators or monitoring dashboards, cron expressions, exception categories of concern, alarm configuration (whether to alarm, alarm recipients), and other information.

[0057] Step S12: Obtain a cron expression from the inspection configuration information.

[0058] The cron expression is used to represent the execution time of the inspection task.

[0059] A corn expression is a string that is used to define a schedule for executing inspection tasks at a specific time or set of times.

[0060] Step S13: triggering the execution of the inspection task corresponding to the inspection configuration information based on the cron expression.

[0061] It can be understood that when the current time is consistent with the execution time of the inspection task represented by the cron expression, the inspection task is executed.

[0062] Step S102: input the billboard image into the target detection model and perform detection using the target detection technology to obtain the detection result output by the target detection model.

[0063] In step S102, object detection technology is an important branch of computer vision, which aims to identify the location and category of specific objects in images or videos. Common object detection technologies include deep learning methods.

[0064] The target detection model is pre-trained using abnormal billboard images containing abnormal information.

[0065] like Figure 3 FIG. 1 is a flowchart of a target detection model training disclosed in an embodiment of the present invention. The specific training process includes the following steps:

[0066] Step S21: Acquire a plurality of abnormal billboard images containing abnormal information.

[0067] In step S21, an abnormal billboard image containing abnormal information of the monitoring billboard is collected, and the abnormal information includes spikes, fluctuations, daily limits, bottom drops, etc.

[0068] Step S22: Annotate each abnormal billboard image to obtain a corresponding annotation file containing annotation information.

[0069] The annotation information includes the anomaly type and coordinate information that characterize the anomaly information in the anomaly signboard image. Coordinate information generally refers to information used to describe the target location (i.e., the location of the anomaly information). Specifically, for target detection tasks, the annotation coordinate information is generally a bounding box that is used to frame the target object of interest in the image.

[0070] Step S23: construct a training set based on each abnormal billboard image and each annotation file.

[0071] In step S23, each abnormal signboard image and each annotation file are packaged into a data set for AI training.

[0072] Step S24: Create a target detection model to be trained.

[0073] Preferably, the target detection model to be trained can be an AI target detection model built using YOLOv8.

[0074] Among them, the YOLO (You Only Look Once) series is a series of real-time target detection algorithms, and YOLOv8 refers to the eighth version of YOLO.

[0075] It should be noted that YOLO is only a preferred embodiment provided by the present invention, and other target detection algorithms can also be used to build a target detection model.

[0076] Step S25: Based on the training set and the pre-written training script, the target detection model to be trained is trained to obtain a trained target detection model.

[0077] In the specific implementation process of step S25, a machine is rented through the cloud computing platform to execute the written training script (due to the special application scenario, the training images in the training set are all very standard monitoring board images, and several hyperparameters such as the probability of image upside down flipping, the probability of image left and right flipping, and the probability of image mosaic are set to 0 in advance), the training set is loaded, and after completing the training objectives in the training script, the trained target detection model is obtained.

[0078] In one embodiment, the trained target detection model is packaged into a back-end AI detection service, which provides an interface for detecting abnormal information in monitoring images. The interface can be called to detect the billboard image.

[0079] Specifically, after obtaining the billboard image, the AI ​​detection service is called. The AI ​​detection service loads the pre-trained target detection model, performs anomaly detection on the billboard image, and returns the detection result of the billboard image. When the detection result is abnormal, the detection result will include a set of all abnormal information. Each piece of abnormal information includes abnormal category, abnormal area coordinate information, confidence level and other information.

[0080] When there is no abnormality in the test result, the test result does not contain abnormal information.

[0081] In one embodiment, after the detection result output by the target detection model is obtained, the detection result is stored in a preset detection result database.

[0082] It is understandable that the records are stored in the database for on-duty personnel to trace back, and when an alarm message is generated later, the abnormal information can be retrieved from the database to prevent the loss of detection results.

[0083] It should be noted that the application of AI target detection models allows on-duty personnel to save a lot of time on inspections, focusing their energy on daily work and solving abnormal problems, greatly improving work efficiency. At the same time, the application of AI target detection models has increased the inspection frequency from twice a day to once every 10 minutes. A higher inspection frequency can help on-duty personnel find problems faster and intervene earlier.

[0084] Step S103: If there is abnormal information in the detection result, alarm information is generated based on the abnormal information, and a preset alarm process is executed.

[0085] In step S103, an alarm strategy is generally configured in the inspection task. If the alarm strategy indicates the alarm recipient and indicates that the alarm function is enabled, alarm information is generated based on the abnormal information and a preset alarm process is executed.

[0086] In the specific implementation process of step S103, if there is abnormal information in the detection result, the abnormal information is marked in the billboard image to obtain an alarm image; alarm information is generated based on the abnormal information and the alarm image, and a preset alarm process is executed.

