Business system operation and maintenance method and device based on image recognition

Abstract

The invention discloses a service system operation and maintenance method and device based on image recognition, relates to the technical field of service system operation and maintenance and intelligent operation and maintenance, can be used in the field of financial science and technology, and comprises the following steps: capturing a current interface screenshot of a service system after receiving error report information in the operation process of the service system; identifying a function object parameter and a problem type parameter according to the current interface screenshot; obtaining an interface object parameter in a code corresponding to the business system according to the function object parameter and the problem type parameter; taking the interface object parameter as a keyword, retrieving an abnormal log file of the business system, and obtaining an error reporting reason and an error reporting parameter of the business system; according to the interface object parameters, the error reporting reasons and the error reporting parameters, determining a plurality of solutions after credibility sorting; and sequentially executing the solutions after credibility sorting until an error is reported and repaired. The error processing efficiency of the business system can be improved, the workload of operation and maintenance personnel is reduced, and the operation and maintenance cost is low.

Description

Technical Field

[0001] The present invention relates to the fields of business system operation and maintenance and intelligent operation and maintenance technology, and can be used in the field of financial technology. In particular, it relates to a business system operation and maintenance method and device based on image recognition. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. No statement herein is admitted to be prior art by virtue of its inclusion in this section.

[0003] Functional errors that occur during the use of business systems (such as banks' business systems) are usually reported by users to the corresponding development support personnel for production operations and maintenance. Support personnel locate the corresponding functional modules and investigate the specific code that caused the error based on the error information, error interface screenshots, and operation descriptions provided by users. Specifically, in the process of resolving business system errors, development support personnel need to devote considerable effort to log code retrieval and logical analysis, and then use their experience to determine the cause of the problem, provide solutions, and repair it. This process takes up a large amount of operation and maintenance time and places high demands on the business operations of development support personnel. In the event of personnel adjustments, solutions to repetitive problems cannot be reused, increasing operation and maintenance costs. Summary of the invention

[0004] In a first aspect, an embodiment of the present invention provides a business system operation and maintenance method based on image recognition. The method uses image recognition technology to extract key feature information, performs similarity analysis in combination with business system logs, and intelligently recommends error reporting solutions for the business system. This improves the efficiency of error processing in the business system, reduces the workload of operation and maintenance personnel, and solves the problem of high operation and maintenance costs. The method includes:

[0005] After receiving the error message during the operation of the business system, capture the current interface screenshot of the business system;

[0006] Based on the current interface screenshot, identify the function object parameters and question type parameters;

[0007] According to the function object parameters and the problem type parameters, obtain the interface object parameters in the code corresponding to the business system;

[0008] Using the interface object parameters as keywords, searching the business system exception log file to obtain the error cause and error parameters of the business system;

[0009] Determine multiple solutions ranked by credibility based on interface object parameters, error causes, and error parameters;

[0010] Execute the solutions ranked by credibility in sequence until the error is fixed.

[0011] In a second aspect, an embodiment of the present invention further provides another operation and maintenance device for a business system based on image recognition. By using image recognition technology to extract key feature information, and combining with the business system log for similarity analysis, it can intelligently recommend solutions for error reports of the business system, improve the efficiency of error handling in the business system, reduce the workload of operation and maintenance personnel, and solve the problem of high operation and maintenance costs. The device includes:

[0012] An image processing module, configured to capture a current interface screenshot of the business system after receiving an error message during the operation of the business system; identify function object parameters and problem type parameters according to the current interface screenshot; and obtain interface object parameters in the code corresponding to the business system according to the function object parameters and the problem type parameters.

[0013] A log processing module, configured to use the interface object parameters as keywords to retrieve the business system exception log file, and obtain the error cause and error parameters of the business system.

[0014] A solution module, configured to determine multiple solutions sorted by credibility according to the interface object parameters, the error cause, and the error parameters, where the solution is a solution for repairing the error.

[0015] An operation and maintenance execution module, configured to sequentially execute the solutions sorted by credibility until the error is repaired.

[0016] In a third aspect, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned operation and maintenance method for a business system based on image recognition.

[0017] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned operation and maintenance method for a business system based on image recognition.

[0018] In a fifth aspect, an embodiment of the present invention proposes a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the operation and maintenance method for a business system based on image recognition.

