An automatic login method and related products suitable for distribution network operation analysis

Abstract

The present invention relates to the field of automation technology, and in particular to an automatic login method and related products suitable for distribution network operation analysis, the method comprising: building a database; determining a system type corresponding to a current login page, loading a corresponding recognition template for matching, and determining the position of key elements of the current login page; obtaining a multi-factor response function corresponding to each key element; dynamically adjusting the input of each key element based on the multi-factor response function; submitting a login instruction after completing the input, the system returns a status code, and determines whether the login is successful based on the status code, and re-logins if the login is unsuccessful; the present invention can maintain high recognition accuracy and input reliability when facing complex and changeable login page structures and system states; and through a mechanism for dynamically adjusting the input method, the system can effectively cope with the influence of external factors such as network delay and system load, and significantly improve the success rate of automatic login.

Description

Technical Field

[0001] The present invention relates to the field of automation technology, and in particular to an automatic login method and related products suitable for distribution network operation analysis. Background Art

[0002] With the widespread application of distribution network operation and analysis systems in the power industry, operators need to frequently log in to multiple different systems (such as power supply and service systems, distribution and self-distribution systems, leakage protection systems, etc.) to obtain and process relevant data. However, due to the different login interfaces and requirements of different systems, manual login operations are not only time-consuming and laborious, but also prone to errors. Some existing automatic login tools often lack sufficient intelligence and flexibility, and are difficult to cope with complex login page layout changes and system status fluctuations.

[0003] In the prior art, although there are some batch login methods implemented through scripts or automation tools, most of these methods rely on fixed page structures and static operation processes and cannot adapt to a variety of complex practical scenarios. For example, in actual applications, since system access may be affected by various factors such as network conditions and server load, the input method and strategy need to be dynamically adjusted during the automatic login process to ensure the success rate of the login operation.

[0004] Therefore, there is an urgent need for an automatic login method that is more intelligent, flexible, and can dynamically adapt to different systems and states, so as to improve the efficiency and success rate of login operations and provide stable and reliable technical support for power distribution network operation analysis. Summary of the invention

[0005] The technical problem to be solved by the present invention is that the existing automatic login method lacks flexibility and intelligence when facing multiple distribution network operation systems, and it is difficult to cope with the differences in login pages and system status fluctuations between different systems, resulting in low login success rate and low operating efficiency. The present invention aims to provide an automatic login method and related products suitable for distribution network operation analysis, which realizes intelligent identification of multiple distribution network operation systems, automatic credential filling, and dynamic adjustment of input strategies, and can still efficiently and reliably complete the automatic login operation when the network status and system load fluctuate.

[0006] The present invention is achieved through the following technical solutions:

[0007] An automatic login method suitable for distribution network operation analysis, comprising:

[0008] Build a database that contains identification templates, account and password information for each system;

[0009] Determine the system type corresponding to the current login page, load the corresponding recognition template for matching, and determine the location of the key elements of the current login page;

[0010] According to the key element positions and the system state vector, the multi-factor response function corresponding to each key element is obtained;

[0011] Dynamically adjust the input of each key element based on a multi-factor response function;

[0012] After completing the input, submit the login command. The system returns a status code and determines whether the login is successful based on the status code. If the login is unsuccessful, log in again.

[0013] Specifically, the method for determining the system type corresponding to the current login page is: presetting the URL of each system in the database, obtaining the URL of the current login interface when logging in, and matching through the URL to determine the system type.

[0014] Optionally, the method for determining the position of the key element of the current login page includes:

[0015] Downsample the current login interface and the recognition template at different scales to form multi-scale pyramids, respectively, to obtain the target image and template image at multiple scales;

[0016] Calculate the gradients of the target image and template image at each scale respectively;

[0017] The regularization term is determined by calculating the gradient difference between the target image and the template image at each scale;

[0018] Combine pixel value differences, gradient differences, and regularization terms to calculate the basic matching model between the target image and the template image at each scale;

[0019] Calculate the gradient direction consistency between the target image and the template image at each scale;

[0020] Combine the basic matching model and gradient direction consistency check to determine the best matching function at each scale;

[0021] By weighted fusion of the best matching functions at all scales, the best matching position of the key element is determined, which is the key element position.

