A user account management method and system based on stroke order, OCR and artificial intelligence

Abstract

The present invention provides a user account management method and system based on stroke order, OCR and artificial intelligence, wherein the method comprises: upon receiving a registration instruction from a first user, entering a registration mode; obtaining user information of the first user, the user information comprising: a combination of one or more of a user name, a registered place of residence, and a registered place of school; generating a login name of the user account based on the user information; obtaining the first handwriting of the user inputting the login name multiple times; and determining a first model as login verification of the user account based on the first handwriting. The user account management method based on stroke order, OCR and artificial intelligence of the present invention realizes that the user does not need to remember the login name and login password, and can log in to a backend server and use the services provided by the backend server according to the user's daily behavior habits.

Description

Technical Field

[0001] The present invention relates to the technical field of database management, and in particular to a user account management method and system based on stroke order, OCR and artificial intelligence. Background Art

[0002] At present, with the rapid development of network information technology, there are more and more network platforms and APPs that provide users with various services. For the privacy and security of user accounts, users are usually required to register accounts. When registering accounts, they need to set account information such as login name and login password, and log in to the backend server based on the registered account to use the services provided by the backend server;

[0003] The management of login names and login passwords depends on the user's memory. In daily life, users need to use many services. Faced with numerous accounts, if a unified login name and login password are set, the security is worrying. If different login names and login passwords are set, it is easy to cause the login name and login password to be forgotten. Therefore, an account management method is urgently needed to achieve that users do not need to remember login names and login passwords, and can log in to the backend server according to their daily behavior habits and use the services provided by the backend server. Summary of the invention

[0004] One of the purposes of the present invention is to provide a user account management method based on stroke order, OCR and artificial intelligence, so that users do not need to remember login names and login passwords, and can log in to a backend server and use the services provided by the backend server according to their daily behavior habits.

[0005] An embodiment of the present invention provides a user account management method based on stroke order, OCR and artificial intelligence, comprising:

[0006] When receiving a registration instruction from the first user, entering a registration mode;

[0007] Obtaining user information of the first user, the user information including: one or more combinations of user name, registered residence, and registered school location;

[0008] Generate a login name for the user account based on the user information;

[0009] Obtain the first handwriting of the user's login name multiple times;

[0010] Based on the first handwriting, determining a first model as a login verification of the user account;

[0011] Calculate the hash value of the first model and save the hash value on the chain.

[0012] Preferably, the user account management method based on stroke order, OCR and artificial intelligence also includes:

[0013] Get the login sign-in information of the user account;

[0014] Configure the first points for the user account according to the login sign-in information and the preset login sign-in point accumulation rules;

[0015] and / or,

[0016] Get original resource information published by user accounts;

[0017] Allocate a second point for the user account according to the original resource information and the preset original resource accumulation rules;

[0018] and / or,

[0019] Get resource processing information of user account;

[0020] Allocate third points for user accounts based on resource processing information and preset resource processing points accumulation rules;

[0021] and / or,

[0022] Get resource sharing information of user accounts;

[0023] Allocate fourth points for user accounts according to resource sharing information and preset resource sharing points rules;

[0024] and / or,

[0025] Get information and comments of user accounts;

[0026] The fifth point is configured for the user account according to the information comment information and the preset information comment point accumulation rules.

[0027] Preferably, the user account management method based on stroke order, OCR and artificial intelligence also includes:

[0028] Obtain a user list input by a second user with high-level permissions;

[0029] Based on the user list, multiple initial accounts are opened; the initial accounts use the name of the user list and the user name in the user list as the login name;

[0030] Receiving a first login operation performed by a third user corresponding to the initial account using the initial account;

[0031] When the third user performs the first login operation, the handwriting collection interface is output multiple times;

[0032] collecting a second handwriting of a third user inputting a login name multiple times through the handwriting collection interface;

[0033] Based on the second handwriting, determining a second model as login verification of the initial account;

[0034] Receiving a processing operation of the third user on the user information of the initial account, the processing operation including: modifying, inputting, deleting or one or a combination thereof;

[0035] Receiving an operation of logging out of the first login by the third user, and saving the initial account as the user account;

[0036] The receiving step of performing a first login operation using the initial account by a third user corresponding to the initial account includes:

[0037] Acquire a second handwriting input by a third user;

[0038] Performing OCR recognition on the second handwriting to obtain recognition information;

[0039] The identification information is matched with the login name of the initial account, and when the match is confirmed, the first login of the initial account is completed.