[0087] It should be noted that the attached alarm image is to facilitate the on-duty personnel to make inspections. Through the visual method of the alarm image, the on-duty personnel can locate the problem more efficiently.

[0088] In one embodiment, the push interface of the communication tool set in the alarm strategy of the inspection task configuration is obtained, and the alarm information is sent through the push interface.

[0089] It can be understood that each push interface corresponds to a duty officer. For example, the push interface of the office communication software is used to notify the corresponding duty officer and the inspection abnormality information synchronization group.

[0090] In one embodiment, after the alarm information is obtained, a life cycle status mark of the initial state is added to the alarm information, and a preset alarm process is executed to send the alarm information with the life cycle status mark added to the on-duty personnel.

[0091] The life cycle status mark is used to indicate the processing progress of the alarm information.

[0092] It is understandable that after receiving the alarm information, the on-duty personnel will perform corresponding emergency processing according to the abnormal information therein, and the current emergency processing progress can be updated in real time by modifying the life cycle status mark.

[0093] In response to a follow-up instruction input by the on-duty personnel, the life cycle status flag is modified based on the follow-up instruction.

[0094] It should be noted that building an exception information lifecycle management process can complete the closed loop of the status of exception information under the follow-up of the on-duty personnel and the stability manager. The status categories include: newly discovered, being followed up, repeated, deferred processing, completed, invalid, etc.

[0095] In one embodiment, multiple pieces of alarm information may be merged according to actual conditions, and a life cycle status mark may be added to the merged alarm information.

[0096] The abnormal life cycle management can help the on-duty personnel and R&D team to manage all abnormalities and hidden dangers and follow up to the end. There will be no incidents where hidden dangers lead to online failures due to manual follow-up negligence.

[0097] Based on the AI-based inspection method disclosed in the above-mentioned embodiment of the present invention, when any preset inspection task is executed, a screenshot of the monitoring board indicated by the inspection task is taken to obtain a board image; the board image is input into the target detection model and detected using the target detection technology to obtain the detection result output by the target detection model; the target detection model is pre-trained using an abnormal board image containing abnormal information; if there is abnormal information in the detection result, an alarm message is generated based on the abnormal information, and a preset alarm process is executed. In this solution, a screenshot of the monitoring board is taken to obtain a board image, and an AI model with target detection function is used to perform target detection on the board image to obtain the detection result, which can greatly improve the work efficiency of the inspection personnel and the accuracy of the inspection results, thereby avoiding missing abnormalities and saving manpower and material resources.

[0098] Corresponding to the AI-based inspection method disclosed in the above embodiment of the present invention, Figure 4 As shown, it is a structural diagram of an AI inspection system disclosed in an embodiment of the present invention, and the system includes: a screenshot module 401, an AI detection module 402 and an alarm module 403.

[0099] The screenshot module 401 is used to take a screenshot of the monitoring board indicated by any preset inspection task when the inspection task is executed, so as to obtain a board image;

[0100] AI detection module 402, used for inputting the billboard image into a target detection model to perform detection using target detection technology, and obtaining a detection result output by the target detection model; the target detection model is pre-trained using an abnormal billboard image containing abnormal information;

[0101] The alarm module 403 is used to generate alarm information based on the abnormal information if the abnormal information exists in the detection result, and execute a preset alarm process.

[0102] In one embodiment, the AI ​​inspection system further includes:

[0103] A training module, used for obtaining a plurality of abnormal billboard images containing abnormal information;

[0104] Annotate each of the abnormal billboard images to obtain a corresponding annotation file containing annotation information; the annotation information includes: abnormality type and coordinate information;

[0105] Based on each of the abnormal billboard images and each of the annotation files, construct a training set;

[0106] Create a target detection model to be trained;

[0107] Based on the training set and a pre-written training script, the target detection model to be trained is trained to obtain a trained target detection model.

[0108] In one embodiment, the alarm module 403 is specifically used to:

[0109] If the abnormal information exists in the detection result, the abnormal information is marked in the billboard image to obtain an alarm image;

[0110] Alarm information is generated based on the abnormal information and the alarm image, and a preset alarm process is executed.

[0111] In one embodiment, the AI ​​inspection system further includes:

[0112] An alarm execution module is used to obtain the push interface of the communication tool set in the alarm strategy of the inspection task configuration; each of the push interfaces corresponds to a duty officer;

[0113] The warning information is sent through the push interface.

[0114] In one embodiment, the AI ​​inspection system further includes:

[0115] A cycle management module, used to add a life cycle state mark of an initial state to the alarm information; the life cycle state mark is used to indicate the processing progress of the alarm information;

[0116] In response to a follow-up instruction input by an on-duty personnel, the life cycle status mark is modified based on the follow-up instruction.