[0022] Optionally, the key elements include a user name input box, a password input box, a verification code input box, a verification code box, and a login button.

[0023] Optionally, the method for determining the position of the key element of the current login page specifically includes:

[0024] Downsample the current login interface at different scales to form a multi-scale pyramid and obtain the Target image at scale ;

[0025] The recognition template is downsampled at different scales to form a multi-scale pyramid and obtain the Template image at scale ;

[0026] Calculate the target image in Gradient at scale ; Calculate the template image in the Gradient at scale ;

[0027] Determine the regularization term at each scale , ,in, For the target image Scale at location The gradient value at For the template image in Under scale The gradient value of the position, and Respectively The width and height of the template image at scale, is the gradient weight coefficient of the regularization term, and is the index value;

[0028] Determine the basic matching model at each scale , ,in, For the target image Scale at location The pixel value at For the template image in Under scale The pixel value of the position, is the gradient weight coefficient of the basic matching function, is the weight coefficient of the regularization term, and is the index value;

[0029] Determine the consistency check model based on the gradient direction at each scale , ,in, For the target image The gradient direction at scale, For the template image in The gradient direction at scale, and is the index value;

[0030] Determine the best matching function at each scale , ,in, is the weight coefficient of the gradient direction consistency measure;

[0031] Get the best matching position of key elements , ,in, is the total number of scales, For the The weight coefficient of the scale, argmin is the minimum function.

[0032] Optionally, judging whether there is a verification code input box and a verification code box according to the position of the key element, and if there is a verification code box, identifying and obtaining the verification code;

[0033] Methods for obtaining the multi-factor response functions corresponding to each key element include:

[0034] Collect the current system state vector ,in, For the Impact factors;

[0035] Through the key element position and system state vector, the multi-factor response function corresponding to each key element is obtained ,in, Enter the position coordinates of the key element input box. Bias for the base level of the adjusted response function for the key element input box, For key element input box The weight of the direction position, For key element input box The weight of the direction position, For the Impact factors, is the index value, , , Is the key element input box position and The coefficients of the second-order and cross terms of is a nonlinear activation function, is the first Impact Factor The weight coefficient of .

[0036] Optionally, the method for dynamically adjusting the input of each key element based on the multi-factor response function includes:

[0037] Calculate the response function value of each key element input box through the multi-factor response function, and arrange the response function values ​​in ascending order;

[0038] Each key element input box is processed in sequence according to the queue. The processing method includes:

[0039] Obtain the corresponding string from the database according to the attributes of the key element input box; if it is a verification code input box, identify the corresponding verification code from the verification code box;

[0040] Determine the size relationship between the response function value of the key element input box and the set threshold; if the response function value is greater than the set threshold, paste input is used for the string or verification code; if the response function value is not greater than the set threshold, word-by-word input is used for the string or verification code;

[0041] If you use word-by-word input, set the input interval time for each character ,in, Enter the interval time as the basis, is a multi-factor response function, is the adjustment factor, Correct positive numbers for division by zero errors;

[0042] After completing the input of the string or verification code, read the content in the key element input box and compare it with the expected characters. If an error occurs, delete the input content and recalculate the response function value before inputting again.

[0043] If the content in the key element input box is the same as the expected characters, the key element input is completed.

[0044] Further, if a login failure is detected according to the status code, the account and password information in the standby database is called, and it is determined whether the account and password information are the same as the account and password information in the database. If they are not the same, the account and password information in the standby database is used to replace the account and password information in the database;

[0045] Wait for a random time interval and try to log in again;

[0046] If the attempt fails after the set number of times, the recognition template in the standby database is called, and the recognition template in the standby database is used to replace the recognition template in the database, and the account and password information in the database and the account and password information in the standby database are used to log in again;

[0047] If the login fails after the set number of attempts, the login operation will be suspended and human intervention will be notified.

[0048] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the automatic login method applicable to distribution network operation analysis as described above is implemented.

[0049] A computer program product includes a computer program or instructions, which, when executed by a processor, implements the automatic login method applicable to distribution network operation analysis as described above.