[0040] Preferably, based on the first handwriting, determining a first model for login verification of the user account includes:

[0041] Parse the first handwriting to obtain stroke order information; the stroke order information includes: a stroke type, an order between stroke types, and parameter information of each stroke, or one or more combinations thereof; the parameter information of each stroke includes: a position relationship parameter between a starting point, an end point, and a preset number of sampling points between the starting point and the end point of each stroke, and / or a handwriting strength parameter of a preset number of sampling points between the starting point, the end point, and the starting point and the end point of each stroke;

[0042] Extracting characteristic values ​​of the stroke order information, inputting the characteristic values ​​into a preset initial neural network model for training, and obtaining a first model for login verification of a user account after the initial neural network model converges;

[0043] or,

[0044] Extract the eigenvalues ​​of the stroke order information and construct a verification vector based on the eigenvalues;

[0045] Building a login verification library based on the verification vector, and using the login verification library as the first model;

[0046] Among them, when the login verification library is used as the first model for login verification, the login verification steps are as follows:

[0047] Obtaining the third handwriting to be verified;

[0048] extracting a feature value of the third handwriting, and constructing a handwriting vector based on the extracted feature value;

[0049] Calculate the similarity between the handwriting vector and the verification vector. The calculation formula is as follows:

[0050]

[0051] Among them, X i is the similarity between the handwriting vector and the i-th verification vector in the login verification library; n is the dimension of the handwriting vector and the verification vector; a i,j is the j-th dimension data value of the i-th verification vector; b j is the j-th dimension data value of the handwriting vector;

[0052] When the similarity between the handwriting vector and all verification vectors in the login verification library is greater than or equal to a preset first threshold and the maximum value of the similarities is greater than a preset second threshold, the login verification is passed.

[0053] Preferably, the user account management method based on stroke order, OCR and artificial intelligence also includes:

[0054] When obtaining first account transfer instruction information input by a third user corresponding to a user account of the third user, sending a first inquiry to a second user associated with the third user;

[0055] receiving a first feedback from a second user regarding the first inquiry;

[0056] When the first feedback is passed, the first account transfer instruction information is parsed to obtain the source data platform of the account transfer;

[0057] Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission;

[0058] Adding the first feedback to the end of the first account transfer instruction information to generate a second query, and sending the second query to a fourth user;

[0059] receiving a second feedback from a fourth user regarding the second inquiry;

[0060] When the second feedback is passed, the account transfer instruction information is parsed again to obtain the target data platform for the account transfer;

[0061] Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission;

[0062] Adding the second feedback to the end of the first account transfer instruction information to generate a third query, and sending the third query to the fifth user;

[0063] receiving a third feedback from a fifth user regarding the third inquiry;

[0064] When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform;

[0065] or,

[0066] When obtaining the second account transfer instruction information input by the second user corresponding to the user account of the third user, parsing the first account transfer instruction information, and obtaining the source data platform of the account transfer;

[0067] Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission;

[0068] generating a second inquiry based on the second account transfer instruction information, and sending the second inquiry to a fourth user;

[0069] receiving a second feedback from a fourth user regarding the second inquiry;

[0070] When the second feedback is passed, the account transfer instruction information is parsed again to obtain the target data platform for the account transfer;

[0071] Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission;

[0072] Adding the second feedback to the end of the first account transfer instruction information to generate a third query, and sending the third query to the fifth user;

[0073] receiving a third feedback from a fifth user regarding the third inquiry;

[0074] When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform.

[0075] The present invention also provides a user account management system based on stroke order, OCR and artificial intelligence, comprising:

[0076] Registration module, the registration module performs the following operations:

[0077] When receiving a registration instruction from the first user, entering a registration mode;

[0078] Obtaining user information of the first user, the user information including: one or more combinations of user name, registered residence, and registered school location;

[0079] Generate a login name for the user account based on the user information;

[0080] Obtain the first handwriting of the user's login name multiple times;

[0081] Based on the first handwriting, determining a first model as a login verification of the user account;

[0082] Calculate the hash value of the first model and save the hash value on the chain.

[0083] Preferably, the user account management system based on stroke order, OCR and artificial intelligence further includes: a points management module, which performs the following operations:

[0084] Get the login sign-in information of the user account;

[0085] Configure the first points for the user account according to the login sign-in information and the preset login sign-in point accumulation rules;

[0086] and / or,

[0087] Get original resource information published by user accounts;

[0088] Allocate a second point for the user account according to the original resource information and the preset original resource accumulation rules;

[0089] and / or,

[0090] Get resource processing information of user account;

[0091] Allocate third points for user accounts based on resource processing information and preset resource processing points accumulation rules;

[0092] and / or,

[0093] Get resource sharing information of user accounts;

[0094] Allocate fourth points for user accounts according to resource sharing information and preset resource sharing points rules;

[0095] and / or,

[0096] Get information and comments of user accounts;

[0097] The fifth point is configured for the user account according to the information comment information and the preset information comment point accumulation rules.