[0117] In one embodiment, the AI ​​inspection system further includes:

[0118] The task trigger module is used to poll the preset configuration information database to obtain the inspection configuration information corresponding to the inspection task;

[0119] Acquire a cron expression from the inspection configuration information; the cron expression is used to represent the execution time of the inspection task;

[0120] The inspection task corresponding to the inspection configuration information is executed based on the cron expression.

[0121] In one embodiment, the AI ​​inspection system further includes:

[0122] The storage module is used to store the detection results output by the target detection model into a preset detection result database after obtaining the detection results.

[0123] Based on the AI-based inspection system disclosed in the above-mentioned embodiment of the present invention, when any preset inspection task is executed, a screenshot of the monitoring board indicated by the inspection task is taken to obtain a board image; the board image is input into the target detection model and detected using the target detection technology to obtain the detection result output by the target detection model; the target detection model is pre-trained using an abnormal board image containing abnormal information; if there is abnormal information in the detection result, an alarm message is generated based on the abnormal information, and a preset alarm process is executed. In this solution, a screenshot of the monitoring board is taken to obtain a board image, and an AI model with target detection function is used to perform target detection on the board image to obtain the detection result, which can greatly improve the work efficiency of the inspection personnel and the accuracy of the inspection results, thereby avoiding missing abnormalities and saving manpower and material resources.

[0124] The present invention also provides an electronic device, please refer to Figure 5 , the electronic device includes a memory 501 and a processor 502.

[0125] Wherein, the memory 501 is used to store computer programs;

[0126] The processor 502 is used to execute a computer program, specifically to implement the AI-based inspection method provided in any embodiment of the present application.

[0127] The present application also provides a computer storage medium for storing a computer program. When the computer program is executed, it is specifically used to implement the AI-based inspection method provided in any embodiment of the present application.

[0128] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0129] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0130] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An AI-based inspection method, characterized in that: The method comprises: When any preset inspection task is executed, a screenshot is taken of the monitoring board indicated by the inspection task to obtain a board image; Inputting the billboard image into a target detection model and performing detection using a target detection technology to obtain a detection result output by the target detection model; the target detection model is pre-trained using an abnormal billboard image containing abnormal information; If the abnormal information exists in the detection result, alarm information is generated based on the abnormal information, and a preset alarm process is executed.

2. The method according to claim 1, characterized in that: The training process of the anomaly detection model includes: Acquire multiple abnormal billboard images containing abnormal information; Annotate each of the abnormal billboard images to obtain a corresponding annotation file containing annotation information; the annotation information includes: abnormality type and coordinate information; Based on each of the abnormal billboard images and each of the annotation files, construct a training set; Create a target detection model to be trained; Based on the training set and a pre-written training script, the target detection model to be trained is trained to obtain a trained target detection model.

3. The method according to claim 1, characterized in that If the abnormal information exists in the detection result, generating alarm information based on the abnormal information and executing a preset alarm process include: If the abnormal information exists in the detection result, the abnormal information is marked in the billboard image to obtain an alarm image; Alarm information is generated based on the abnormal information and the alarm image, and a preset alarm process is executed.

4. The method according to claim 1, characterized in that: The execution of the preset alarm process includes: Obtain the push interface of the communication tool set in the alarm strategy of the inspection task configuration; each of the push interfaces corresponds to a duty officer; The warning information is sent through the push interface.

5. The method according to claim 1, characterized in that The method further comprises: Adding a life cycle status mark of an initial state to the alarm information; the life cycle status mark is used to indicate the processing progress of the alarm information; In response to a follow-up instruction input by an on-duty personnel, the life cycle status mark is modified based on the follow-up instruction.

6. The method according to claim 1, characterized in that The method further comprises: Poll the preset configuration information database to obtain the inspection configuration information corresponding to the inspection task; Acquire a cron expression from the inspection configuration information; the cron expression is used to represent the execution time of the inspection task; The inspection task corresponding to the inspection configuration information is executed based on the cron expression.

7. The method according to any one of claims 1 to 6, characterized in that: After obtaining the detection result output by the target detection model, the method further includes: The detection results are stored in a preset detection result database.

8. An AI-based inspection system, characterized in that: The system comprises: A screenshot module is used to take a screenshot of the monitoring dashboard indicated by any preset inspection task when the inspection task is executed, so as to obtain a dashboard image; An AI detection module is used to input the billboard image into a target detection model and perform detection using target detection technology to obtain a detection result output by the target detection model; the target detection model is pre-trained using an abnormal billboard image containing abnormal information; The alarm module is used to generate alarm information based on the abnormal information if the abnormal information exists in the detection result, and execute a preset alarm process.

9. A storage medium, characterized in that: Used to store computer programs, which, when executed, are specifically used to implement the AI-based inspection method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer programs; The processor is used to execute the computer program, specifically to implement the AI-based inspection method as described in any one of claims 1 to 7.