[0019] In an embodiment of the present invention, after receiving error information during the operation of a business system, a screenshot of the current interface of the business system is captured; based on the current interface screenshot, function object parameters and problem type parameters are identified; based on the function object parameters and problem type parameters, interface object parameters in the code corresponding to the business system are obtained; using the interface object parameters as keywords, the business system abnormality log file is retrieved to obtain the error cause and error parameters of the business system; based on the interface object parameters, the error cause and the error parameters, multiple solutions ranked by credibility are determined; and the solutions ranked by credibility are executed in sequence until the error is repaired. Compared with the existing solution of manual operation and maintenance by development support personnel, the method proposed in the embodiment of the present invention uses image recognition technology to extract key feature information, combines it with the business system log for similarity analysis, and intelligently recommends error solutions for the business system, thereby improving the efficiency of error processing in the business system, reducing the workload of operation and maintenance personnel, and solving the problem of high operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0021] Figure 1 Flowchart of a business system operation and maintenance method based on image recognition in an embodiment of the present invention;

[0022] Figure 2 The detailed process of business system operation and maintenance based on image recognition in an embodiment of the present invention;

[0023] Figure 3 Schematic diagram of a business system operation and maintenance interface comparison table in an embodiment of the present invention;

[0024] Figure 4 Schematic diagram of the generation principle of error cause R and error parameter P in an embodiment of the present invention;

[0025] Figure 5 A schematic diagram of a solution set obtained in an embodiment of the present invention;

[0026] Figure 6 Schematic diagram of a business system operation and maintenance device based on image recognition in an embodiment of the present invention;

[0027] Figure 7 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer and more understandable, the following further elaborates on the embodiments of the present invention with reference to the accompanying drawings. Herein, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but not to limit the present invention.

[0029] In the technical solution of this application, the acquisition, storage, use, processing, etc. of data all comply with the relevant provisions of national laws and regulations.

[0030] Figure 1 The following is a flowchart of the operation and maintenance method for a business system based on image recognition in an embodiment of the present invention, including:

[0031] Step 101: After receiving an error message during the operation of the business system, capture a screenshot of the current interface of the business system.

[0032] Step 102: According to the current interface screenshot, identify the function object parameters and problem type parameters.

[0033] Step 103: According to the function object parameters and problem type parameters, obtain the interface object parameters in the corresponding code of the business system.

[0034] Step 104: Use the interface object parameters as keywords to retrieve the business system exception log file, and obtain the error cause and error parameters of the business system.

[0035] Step 105: According to the interface object parameters, error cause, and error parameters, determine multiple solutions sorted by credibility, where the solution is a solution for fixing the error.

[0036] Step 106: Sequentially execute the solutions sorted by credibility until the error is fixed.

[0037] In the embodiment of the present invention, image recognition technology is used to extract key feature information, combined with similarity analysis of the business system logs, to intelligently recommend error solutions for the business system, improve the efficiency of error handling in the business system, reduce the workload of operation and maintenance personnel, and solve the problem of high operation and maintenance costs.

[0038] The following details each step. See Figure 2 The following is the detailed process of the operation and maintenance of a business system based on image recognition in an embodiment of the present invention.

[0039] In step 101: After receiving an error message during the operation of the business system, capture a screenshot of the current interface of the business system.

[0040] Business system interfaces all have unified design requirements. The order in which the top-level menus are opened step by step to the functional pages is usually displayed by breadcrumbs or menu tree-style tracking, and each page can accurately correspond to a specific functional module. Image recognition technology is used to uniformly process the business system interfaces. Any screenshot of the business system can correspond to a functional module, locking the functional scope where the abnormality occurred. In addition, the business system error prompt information has different color identification specifications. For example: a green background color represents a successful operation, a yellow background color represents an illegal operation warning, a gray background color represents a guidance reminder prompt, and a red background color represents a business system execution error. During the image recognition process, the color can be used to determine the type of user operation problem, thereby determining the next operation of the intelligent operation and maintenance model.