[0098] Preferably, the registration module also performs the following operations:

[0099] Obtain a user list input by a second user with high-level permissions;

[0100] Based on the user list, multiple initial accounts are opened; the initial accounts use the name of the user list and the user name in the user list as the login name;

[0101] Receiving a first login operation performed by a third user corresponding to the initial account using the initial account;

[0102] When the third user performs the first login operation, the handwriting collection interface is output multiple times;

[0103] collecting a second handwriting of a third user inputting a login name multiple times through the handwriting collection interface;

[0104] Based on the second handwriting, determining a second model as login verification of the initial account;

[0105] Receiving a processing operation of the third user on the user information of the initial account, the processing operation including: modifying, inputting, deleting or one or a combination thereof;

[0106] Receiving an operation of logging out of the first login by the third user, and saving the initial account as the user account;

[0107] The receiving step of performing a first login operation using the initial account by a third user corresponding to the initial account includes:

[0108] Acquire a second handwriting input by a third user;

[0109] Performing OCR recognition on the second handwriting to obtain recognition information;

[0110] The identification information is matched with the login name of the initial account, and when the match is confirmed, the first login of the initial account is completed.

[0111] Preferably, based on the first handwriting, determining a first model for login verification of the user account includes:

[0112] Parse the first handwriting to obtain stroke order information; the stroke order information includes: a stroke type, an order between stroke types, and parameter information of each stroke, or one or more combinations thereof; the parameter information of each stroke includes: a position relationship parameter between a starting point, an end point, and a preset number of sampling points between the starting point and the end point of each stroke, and / or a handwriting strength parameter of a preset number of sampling points between the starting point, the end point, and the starting point and the end point of each stroke;

[0113] Extracting characteristic values ​​of the stroke order information, inputting the characteristic values ​​into a preset initial neural network model for training, and obtaining a first model for login verification of a user account after the initial neural network model converges;

[0114] or,

[0115] Extract the eigenvalues ​​of the stroke order information and construct a verification vector based on the eigenvalues;

[0116] Building a login verification library based on the verification vector, and using the login verification library as the first model;

[0117] Among them, when the login verification library is used as the first model for login verification, the login verification steps are as follows:

[0118] Obtaining the third handwriting to be verified;

[0119] extracting a feature value of the third handwriting, and constructing a handwriting vector based on the extracted feature value;

[0120] Calculate the similarity between the handwriting vector and the verification vector. The calculation formula is as follows:

[0121]

[0122] Among them, X i is the similarity between the handwriting vector and the i-th verification vector in the login verification library; n is the dimension of the handwriting vector and the verification vector; a i,j is the j-th dimension data value of the i-th validation vector; b j is the j-th dimension data value of the handwriting vector;

[0123] When the similarity between the handwriting vector and all verification vectors in the login verification library is greater than or equal to a preset first threshold and the maximum value of the similarities is greater than a preset second threshold, the login verification is passed.

[0124] Preferably, the user account management system based on stroke order, OCR and artificial intelligence further includes: an account transfer module, which performs the following operations:

[0125] When obtaining first account transfer instruction information input by a third user corresponding to a user account of the third user, sending a first inquiry to a second user associated with the third user;

[0126] receiving a first feedback from a second user regarding the first inquiry;

[0127] When the first feedback is passed, the first account transfer instruction information is parsed to obtain the source data platform of the account transfer;

[0128] Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission;

[0129] Adding the first feedback to the end of the first account transfer instruction information to generate a second query, and sending the second query to a fourth user;

[0130] receiving a second feedback from a fourth user regarding the second inquiry;

[0131] When the second feedback is passed, the account transfer instruction information is parsed again to obtain the target data platform for the account transfer;

[0132] Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission;

[0133] Adding the second feedback to the end of the first account transfer instruction information to generate a third query, and sending the third query to the fifth user;

[0134] receiving a third feedback from a fifth user regarding the third inquiry;

[0135] When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform;

[0136] or,

[0137] When obtaining the second account transfer instruction information input by the second user corresponding to the user account of the third user, parsing the first account transfer instruction information, and obtaining the source data platform of the account transfer;

[0138] Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission;

[0139] generating a second inquiry based on the second account transfer instruction information, and sending the second inquiry to a fourth user;

[0140] receiving a second feedback from a fourth user regarding the second inquiry;

[0141] When the second feedback is passed, the account transfer instruction information is parsed again to obtain the target data platform for the account transfer;

[0142] Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission;

[0143] Adding the second feedback to the end of the first account transfer instruction information to generate a third query, and sending the third query to the fifth user;

[0144] receiving a third feedback from a fifth user regarding the third inquiry;

[0145] When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform.