[0050] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0051] The automatic login method suitable for distribution network operation analysis proposed in the present invention determines the system type to which the current login page belongs and loads the corresponding recognition template; then accurately locates the key elements of the login page according to the recognition template, such as the user name input box, password input box, verification code box, etc.; on this basis, based on the multi-factor response function, the system dynamically adjusts the input method of each key element. After completing the credential input, the system submits the login request to achieve automatic login.

[0052] By combining the multi-scale pyramid technology with the multi-factor response function, the system can maintain high recognition accuracy and input reliability when facing complex and changeable login page structures and system states; and by dynamically adjusting the input mode mechanism, the system can effectively cope with the influence of external factors such as network delay and system load, thereby significantly improving the success rate of automatic login. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings illustrate exemplary embodiments of the present invention and, together with the description thereof, are used to explain the principles of the present invention. These drawings are included to provide a further understanding of the present invention, and the accompanying drawings are included in and constitute a part of this specification and do not constitute a limitation of the embodiments of the present invention.

[0054] Figure 1 It is a flow chart of an automatic login method suitable for distribution network operation analysis according to the present invention.

[0055] Figure 2 It is a flowchart of the method for dynamically adjusting the input of each key element based on a multi-factor response function according to the present invention. DETAILED DESCRIPTION

[0056] To make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and implementation methods. It is understood that the specific implementation methods described herein are only used to explain the relevant content, rather than to limit the present invention.

[0057] It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings.

[0058] In the absence of conflict, the embodiments and features of the embodiments of the present invention may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0059] Embodiment 1

[0060] like Figure 1 As shown, this embodiment provides an automatic login method suitable for distribution network operation analysis, including:

[0061] Build a database that contains the recognition templates and account password information of each system; the database stores the recognition templates of different systems and the corresponding account password information. Recognition templates refer to standard images or layouts used to identify login pages of different systems. These templates contain the specific positions and structural features of key elements in the page (such as username input box, password input box, verification code box, etc.). By pre-storing these templates, the system can quickly match and identify the system type to which the current page belongs during actual login. Account password information is the credential data required to log in to each system, which is securely stored in the database for use in subsequent automatic login processes.

[0062] Determine the system type corresponding to the current login page, load the corresponding recognition template for matching, and determine the key element position of the current login page; by obtaining the system type of the current login page, the system matches the corresponding recognition template from the database, and then loads these templates for page matching. The matching process involves multi-scale pyramid technology, which generates image pyramids of different resolutions by multi-level downsampling of images, so as to match at multiple scales to improve the robustness and accuracy of recognition. The key element position refers to the specific coordinates of each functional element in the login page (such as the user name input box, password input box, verification code box, login button, etc.). Through template matching, the system can accurately determine the position of these elements in the current page, laying the foundation for subsequent input operations.

[0063] According to the key element position and system state vector, the multi-factor response function corresponding to each key element is obtained; the system will calculate the multi-factor response function of each key element in combination with the current system state vector. The system state vector is a vector composed of multiple influencing factors (such as network delay, system load, page response time, etc.), which reflect the current operating status of the system. The multi-factor response function is used to dynamically adjust the input behavior of the system, and reflects the impact of various factors on the input process through different weight parameters, and finally outputs a response value to guide specific input operations.

[0064] Dynamically adjust the input of each key element based on the multi-factor response function; based on the calculated multi-factor response function, the system will dynamically adjust the input method of each key element. This adjustment is reflected in many aspects, such as input speed, input order, and whether to use paste input or word-by-word input. Through this dynamic adjustment, the system can adapt to the current system state and ensure the stability and accuracy of the input. For example, when the network delay is high, the system may choose word-by-word input to reduce the risk of errors.

[0065] After completing the input, submit the login instruction. The system returns a status code and determines whether the login is successful based on the status code. If the login is unsuccessful, log in again. After all key elements are entered, the system submits the login request. The status code returned by the server is used to indicate the result of the login. The status code is a standardized number or symbol that indicates the server's response status to the request. Common status codes include 200 (successful request), 401 (unauthorized), etc. The system determines whether the login is successful based on the returned status code. If it is unsuccessful, the login process will be re-executed, which usually includes readjusting the input method, using an alternative account or template, etc., until the login is successful.