[0041] In one embodiment, (step 102) identifying function object parameters and question type parameters based on the current interface screenshot includes:

[0042] Capture screenshots of the current interface of the business system;

[0043] Based on the menu structure characteristics preset by the business system, identify the current menu from the current interface screenshot;

[0044] Determine function object parameters according to the menu;

[0045] According to the error prompt logo in the current interface screenshot, the problem type parameter is identified, and the error prompt logo includes a background color and / or text.

[0046] When a business system error occurs, a screenshot of the current interface is automatically captured. Based on the system's pre-set menu structure, the current menu is identified and the functional scope is located. The error message background color and text in the screenshot are used to identify the type of error. This image recognition process generates two parameters: the function object parameter F and the problem type parameter T.

[0047] In one embodiment, (step 103) obtaining interface object parameters in the code corresponding to the business system according to the function object parameters and the problem type parameters includes:

[0048] According to the functional object parameters and problem type parameters, the preset business system operation and maintenance interface comparison table is searched to obtain the interface object parameters in the code corresponding to the business system, wherein each row of the business system operation and maintenance interface comparison table represents the prompt information returned when an exception occurs in the interface corresponding to the interface object parameter, and the prompt information corresponds to the problem type parameter, and each column of the business system operation and maintenance interface comparison table represents the functional object parameter to which the prompt information belongs.

[0049] During implementation, the function object parameters and problem type parameters are compared with the preset business system operation and maintenance interface comparison table. This will reveal the interface object parameters specified in the code, such as the English name of the interface object. This can be used as a keyword to retrieve the corresponding log content in the large exception log file, thus narrowing the scope of the exception search. The business system operation and maintenance interface comparison table is a data asset formed according to the naming conventions of business system development, with the business system interface as the dimension and the function and problem type as the parameters. Figure 3 This is a schematic diagram of the business system operation and maintenance interface comparison table in an embodiment of the present invention. Each page error message can be mapped to one or more interface object parameters I in this table, forming a parameter set [FTI1, FTI2, ..., FTI n ]It is convenient to quickly locate the exception log content in the exception log file error_log.

[0050] In step 104, the interface object parameter is used as a keyword to search the business system exception log file to obtain the error cause and error parameters of the business system;

[0051] The exception log file is the error_log file. When the business system reports an error during operation, the latest error_log file is automatically obtained, and the log is retrieved according to the "function object parameter F", "problem type parameter T", and "interface object parameter I". By parsing the relevant logs, the error cause and specific error parameters can be analyzed and extracted. The process is equivalent to the operation and maintenance developers manually searching the logs and extracting error information, such as a null pointer causing a save failure, a query returning too long information causing a business system response timeout, etc. After parsing the error_log file, two corresponding parameters will be generated for each object in the parameter set generated in the previous step: the error cause R and the error parameter P, forming a set with a higher parameter dimension [FTI1R1P1, FTI2R2P2,..., FTI n R n P n ] is used to generate a solution in the next step. Since the I, R, and P parameters are already the error parsing parameters after precise positioning, the F and T in the set are the most original parameters recognized by the image processing domain and can be omitted. Therefore, the parameter set passed to the next step is [I1R1P1,I2R2P2,...,I n R n P n ], see Figure 4 2 is a diagram showing the principle of generating the error reason R and the error parameter P in an embodiment of the present invention.

[0052] In one embodiment, (step 105) determining multiple solutions ranked by credibility based on the interface object parameters, the error cause, and the error parameters includes:

[0053] Generate a set of solutions according to the interface object parameters, error cause, and error parameters;

[0054] Rank all the solutions in the set of solutions according to their credibility.

[0055] In one embodiment, generating a set of solutions according to the interface object parameters, error cause, and error parameters includes:

[0056] Form a sequence with each interface object parameter and the corresponding retrieved error cause and error parameter;

[0057] Match and obtain the solution S corresponding to the i-th sequence represented by the following formula from the system operation and maintenance result set i ;

[0058] S i =(s i0 I + s i1 R + s i2 P)

[0059] wherein, I, R, and P are the interface object parameter, error cause, and error parameter respectively; s i0 , s i1 and s i2 are the credibility weights of the interface object parameter, error cause, and error parameter respectively.