[0146] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0147] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0148] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0149] Figure 1 A schematic diagram of a user account management method based on stroke order, OCR and artificial intelligence in an embodiment of the present invention;

[0150] Figure 2 Schematic diagram of another user account management method based on stroke order, OCR and artificial intelligence in an embodiment of the present invention;

[0151] Figure 3 Schematic diagram of a user account management system based on stroke order, OCR and artificial intelligence in an embodiment of the present invention. DETAILED DESCRIPTION

[0152] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0153] The embodiment of the present invention provides a user account management method based on stroke order, OCR and artificial intelligence, such as Figure 1 As shown, including:

[0154] Step S1: upon receiving a registration instruction from the first user, entering a registration mode;

[0155] Step S2: obtaining user information of the first user, the user information including: one or more combinations of user name, registered residence, and registered school location;

[0156] Step S3: Generate a login name for the user account based on the user information;

[0157] Step S4: obtaining the first handwriting of the user inputting the login name multiple times;

[0158] Step S5: Based on the first handwriting, determine a first model for login verification of the user account;

[0159] Step S6: Calculate the hash value of the first model and save the hash value on the chain.

[0160] The working principle and beneficial effects of the above technical solution are:

[0161] This application uses a method of generating a login name based on user information when registering a user account. For example, the user's place of origin + user name can be used as the login name. The handwriting of the user inputting the login name is used to establish the first login verification model. When the user logs in, the user only needs to write the login name, and the backend server will judge and verify it based on the handwriting written by the user. When the verification is passed, the user account is successfully logged in; the login name generation rules are prompted on the login interface, so that the user does not need to remember the login name, and can know his login name according to the prompt; the handwriting written by the user is used as the login password, and there is no need to remember the password; the user does not need to remember the login name and login password, and can log in to the backend server according to the user's daily behavior habits, and use the services provided by the backend server, thereby improving the user's service experience. By obtaining the first handwriting of the user inputting the login name multiple times, multiple first handwritings are obtained, and the accuracy of the first model finally determined is guaranteed based on multiple first handwritings. The hash value of the first model is calculated and saved on the chain; the data security of the first model is guaranteed, that is, when the user account logs in, the hash value of the first model is obtained from the blockchain, and the first model is verified based on the hash value. When the first model is tampered with, it can be discovered and processed in time, further ensuring the security of the user account.

[0162] The user account management method based on stroke order, OCR and artificial intelligence of the present invention can log in to a backend server and use services provided by the backend server according to the user's daily behavior habits without the user having to remember a login name and a login password.

[0163] In one embodiment, the user account management method based on stroke order, OCR and artificial intelligence further includes:

[0164] Get the login sign-in information of the user account;

[0165] Configure the first points for the user account according to the login sign-in information and the preset login sign-in point accumulation rules;

[0166] and / or,

[0167] Get original resource information published by user accounts;

[0168] Allocate a second point for the user account according to the original resource information and the preset original resource accumulation rules;

[0169] and / or,

[0170] Get resource processing information of user account;

[0171] Allocate third points for user accounts based on resource processing information and preset resource processing points accumulation rules;

[0172] and / or,

[0173] Get resource sharing information of user accounts;

[0174] Allocate fourth points for user accounts according to resource sharing information and preset resource sharing points rules;

[0175] and / or,

[0176] Get information and comments of user accounts;

[0177] The fifth point is configured for the user account according to the information comment information and the preset information comment point accumulation rules.

[0178] The working principle and beneficial effects of the above technical solution are:

[0179] All users are provided with two virtual currency systems: learning coins and learning points. System administrators can configure users to obtain corresponding points through daily login sign-in, resource creation, resource processing, resource sharing, information commenting, etc. At the same time, points can be used to view specific information and download high-quality original resources. The details are as follows: according to the login sign-in information of the user account, the first point is configured for the user. For example, if the user signs in, one point is obtained on the first day, two points are obtained by continuing to sign in on the second day, and so on, and seven points are obtained on the seventh day; if the continuous sign-in is interrupted, it will be restored to one point per day; if you do not sign in after logging in on the same day, you can make up for it within the preset time (for example, two days). According to the original resource information published by the user, points can be accumulated, and the corresponding points can be configured according to the click situation; the corresponding points can be configured according to the user's resource processing situation; the corresponding points can be configured according to the resource sharing situation; and the corresponding points can be configured according to the information comment situation. In addition, learning coins are allocated according to the user's recharge situation; learning points and learning coins can be exchanged with each other according to preset rules.