[0066] Embodiment 2

[0067] The method for determining the system type corresponding to the current login page is: presetting the URL of each system in the database, obtaining the URL of the current login interface when logging in, and matching through the URL to determine the system type.

[0068] Use URL (Uniform Resource Locator) to identify the system to which the current login page belongs. URL is an address pointing to a specific web resource, usually including information such as protocol (such as HTTP / HTTPS), domain name, path, etc. In this embodiment, the URL patterns of various distribution network operation systems are pre-stored in the database. For example, the URL of the supply and service system may be https: / / gongfu.example.com / login, the URL of the distribution system may be https: / / peizi.example.com / login, and the URL of the leakage protection system may be https: / / loubao.example.com / login.

[0069] When a user opens the login page, the system automatically obtains the URL of the current page and matches it with the preset URL in the database. Through this matching, the system can quickly determine the system type corresponding to the current page and load the corresponding recognition template for subsequent operations. The advantage of this method is that it is simple and efficient, and is particularly suitable for systems with clear and fixed URLs.

[0070] Of course, in addition to matching by URL, you can also use image recognition methods based on page features to determine the system type. This method analyzes the visual features of the login page (such as page layout, logo, color scheme, etc.) and compares them with the preset templates in the database. The specific implementation can use computer vision technology, such as template matching, feature point matching, or convolutional neural network (CNN). For example, when the system detects a specific logo icon, it can be determined that the page belongs to a specific distribution network operation system.

[0071] The system type can also be determined based on text analysis of web page content. The system can extract keywords or specific identifiers (such as page titles, description text) by parsing the web page source code or the text content on the page. This information can be matched with preset keywords or identifiers in the database to determine the system type. For example, if the page title contains "supply and service system", the system can determine that the page belongs to the supply and service system. This method is suitable for situations where there are clear identification or specific text content on the page, but its applicability may be limited for systems with dynamically generated content or multi-language support.

[0072] In some cases, the system can identify the system to which the current login page belongs through Cookies or Session information stored in the browser. Cookies and Session are mechanisms used by web servers to track and identify user status. By reading this information, the system can understand the systems or pages that the user has visited before, and thus infer the current system type. For example, if a Cookie indicates that the user has previously visited a self-configured system, the system can infer that the current login page may belong to a self-configured system. This method is suitable for situations where there is a shared session mechanism between multiple systems, but its accuracy depends on the validity and update of Cookies or Session.

[0073] Embodiment 3

[0074] The key elements include the username input box, password input box, verification code input box, verification code box, and login button. The methods for determining the positions of the key elements of the current login page include:

[0075] Downsample the current login interface and the recognition template at different scales to form multi-scale pyramids, respectively, to obtain the target image and template image at multiple scales;

[0076] Calculate the gradients of the target image and template image at each scale respectively;

[0077] The regularization term is determined by calculating the gradient difference between the target image and the template image at each scale;

[0078] Combine pixel value differences, gradient differences, and regularization terms to calculate the basic matching model between the target image and the template image at each scale;

[0079] Calculate the gradient direction consistency between the target image and the template image at each scale;

[0080] Combine the basic matching model and gradient direction consistency check to determine the best matching function at each scale;

[0081] By weighted fusion of the best matching functions at all scales, the best matching position of the key element is determined, which is the key element position.

[0082] The methods for determining the positions of key elements of the current login page specifically include:

[0083] Downsample the current login interface at different scales to form a multi-scale pyramid and obtain the Target image at scale The image of the login page is downsampled at different scales (i.e., the resolution is reduced) to form a multi-scale pyramid structure. The multi-scale pyramid generates a series of images of different resolutions by downsampling layer by layer, which is convenient for matching and analysis at different scales.

[0084] The recognition template is downsampled at different scales to form a multi-scale pyramid and obtain the Template image at scale ; Similarly, the preset recognition template image is downsampled at multiple scales to form a template image pyramid corresponding to the target image pyramid.

[0085] Calculate the target image in Gradient at scale ; Calculate the template image in the Gradient at scale ; Gradient represents the rate of change of pixel values ​​in an image and is usually used to capture edges or features in an image. In actual calculations, filters such as the Sobel operator and the Prewitt operator are usually used to perform convolution operations on the image to obtain the gradient of the image.