[0060] See Figure 5 which is the schematic diagram for obtaining the set of solutions in the embodiment of the present invention. Specifically in implementation, the set of solutions S can be S = [(s 00 I + s 01 R + s 02 P)S0, (s 10 I + s 11 R + s 12 P)S1,..., (s n0 I + s n1 R + s n2 P)S n , and s n(0～2) is used for the reliability analysis of the solutions. In the process of searching for solutions in the business system operation and maintenance solution set, it is necessary to match I first, then match R and P, and put all possible solutions into the set S. In this link, the credibility weights of the solutions corresponding to the I, R, and P parameters are different.

[0061] In one embodiment, the method further includes:

[0062] Based on the historical data of the credibility weights of interface object parameters, the credibility weights of error causes and the credibility weights of error parameters, a neural network model is used to train the credibility weights of interface object parameters, the credibility weights of error causes and the credibility weights of error parameters.

[0063] Through the above training, the credibility weight coefficient is continuously adjusted, so that the positioning of the solution is more accurate and more credible.

[0064] In one embodiment, all solutions in the solution set are ranked according to their credibility, including:

[0065] Add the solution with the credibility weight of the interface object parameter as 1 to the automatic operation and maintenance solution set;

[0066] Add the solution with a credibility weight of 0 for the interface object parameter to the manual processing operation and maintenance solution set;

[0067] Sequentially merge the set of automatic operation and maintenance solutions and the set of manual operation and maintenance solutions to form a hierarchically sorted solution;

[0068] In the automatic operation and maintenance solution set and the manual operation and maintenance solution set, the solutions are sorted according to the following steps:

[0069] Calculate the sum of the credibility weight of the error cause and the credibility weight of the error parameter;

[0070] Sort the solutions by sum value from largest to smallest.

[0071] In specific implementation, the I parameter must be 100% matched because the error information can only come from one front-end and back-end interface, so s0 = 1; if s0 = 0, that is, the I parameter fails to match a solution, then s1 and s2 are obtained based on similar solutions matched by the R and P parameters in the result set. The larger the sum of the values, the higher the credibility.

[0072] The credibility weight coefficients s0, s1, and s2 in the solution set S are used to rank the solutions. First, based on s0, solutions with s0 = 1 are added to the set of automated O&M solutions. These solutions can be automatically invoked through the interface and are ranked before the set of manual O&M solutions with s0 = 0. The two sets are then ranked from largest to smallest based on the sum of γ = s1 + s2. Solutions ranked first are more credible than those ranked later.

[0073] In step 106, the solutions ranked by credibility are executed in sequence until the error is fixed.

[0074] That is, according to the classification, first take out the solution with the highest ranking. If there is an error and it can be repaired, otherwise, take out the solution with the second highest ranking, and so on. Assume that the set of solutions after sorting is S order =[γ0IS0,γ1IS1,...,γ k IS k +[γ0S0,γ1S1,...,γ j S j , where the first array is the set of automated operation and maintenance solutions, which contains interface parameters and can be automatically executed in the order of credibility. Problem repair is performed through the method of calling entity interfaces to ensure user use. If the problem is still not repaired after all the solutions in the first set are executed, the second set of solutions is enabled. The second array is the set of manual processing operation and maintenance solutions. Since the solutions in it cannot correspond to a specific interface, they can only be sorted in descending order of credibility γ. The second set of solutions can be sent to development support personnel via email for manual collection, analysis, and repair. The solutions for error repair are inserted into the set of operation and maintenance solutions of the business system, realizing the accumulation and improvement of operation and maintenance assets.

[0075] An embodiment of the present invention also proposes an operation and maintenance device for a business system based on image recognition. Its principle is similar to the operation and maintenance method for a business system based on image recognition, and will not be elaborated here.

[0076] Figure 6 is a schematic diagram of the operation and maintenance device for a business system based on image recognition in an embodiment of the present invention, including:

[0077] An image processing module 601, configured to capture a current interface screenshot of the business system after receiving error information during the operation of the business system; identify functional object parameters and problem type parameters according to the current interface screenshot; and obtain interface object parameters in the code corresponding to the business system according to the functional object parameters and problem type parameters;

[0078] A log processing module 602, configured to use the interface object parameters as keywords to retrieve the business system exception log file to obtain the error cause and error parameters of the business system;

[0079] A solution module 603, configured to determine multiple solutions sorted by credibility according to the interface object parameters, error cause, and error parameters, where the solutions are solutions for repairing errors;

[0080] An operation and maintenance execution module 604, configured to sequentially execute the solutions sorted by credibility until the error is repaired.