[0180] In one embodiment, a user account management method based on stroke order, OCR and artificial intelligence, such as Figure 2 As shown, it also includes:

[0181] Step S11: obtaining a user list input by a second user with high-level authority;

[0182] Step S12: Based on the user list, multiple initial accounts are opened; the initial accounts use the name of the user list and the user name in the user list as login names;

[0183] Step S13: receiving a first login operation performed by a third user corresponding to the initial account using the initial account;

[0184] Step S14: when the third user performs the first login operation, the handwriting collection interface is output multiple times;

[0185] Step S15: collecting the second handwriting of the third user inputting the login name multiple times through the handwriting collection interface;

[0186] Step S16: Based on the second handwriting, determine a second model for login verification of the initial account;

[0187] Step S17: receiving a processing operation of the third user on the user information of the initial account, the processing operation including: modifying, inputting, deleting or one or a combination thereof;

[0188] Step S18: receiving the operation of the third user logging out of the first login, and saving the initial account as the user account;

[0189] The receiving step of performing a first login operation using the initial account by a third user corresponding to the initial account includes:

[0190] Acquire a second handwriting input by a third user;

[0191] Performing OCR recognition on the second handwriting to obtain recognition information;

[0192] The identification information is matched with the login name of the initial account, and when the match is confirmed, the first login of the initial account is completed.

[0193] The working principle and beneficial effects of the above technical solution are:

[0194] The second user with high-level permissions is the class teacher, and the user list is the class student list. Based on the user list, user accounts for each student (third user) can be opened in batches at the same time in units of classes, thereby increasing the rate of account opening; further, the user list can be filled with specific information of each user, and the information can also be supplemented when the user account logs in for the first time.

[0195] In one embodiment, based on the first handwriting, determining a first model as a login verification of a user account includes:

[0196] Parse the first handwriting to obtain stroke order information; the stroke order information includes: a stroke type, an order between stroke types, and parameter information of each stroke, or one or more combinations thereof; the parameter information of each stroke includes: a position relationship parameter between a starting point, an end point, and a preset number of sampling points between the starting point and the end point of each stroke, and / or a handwriting strength parameter of a preset number of sampling points between the starting point, the end point, and the starting point and the end point of each stroke;

[0197] Extracting characteristic values ​​of the stroke order information, inputting the characteristic values ​​into a preset initial neural network model for training, and obtaining a first model for login verification of a user account after the initial neural network model converges;

[0198] or,

[0199] Extract the eigenvalues ​​of the stroke order information and construct a verification vector based on the eigenvalues;

[0200] Building a login verification library based on the verification vector, and using the login verification library as the first model;

[0201] Among them, when the login verification library is used as the first model for login verification, the login verification steps are as follows:

[0202] Obtaining the third handwriting to be verified;

[0203] extracting a feature value of the third handwriting, and constructing a handwriting vector based on the extracted feature value;

[0204] Calculate the similarity between the handwriting vector and the verification vector. The calculation formula is as follows:

[0205]

[0206] Among them, X i is the similarity between the handwriting vector and the i-th verification vector in the login verification library; n is the dimension of the handwriting vector and the verification vector; a i,j is the j-th dimension data value of the i-th verification vector; b j is the j-th dimension data value of the handwriting vector;

[0207] When the similarity between the handwriting vector and all verification vectors in the login verification library is greater than or equal to a preset first threshold and the maximum value of the similarities is greater than a preset second threshold, the login verification is passed.

[0208] The working principle and beneficial effects of the above technical solution are:

[0209] The present embodiment provides two models for verifying the login of the user account. Model 1, a neural network model, is established in the following manner: using an initial neural network, the handwriting input by the user is machine-learned, and the handwriting of the user is accurately identified when logging in, thereby realizing verification. Model 2, a login verification library, analyzes the handwriting, determines the stroke order information, and then extracts features from the stroke order information, and establishes a verification vector based on the extracted feature values. Whether it is a neural network model or a login verification library, when writing the login name, it is also possible not to follow the standard, realize that the stroke order is used as a verification point according to user customization, improve the granularity of verification, and realize the operability of user customization. In addition, considering the handwriting strength parameters between the strokes, the recognition of the same stroke of the same word written by different people is realized, and the recognition of personnel based on the stroke order information and the identification information scheme is improved. Further, after each successful login, the handwriting of this login is used as material to update the first model.

[0210] In one embodiment, the user account management method based on stroke order, OCR and artificial intelligence further includes:

[0211] When obtaining first account transfer instruction information input by a third user corresponding to a user account of the third user, sending a first inquiry to a second user associated with the third user;

[0212] receiving a first feedback from a second user regarding the first inquiry;

[0213] When the first feedback is passed, the first account transfer instruction information is parsed to obtain the source data platform of the account transfer;

[0214] Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission;

[0215] Adding the first feedback to the end of the first account transfer instruction information to generate a second query, and sending the second query to a fourth user;

[0216] receiving a second feedback from a fourth user regarding the second inquiry;

[0217] When the second feedback is passed, the account transfer instruction information is parsed again to obtain the target data platform for the account transfer;

[0218] Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission;

[0219] Adding the second feedback to the end of the first account transfer instruction information to generate a third query, and sending the third query to the fifth user;