[0086] Determine the regularization term at each scale , ,in, For the target image Scale at location The gradient value at For the template image in Under scale The gradient value of the position, and Respectively The width and height of the template image at scale, is the gradient weight coefficient of the regularization term, and is the index value; the regularization term is used to prevent overfitting problems in the matching process and ensure the smoothness and stability of the matching results.

[0087] Determine the basic matching model at each scale , ,in, For the target image Scale at location The pixel value at For the template image in Under scale The pixel value of the position, is the gradient weight coefficient of the basic matching function, is the weight coefficient of the regularization term, and is the index value; the basic matching model combines the differences in pixel values ​​and gradients between the target image and the template image, while taking into account the regularization term. This model determines the best matching position by minimizing the difference between the target image and the template image.

[0088] Determine the consistency check model based on the gradient direction at each scale , ,in, For the target image The gradient direction at scale, For the template image in The gradient direction at scale, and is the index value; the gradient direction consistency check model is used to evaluate the consistency of the target image and the template image in the gradient direction. The gradient direction represents the direction of the gradient vector in the image. By calculating the cosine similarity between the gradient directions of the target image and the template image, their matching degree is determined.

[0089] Determine the best matching function at each scale , ,in, is the weight coefficient of the gradient direction consistency measurement; the best matching function is based on the basic matching model and further optimizes the matching result by combining the gradient direction consistency check model.

[0090] Get the best matching position of key elements , ,in, is the total number of scales, For the The weight coefficient of the scale, argmin is the minimum function. The best matching position of the key element is determined by weighted fusion of the best matching functions at all scales.

[0091] Embodiment 4

[0092] Determine whether there is a verification code input box and a verification code box according to the position of the key element. If there is a verification code box, identify and obtain the verification code;

[0093] After determining the location of each key element in the login page, the system needs to determine whether there is a verification code input box and a verification code box in the page. This is achieved by checking whether there is a specific identifier or element tag related to the verification code in the page. If the existence of the verification code box is detected, the system will start the recognition module and extract the content of the verification code through image recognition or OCR (optical character recognition) technology. The result of verification code recognition will be used for subsequent automatic filling operations to ensure that the login operation can be completed smoothly.

[0094] It is also possible to perform CAPTCHA recognition based on deep learning, especially convolutional neural networks (CNNs), which perform well in image recognition tasks. A deep learning model specifically for recognizing CAPTCHAs can be trained with a large amount of labeled CAPTCHA data, so that the model can automatically learn the features of CAPTCHAs and recognize them.

[0095] Verification code recognition can also be performed based on image segmentation. Image segmentation technology can be used to segment the characters in the verification code image one by one, and then perform single character recognition. This method is suitable for verification codes with multiple characters. First, the verification code image is grayed and binarized to remove background noise, and then each character in the verification code is segmented into an independent small image using an image segmentation algorithm (such as vertical projection, contour detection, etc.). Each segmented character is then recognized separately. Character recognition can be combined with template matching or machine learning algorithms. Finally, the recognized characters are combined to generate the final verification code text.

[0096] Verification codes can also be recognized based on joint denoising. For verification codes with more noise or complex backgrounds, joint denoising and recognition technology can be used. First, the noise in the image is cleaned up by a denoising algorithm, and then combined with OCR or deep learning models for recognition. That is, image denoising algorithms, such as median filtering and Gaussian filtering, are used to remove noise from the verification code image; the contrast or sharpness of the verification code image is enhanced to improve the recognizability of the characters, and then OCR or deep learning models are applied to the denoised image for verification code recognition. Finally, the recognized characters are combined to obtain the verification code text.

[0097] Methods for obtaining the multi-factor response functions corresponding to each key element include:

[0098] Collect the current system state vector ,in, For the The system state vector contains multiple factors that affect input operations, which may include network latency, system load, page response time, etc. Each state factor represents a specific influencing factor, and the system uses this state information as input to help adjust subsequent input strategies.