[0081] In an embodiment, the image processing module is specifically configured to:

[0082] Capture a current interface screenshot of the business system;

[0083] Identify the current menu from the current interface screenshot based on the menu structure features preset in the business system;

[0084] Determine the function object parameters according to the menu;

[0085] Identify the problem type parameters based on the error prompt identifier in the current interface screenshot, where the error prompt identifier includes the background color and / or text.

[0086] In one embodiment, the image processing module is specifically configured to:

[0087] According to the function object parameters and the problem type parameters, search for the preset business system operation and maintenance interface comparison table to obtain the interface object parameters in the code corresponding to the business system. Each row of the business system operation and maintenance interface comparison table represents the prompt information returned when an exception occurs in the interface corresponding to the interface object parameter, and the prompt information corresponds to the problem type parameter. Each column of the business system operation and maintenance interface comparison table represents the function object parameter to which the prompt information belongs.

[0088] In one embodiment, the solution module is specifically configured to:

[0089] Generate a solution set according to the interface object parameters, the error cause, and the error parameters;

[0090] Rank all the solutions in the solution set according to their credibility.

[0091] In one embodiment, the solution module is specifically configured to:

[0092] Form a sequence with each interface object parameter and the corresponding error cause and error parameters retrieved;

[0093] Match and obtain the solution S corresponding to the i-th sequence represented by the following formula from the system operation and maintenance result set i ;

[0094] S i =(s i0 I + s i1 R + s i2 P)

[0095] where I, R, and P are the interface object parameter, the error cause, and the error parameter respectively; s i0 , s i1 and s i2 are the credibility weights of the interface object parameter, the credibility weight of the error cause, and the credibility weight of the error parameter respectively.

[0096] In one embodiment, the solution module is specifically configured to:

[0097] Based on the historical data of the credibility weights of interface object parameters, the credibility weights of error reporting reasons, and the credibility weights of error reporting parameters, a neural network model is used to train the credibility weights of interface object parameters, the credibility weights of error reporting reasons, and the credibility weights of error reporting parameters.

[0098] In one embodiment, the solution module is specifically configured to:

[0099] Add the solutions with a credibility weight of 1 for interface object parameters to the automatic execution operation and maintenance solution set;

[0100] Add the solutions with a credibility weight of 0 for interface object parameters to the manual processing operation and maintenance solution set;

[0101] Sequentially merge the automatic execution operation and maintenance solution set and the manual processing operation and maintenance solution set to form a sorted solution after hierarchical sorting;

[0102] Among them, within the automatic execution operation and maintenance solution set and the manual processing operation and maintenance solution set, the solutions are sorted according to the following steps:

[0103] Calculate the sum value of the credibility weight of the error reporting reason and the credibility weight of the error reporting parameter;

[0104] Sort the solutions from large to small according to the sum value.

[0105] In summary, in the method and device proposed in the embodiments of the present invention, through image processing, log processing, solutions, and operation and maintenance execution, the error screenshots of the business system and the business system logs are automatically parsed to generate a solution set for business system errors, and each solution in the set is sorted according to credibility, and the business system is self-repaired from high to low. The solutions proposed in this solution will effectively reduce the dependence of development support personnel on experience, reduce labor costs and the time for manual positioning and troubleshooting problems, reduce repetitive labor, and automatically repair problems according to the solution set, improving the efficiency of business system operation and maintenance. For solutions that cannot be self-repaired, an email reminder is sent to the development support personnel for manual operation and maintenance, and new operation and maintenance solutions are supplemented to the business system operation and maintenance solution set to achieve iterative improvement and accumulation of operation and maintenance assets.

[0106] The embodiments of the present invention also provide a computer device, Figure 7 is a schematic diagram of the computer device in the embodiments of the present invention. The computer device 700 includes a memory 710, a processor 720, and a computer program 730 stored on the memory 710 and executable on the processor 720. When the processor 720 executes the computer program 730, the above-mentioned business system operation and maintenance method based on image recognition is implemented.

[0107] An embodiment of the present invention also provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the above-mentioned operation and maintenance method for an image recognition-based business system.