[0220] receiving a third feedback from a fifth user regarding the third inquiry;

[0221] When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform;

[0222] or,

[0223] When obtaining the second account transfer instruction information input by the second user corresponding to the user account of the third user, parsing the first account transfer instruction information, and obtaining the source data platform of the account transfer;

[0224] Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission;

[0225] generating a second inquiry based on the second account transfer instruction information, and sending the second inquiry to a fourth user;

[0226] receiving a second feedback from a fourth user regarding the second inquiry;

[0227] When the second feedback is passed, the account transfer instruction information is parsed again to obtain the target data platform for the account transfer;

[0228] Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission;

[0229] Adding the second feedback to the end of the first account transfer instruction information to generate a third query, and sending the third query to the fifth user;

[0230] receiving a third feedback from a fifth user regarding the third inquiry;

[0231] When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform.

[0232] The working principle and beneficial effects of the above technical solution are:

[0233] The specific application scenario of this embodiment is student transfer. The existing conventional operation is to re-open an account on the education platform of the place where the transfer takes place. To open an account, the information and password need to be reset, which is extremely inconvenient. Therefore, a transfer method is adopted. Through the application of the student (third user) or the class teacher in charge of the student (second user), with the consent of the principal (fourth user), the user account is packaged and transferred out of the source data platform, and after the consent of the principal (fifth user) of the receiving party, it is transferred to the target data platform. The tedious processing of account cancellation and re-application is avoided, and the account is transferred with the student after the student transfers; the place of belonging in the login name is also relatively changed after the transfer.

[0234] In one embodiment, the user account management method based on stroke order, OCR and artificial intelligence further includes:

[0235] Get the historical login information of the user account;

[0236] Analyze historical login information to obtain the number of days the user has logged in within a preset time period from the current time, the daily login time, and a list of the data in the database called when logging in every day;

[0237] Based on the data in the list, the preset call data and time comparison table, the daily login time is corrected to determine the effective time; specifically as follows: the total call time of the data in the call list is counted, and when the total call time is greater than or equal to the daily login time, the daily login time is the effective time; when the total call time is less than or equal to the daily login time, the total call time is the effective time;

[0238] The user's activity is determined based on the current user's effective time, the sum of all users' effective time, the current user's login days, and the number of days corresponding to the preset time period. The calculation formula is as follows:

[0239]

[0240] Among them, Hk is the activity of the kth user in the big data platform, D k is the number of days the kth user logged in; D0 is the number of days corresponding to the preset time period; B k is the effective time of the kth user; M is the number of users on the big data platform; ε1 and ε2 are preset weights;

[0241] Points are allocated to user accounts based on activity and preset activity allocation rules.

[0242] The working principle and beneficial effects of the above technical solution are:

[0243] The user activity is calculated based on the number of login days and the effective login time of each day to ensure the accuracy of the activity. In addition, the effective time is introduced to correct the login time through the data called each time the user logs in, eliminating the impact of the user's inactivity time on the grading, making the activity more accurate and objective. Points are allocated based on activity to increase users' enthusiasm for using the platform.

[0244] The present invention also provides a user account management system based on stroke order, OCR and artificial intelligence, such as Figure 3 As shown, including:

[0245] The registration module 11 performs the following operations:

[0246] When receiving a registration instruction from the first user, entering a registration mode;

[0247] Obtaining user information of the first user, the user information including: one or more combinations of user name, registered residence, and registered school location;

[0248] Generate a login name for the user account based on the user information;

[0249] Obtain the first handwriting of the user's login name multiple times;

[0250] Based on the first handwriting, determining a first model as a login verification of the user account;

[0251] Calculate the hash value of the first model and save the hash value on the chain.

[0252] The working principle and beneficial effects of the above technical solution are:

[0253] This application adopts the method of generating a login name based on user information when registering a user account. For example, the user's place of origin + user name can be used as the login name. The first login verification model is established by the handwriting of the user inputting the login name. When the user logs in, the user only needs to write the login name, and the backend server judges and verifies it according to the handwriting written by the user. When the verification is passed, the user account is successfully logged in; the login name generation rule is prompted on the login interface, so that the user does not need to remember the login name, and can know his login name according to the prompt; the handwriting written by oneself is used as the login password, and there is no need to remember the password; it is realized that the user does not need to remember the login name and login password, and can log in to the backend server according to the user's daily behavior habits, and use the services provided by the backend server, thereby improving the user's service experience. The hash value of the first model is calculated, and the hash value is saved on the chain; the data security of the first model is guaranteed, that is, when the user account logs in, the hash value of the first model is obtained from the blockchain, and the first model is verified based on the hash value. When the first model is tampered, it can be discovered and processed in time, further ensuring the security of the user account.