[0099] Through the key element position and system state vector, the multi-factor response function corresponding to each key element is obtained ,in, Enter the position coordinates of the key element input box to determine the position of the element on the page; The base level offset for the adjusted response function for the key element input box represents the base response value when there is no effect of system status. For key element input box The weight of the direction position, For key element input box The weight of the direction position; For the Impact factors; is the index value; , , Is the key element input box position and The coefficients of the second-order and cross terms of is a nonlinear activation function; It is the hyperbolic tangent function of the state factor, which shows the regulatory effect of the system state on the response function. is the first Impact Factor The weight coefficient of .

[0100] Through the above calculations, the system can dynamically generate a multi-factor response function based on the position of key elements and the current system status. This function will directly affect the way subsequent input operations are performed, including input speed, input order, etc. Ultimately, using this function, the system can flexibly adjust the automatic login process in different environments and states to ensure the success rate and stability of the login operation.

[0101] Embodiment 5

[0102] like Figure 2 As shown, the method for dynamically adjusting the input of each key element based on the multi-factor response function includes:

[0103] The response function value of each key element input box is calculated through a multi-factor response function, and the response function values ​​are arranged in ascending order; the response function value reflects the difficulty or priority of operating a specific input box in the current environment. By calculating the response function value of each key element and arranging these values ​​in ascending order, the system can determine the most suitable element for priority processing to ensure the accuracy and efficiency of the operation.

[0104] Each key element input box is processed in sequence according to the queue. The processing method includes:

[0105] According to the attributes of the key element input box (such as the username input box and the password input box), the corresponding string is obtained from the database; if it is a verification code input box, the corresponding verification code is identified from the verification code box.

[0106] Determine the relationship between the response function value of the key element input box and the set threshold; if the response function value is greater than the set threshold, it means that the current environment allows fast input, then the string or verification code is pasted for input, and the entire string or verification code is entered into the corresponding input box at one time; if the response function value is not greater than the set threshold, it means that the network condition is poor or the system load is high, then the string or verification code is input word by word to reduce the risk of input errors.

[0107] If you use word-by-word input, set the input interval time for each character ,in, The basic input interval time, which means the time interval between each character input under ideal conditions. is a multi-factor response function, To adjust the factor, control the intensity of the effect of the response function value on the input interval time. It is a small positive number to correct the division by zero error, which is used to prevent the denominator from being zero. Therefore, the input interval time will be dynamically adjusted according to the current system status. When the response function value is low (that is, the operating environment is poor), the input interval time will increase, thereby reducing the possibility of input errors.

[0108] After completing the input of the string or verification code, read the content in the key element input box and compare it with the expected characters. If an error occurs, delete the input content and recalculate the response function value before inputting again.

[0109] If the content in the key element input box is the same as the expected characters, the key element input is completed.

[0110] The system can dynamically adjust the input method of each key element according to the current environmental conditions, thereby ensuring that the automatic login operation can be completed efficiently and accurately under various complex system conditions.

[0111] Embodiment 6

[0112] If a login failure is detected according to the status code, the account and password information in the standby database is called to determine whether it is the same as the account and password information in the database. If they are not the same, the account and password information in the standby database is used to replace the account and password information in the database.

[0113] When a login failure is detected based on the status code, the system first calls the account and password information in the standby database. The standby database usually stores a second set of credentials for login, which may be used to deal with the situation where the primary database information is invalid or cannot log in for other reasons. The system compares the account and password in the standby database with the information in the primary database. If the account and password in the standby database are different from those in the primary database, it means that the information in the primary database may have expired or is no longer valid. At this time, the system replaces the information in the primary database with the information in the standby database to ensure that the latest or valid credentials are used the next time you try to log in.

[0114] Wait for a random time interval before trying to log in again; after replacing the account and password information, the system will not immediately retry to log in, but will wait for a random time interval. The purpose is to avoid continuous login attempts in a short period of time that may cause system locks or other security measures. Random time intervals also help reduce the risk of the system detecting automated behavior. The random time interval is based on a random number within a certain range (for example, a few seconds to tens of seconds).