[0108] An embodiment of the present invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-mentioned operation and maintenance method for an image recognition-based business system.

[0109] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a business system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0110] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (business systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0111] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that realizes the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0112] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0113] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for operation and maintenance of a business system based on image recognition, characterized in that Including: After receiving the error message during the operation of the business system, capture the current interface screenshot of the business system; According to the current interface screenshot, identify the function object parameter and the problem type parameter; According to the function object parameter and the problem type parameter, obtain the interface object parameter in the corresponding code of the business system; Use the interface object parameter as a keyword to retrieve the business system exception log file, and obtain the error cause and error parameter of the business system; According to the interface object parameter, error cause and error parameter, determine multiple solutions sorted by credibility, and the solution is a solution to repair the error; Execute the solutions sorted by credibility in sequence until the error is repaired.

2. The method according to claim 1, characterized in that According to the current interface screenshot, identify the function object parameter and the problem type parameter, including: Capture the current interface screenshot of the business system; Based on the preset menu structure feature of the business system, identify the current menu from the current interface screenshot; According to the menu, determine the function object parameter; Based on the error prompt identifier in the current interface screenshot, identify the problem type parameter, and the error prompt identifier includes the background color and / or text.

3. The method according to claim 1, wherein According to the function object parameter and the problem type parameter, obtain the interface object parameter in the corresponding code of the business system, including: According to the function object parameter and the problem type parameter, search the preset business system operation and maintenance interface comparison table to obtain the interface object parameter in the corresponding code of the business system. Each row of the business system operation and maintenance interface comparison table represents the prompt information returned when the interface corresponding to the interface object parameter has an exception, and the prompt information corresponds to the problem type parameter. Each column of the business system operation and maintenance interface comparison table represents the function object parameter to which the prompt information belongs.

4. The method according to claim 1, wherein According to the interface object parameter, error cause and error parameter, determine multiple solutions sorted by credibility, including: Generate a solution set according to the interface object parameter, error cause and error parameter; Sort all the solutions in the solution set by credibility.

5. The method according to claim 4, characterized in that, Generate a solution set according to the interface object parameter, error cause and error parameter, including: Form a sequence with each interface object parameter and the corresponding retrieved error cause and error parameter; Match and obtain the solution S corresponding to the i-th sequence represented by the following formula from the system operation and maintenance result set i ; S i = (s i0 I + s i1 R + s i2 P) Among them, I, R, and P are the interface object parameter, error cause, and error parameter respectively; si0, si1, and si2 are the credibility weights of the interface object parameter, error cause, and error parameter respectively.

6. The method according to claim 5, wherein It also includes: Based on the historical data of the credibility weights of the interface object parameter, error cause, and error parameter, use a neural network model to train the credibility weights of the interface object parameter, error cause, and error parameter.

7. The method according to claim 4, characterized in that Sort all the solutions in the solution set by credibility, including: Add the solutions with the credibility weight of the interface object parameter being 1 to the automatic execution operation and maintenance solution set; Add the solutions with the credibility weight of the interface object parameter being 0 to the manual processing operation and maintenance solution set; Sequentially merge the automatically executed operation and maintenance solution set and the manually processed operation and maintenance solution set to form a sorted solution after hierarchical sorting; Among them, within the automatically executed operation and maintenance solution set and the manually processed operation and maintenance solution set, the solutions are sorted according to the following steps: Calculate the sum of the credibility weights of the error cause and the credibility weights of the error parameters; Sort the solutions from largest to smallest according to the sum value.

8. An operation and maintenance device for a business system based on image recognition, characterized in that, Including: An image processing module, configured to capture a current interface screenshot of the business system after receiving an error message during the operation of the business system; Identify the function object parameters and problem type parameters according to the current interface screenshot; Obtain the interface object parameters in the code corresponding to the business system according to the function object parameters and problem type parameters; A log processing module, configured to use the interface object parameters as keywords to retrieve the business system exception log file to obtain the error cause and error parameters of the business system; A solution module, configured to determine multiple solutions sorted by credibility according to the interface object parameters, error cause, and error parameters, and the solution is a solution for fixing the error; An operation and maintenance execution module, configured to sequentially execute the solutions sorted by credibility until the error is fixed.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the method according to any one of claims 1 to 7.

Images