[0254] The user account management system based on stroke order, OCR and artificial intelligence of the present invention can log in to a backend server and use services provided by the backend server according to the user's daily behavior habits without the user having to remember a login name and a login password.

[0255] In one embodiment, the user account management system based on stroke order, OCR and artificial intelligence further includes: a point management module 12, which performs the following operations:

[0256] Get the login sign-in information of the user account;

[0257] Configure the first points for the user account according to the login sign-in information and the preset login sign-in point accumulation rules;

[0258] and / or,

[0259] Get original resource information published by user accounts;

[0260] Allocate a second point for the user account according to the original resource information and the preset original resource accumulation rules;

[0261] and / or,

[0262] Get resource processing information of user account;

[0263] Allocate third points for user accounts based on resource processing information and preset resource processing points accumulation rules;

[0264] and / or,

[0265] Get resource sharing information of user accounts;

[0266] Allocate fourth points for user accounts according to resource sharing information and preset resource sharing points rules;

[0267] and / or,

[0268] Get information and comments of user accounts;

[0269] The fifth point is configured for the user account according to the information comment information and the preset information comment point accumulation rules.

[0270] The working principle and beneficial effects of the above technical solution are:

[0271] All users are provided with two virtual currency systems, learning coins and learning points. System administrators can configure users to obtain corresponding points through daily login and sign-in, resource creation, resource processing, resource sharing, information comments, etc. At the same time, points can be used to view specific information and download high-quality original resources. The details are as follows: According to the login and sign-in information of the user account, the first point is configured for the user. For example, if the user signs in, one point is obtained on the first day, two points are obtained by continuing to sign in on the second day, and so on, and seven points are obtained on the seventh day; if the continuous sign-in is interrupted, it will be restored to one point per day; if you do not sign in after logging in on the same day, you can make up for it within the preset time (for example, two days). According to the original resource information published by the user, points can be accumulated, and the corresponding points can be configured according to the click situation; the corresponding points can be configured according to the user's resource processing situation; the corresponding points can be configured according to the resource sharing situation; and the corresponding points can be configured according to the information comment situation. In addition, learning coins are allocated according to the user's recharge situation; learning points and learning coins can be exchanged with each other according to preset rules.

[0272] In one embodiment, the registration module 11 further performs the following operations:

[0273] Obtain a user list input by a second user with high-level permissions;

[0274] Based on the user list, multiple initial accounts are opened; the initial accounts use the name of the user list and the user name in the user list as the login name;

[0275] Receiving a first login operation performed by a third user corresponding to the initial account using the initial account;

[0276] When the third user performs the first login operation, the handwriting collection interface is output multiple times;

[0277] collecting a second handwriting of a third user inputting a login name multiple times through the handwriting collection interface;

[0278] Based on the second handwriting, determining a second model as login verification of the initial account;

[0279] Receiving a processing operation of the third user on the user information of the initial account, the processing operation including: modifying, inputting, deleting or one or a combination thereof;

[0280] Receiving an operation of logging out of the first login by the third user, and saving the initial account as the user account;

[0281] The receiving step of performing a first login operation using the initial account by a third user corresponding to the initial account includes:

[0282] Acquire a second handwriting input by a third user;

[0283] Performing OCR recognition on the second handwriting to obtain recognition information;

[0284] The identification information is matched with the login name of the initial account, and when the match is confirmed, the first login of the initial account is completed.

[0285] The working principle and beneficial effects of the above technical solution are:

[0286] The second user with high-level permissions is the class teacher, and the user list is the class student list. Based on the user list, user accounts for each student (third user) can be opened in batches at the same time in units of classes, thereby increasing the rate of account opening; further, the user list can be filled with specific information of each user, and the information can also be supplemented when the user account logs in for the first time.

[0287] In one embodiment, based on the first handwriting, determining a first model as a login verification of a user account includes:

[0288] Parse the first handwriting to obtain stroke order information; the stroke order information includes: a stroke type, an order between stroke types, and parameter information of each stroke, or one or more combinations thereof; the parameter information of each stroke includes: a position relationship parameter between a starting point, an end point, and a preset number of sampling points between the starting point and the end point of each stroke, and / or a handwriting strength parameter of a preset number of sampling points between the starting point, the end point, and the starting point and the end point of each stroke;

[0289] Extracting characteristic values ​​of the stroke order information, inputting the characteristic values ​​into a preset initial neural network model for training, and obtaining a first model for login verification of a user account after the initial neural network model converges;

[0290] or,

[0291] Extract the eigenvalues ​​of the stroke order information and construct a verification vector based on the eigenvalues;

[0292] Building a login verification library based on the verification vector, and using the login verification library as the first model;

[0293] Among them, when the login verification library is used as the first model for login verification, the login verification steps are as follows:

[0294] Obtaining the third handwriting to be verified;

[0295] extracting a feature value of the third handwriting, and constructing a handwriting vector based on the extracted feature value;

[0296] Calculate the similarity between the handwriting vector and the verification vector. The calculation formula is as follows:

[0297]

[0298] Among them, X i is the similarity between the handwriting vector and the i-th verification vector in the login verification library; n is the dimension of the handwriting vector and the verification vector; a i,j is the j-th dimension data value of the i-th verification vector; b j is the j-th dimension data value of the handwriting vector;

[0299] When the similarity between the handwriting vector and all verification vectors in the login verification library is greater than or equal to a preset first threshold and the maximum value of the similarities is greater than a preset second threshold, the login verification is passed.