[0115] If the attempt still fails after the set number of times, the recognition template in the standby database will be called, and after replacing the recognition template in the database with the recognition template in the standby database, the system will replace the template in the main database with the recognition template in the standby database. This operation is usually to deal with the situation where the page structure has changed (such as page upgrade or revision), and the original template is no longer applicable. After the replacement, the system will try to log in again using the new template and account password information, that is, log in again using the account password information in the database and the account password information in the standby database in turn.

[0116] If the login fails after the set number of attempts, the login operation will be suspended and human intervention will be notified. If the login fails after multiple attempts using all available account password information and recognition templates, the system will suspend the automatic login operation to prevent the system from falling into an infinite retry loop. The system will then notify the administrator or relevant operator to intervene manually. Notifications can be sent in a variety of ways, such as email, SMS or system warnings.

[0117] Embodiment 7

[0118] An automatic login terminal suitable for distribution network operation analysis includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the automatic login method suitable for distribution network operation analysis is implemented.

[0119] The memory can be used to store software programs and modules. The processor executes various functional applications and data processing of the terminal by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an execution program required for at least one function, etc.

[0120] The data storage area can store data created according to the use of the terminal, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0121] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned automatic login method applicable to distribution network operation analysis.

[0122] Without loss of generality, computer readable media may include computer storage media and communication media. Computer storage media include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information such as computer readable instruction data structures, program modules or other data. Computer storage media include RAM, ROM, EPROM, EEPROM, flash memory or other solid-state storage technology, CD-ROM, DVD or other optical storage, cassettes, magnetic tapes, disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that computer storage media are not limited to the above. The above-mentioned system memory and mass storage devices can be collectively referred to as memory.

[0123] A computer program product includes a computer program or instructions, which, when executed by a processor, implements the above automatic login method applicable to distribution network operation analysis.

[0124] A computer program product includes a computer program or set of instructions for performing specific tasks or implementing specific functions. These programs or instructions are designed to be executed by a processor to implement a series of predefined steps or operations. The program product may be stored in various forms of computer storage media, such as memory, hard disk, solid-state drive, optical disk or other forms of digital storage devices. It may exist in the form of compiled binary code or in the form of scripts or bytecodes that can be executed by an interpreter. The program product uses carefully designed algorithms and logical instructions to enable the processor to process data in a specific order and manner to complete various functions such as data analysis, user interaction, device control, etc.

[0125] In the description of this specification, the description with reference to the terms "one embodiment / method", "some embodiments / methods", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment / method or example are included in at least one embodiment / method or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment / method or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments / methods or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments / methods or examples described in this specification and the features of the different embodiments / methods or examples, unless they are contradictory.

[0126] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0127] It should be understood by those skilled in the art that the above embodiments are only for the purpose of clearly illustrating the present invention, and are not intended to limit the scope of the present invention. For those skilled in the art, other changes or modifications may be made based on the above invention, and these changes or modifications are still within the scope of the present invention.

Claims

1. An automatic login method suitable for distribution network operation analysis, characterized in that: include: Build a database that contains identification templates, account and password information for each system; Determine the system type corresponding to the current login page, load the corresponding recognition template for matching, and determine the location of the key elements of the current login page; According to the key element positions and the system state vector, the multi-factor response function corresponding to each key element is obtained; Dynamically adjust the input of each key element based on a multi-factor response function; After completing the input, submit the login command. The system returns a status code and determines whether the login is successful based on the status code. If the login is unsuccessful, log in again; Among them, the key elements include a user name input box, a password input box, a verification code input box, a verification code box, and a login button. The method for determining the position of the key elements of the current login page includes: Downsample the current login interface and the recognition template at different scales to form multi-scale pyramids, respectively, to obtain the target image and template image at multiple scales; Calculate the gradients of the target image and template image at each scale respectively; The regularization term is determined by calculating the gradient difference between the target image and the template image at each scale; Combine pixel value differences, gradient differences, and regularization terms to calculate the basic matching model between the target image and the template image at each scale; Calculate the gradient direction consistency between the target image and the template image at each scale; Combine the basic matching model and gradient direction consistency check to determine the best matching function at each scale; By weighted fusion of the best matching functions at all scales, the best matching position of the key element is determined, which is the key element position.

2. The automatic login method for distribution network operation analysis according to claim 1, characterized in that: The method for determining the system type corresponding to the current login page is: presetting the URL of each system in the database, obtaining the URL of the current login interface when logging in, and matching through the URL to determine the system type.

3. The automatic login method suitable for distribution network operation analysis according to claim 1, characterized in that: The methods for determining the positions of key elements of the current login page specifically include: Downsample the current login interface at different scales to form a multi-scale pyramid and obtain the Target image at scale ; The recognition template is downsampled at different scales to form a multi-scale pyramid and obtain the Template image at scale ; Calculate the target image in Gradient at scale ; Calculate the template image in the Gradient at scale ; Determine the regularization term at each scale , ,in, For the target image Scale at location The gradient value at For the template image in Under scale The gradient value of the position, and Respectively The width and height of the template image at scale, is the gradient weight coefficient of the regularization term, and is the index value; Determine the basic matching model at each scale , ,in, For the target image Scale at location The pixel value at For the template image in Under scale The pixel value of the position, is the gradient weight coefficient of the basic matching function, is the weight coefficient of the regularization term, and is the index value; Determine the consistency check model based on the gradient direction at each scale , ,in, For the target image The gradient direction at scale, For the template image in The gradient direction at scale, and is the index value; Determine the best matching function at each scale , ,in, is the weight coefficient of the gradient direction consistency measure; Get the best matching position of key elements , ,in, is the total number of scales, For the The weight coefficient of the scale, argmin is the minimum function.

4. The automatic login method suitable for distribution network operation analysis according to claim 1, characterized in that: Determine whether there is a verification code input box and a verification code box according to the position of the key element. If there is a verification code box, identify and obtain the verification code; Methods for obtaining the multi-factor response functions corresponding to each key element include: Collect the current system state vector ,in, For the Impact factors; Through the key element position and system state vector, the multi-factor response function corresponding to each key element is obtained ,in, Enter the position coordinates of the key element input box. Bias for the base level of the adjusted response function for the key element input box, For key element input box The weight of the direction position, For key element input box The weight of the direction position, For the Impact factors, is the index value, , , Is the key element input box position and The coefficients of the second-order and cross terms of is a nonlinear activation function, is the first Impact Factor The weight coefficient of .

5. The automatic login method suitable for distribution network operation analysis according to claim 1, characterized in that: Methods for dynamically adjusting the input of each key element based on a multi-factor response function include: Calculate the response function value of each key element input box through the multi-factor response function, and arrange the response function values ​​in ascending order; Each key element input box is processed in sequence according to the queue. The processing method includes: Obtain the corresponding string from the database according to the attributes of the key element input box; if it is a verification code input box, identify the corresponding verification code from the verification code box; Determine the size relationship between the response function value of the key element input box and the set threshold; if the response function value is greater than the set threshold, paste input is used for the string or verification code; if the response function value is not greater than the set threshold, word-by-word input is used for the string or verification code; If you use word-by-word input, set the input interval time for each character ,in, Enter the interval time as the basis, is a multi-factor response function, is the adjustment factor, Correct positive numbers for division by zero errors; After completing the input of the string or verification code, read the content in the key element input box and compare it with the expected characters. If an error occurs, delete the input content and recalculate the response function value before inputting again. If the content in the key element input box is the same as the expected characters, the key element input is completed.

6. The automatic login method suitable for distribution network operation analysis according to claim 1, characterized in that: If a login failure is detected according to the status code, the account and password information in the standby database is called to determine whether it is the same as the account and password information in the database. If they are not the same, the account and password information in the standby database is used to replace the account and password information in the database. Wait for a random time interval and try to log in again; If the attempt fails after the set number of times, the recognition template in the standby database is called, and the recognition template in the standby database is used to replace the recognition template in the database, and the account and password information in the database and the account and password information in the standby database are used to log in again; If the login fails after the set number of attempts, the login operation will be suspended and human intervention will be notified.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the automatic login method suitable for distribution network operation analysis as described in any one of claims 1 to 6 is implemented.

8. A computer program product comprising a computer program or instructions, characterized in that When the computer program / instruction is executed by the processor, the automatic login method suitable for distribution network operation analysis as described in any one of claims 1 to 6 is implemented.

Images