[0300] The working principle and beneficial effects of the above technical solution are:

[0301] The present embodiment provides two models for verifying the login of the user account. Model 1, a neural network model, is established in the following manner: using an initial neural network, the handwriting input by the user is machine-learned, and the handwriting of the user is accurately identified when logging in, thereby realizing verification. Model 2, a login verification library, analyzes the handwriting, determines the stroke order information, and then extracts features from the stroke order information, and establishes a verification vector based on the extracted feature values. Whether it is a neural network model or a login verification library, when writing the login name, it is also possible not to follow the standard, realize that the stroke order is used as a verification point according to user customization, improve the granularity of verification, and realize the operability of user customization. In addition, considering the handwriting strength parameters between the strokes, the recognition of the same stroke of the same word written by different people is realized, and the recognition of personnel based on the stroke order information and the identification information scheme is improved. Further, after each successful login, the handwriting of this login is used as material to update the first model.

[0302] In one embodiment, the user account management system based on stroke order, OCR and artificial intelligence further includes: an account transfer module 13, which performs the following operations:

[0303] When obtaining first account transfer instruction information input by a third user corresponding to a user account of the third user, sending a first inquiry to a second user associated with the third user;

[0304] receiving a first feedback from a second user regarding the first inquiry;

[0305] When the first feedback is passed, the first account transfer instruction information is parsed to obtain the source data platform of the account transfer;

[0306] Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission;

[0307] Adding the first feedback to the end of the first account transfer instruction information to generate a second query, and sending the second query to a fourth user;

[0308] receiving a second feedback from a fourth user regarding the second inquiry;

[0309] When the second feedback is passed, the account transfer instruction information is parsed again to obtain the target data platform for the account transfer;

[0310] Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission;

[0311] Adding the second feedback to the end of the first account transfer instruction information to generate a third query, and sending the third query to the fifth user;

[0312] receiving a third feedback from a fifth user regarding the third inquiry;

[0313] When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform;

[0314] or,

[0315] When obtaining the second account transfer instruction information input by the second user corresponding to the user account of the third user, parsing the first account transfer instruction information, and obtaining the source data platform of the account transfer;

[0316] Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission;

[0317] generating a second inquiry based on the second account transfer instruction information, and sending the second inquiry to a fourth user;

[0318] receiving a second feedback from a fourth user regarding the second inquiry;

[0319] When the second feedback is passed, the account transfer instruction information is parsed again to obtain the target data platform for the account transfer;

[0320] Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission;

[0321] Adding the second feedback to the end of the first account transfer instruction information to generate a third query, and sending the third query to the fifth user;

[0322] receiving a third feedback from a fifth user regarding the third inquiry;

[0323] When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform.

[0324] The working principle and beneficial effects of the above technical solution are:

[0325] The specific application scenario of this embodiment is student transfer. The existing routine operation is to re-open an account on the education platform of the place where the transfer takes place. To open an account, the information and password need to be reset, which is extremely inconvenient. Therefore, a transfer method is adopted. Through the application of the student (third user) or the application of the head teacher in charge of the student (second user), with the consent of the principal (fourth user), the user account is packaged and transferred out of the source data platform, and with the consent of the principal of the recipient (fifth user), it is transferred to the target data platform. It avoids the tedious processing of account cancellation and re-application, and realizes that the account follows the student after the student transfers; the place of affiliation in the login name is also relatively changed after the transfer. Among them, the first-level authority is greater than the high-level authority.

[0326] In one embodiment, the points management module 12 further performs the following operations:

[0327] Get the historical login information of the user account;

[0328] Analyze historical login information to obtain the number of days the user has logged in within a preset time period from the current time, the daily login time, and a list of the data in the database called when logging in every day;

[0329] Based on the data in the list, the preset call data and time comparison table, the daily login time is corrected to determine the effective time; specifically as follows: the total call time of the data in the call list is counted, and when the total call time is greater than or equal to the daily login time, the daily login time is the effective time; when the total call time is less than or equal to the daily login time, the total call time is the effective time;

[0330] The user's activity is determined based on the current user's effective time, the sum of all users' effective time, the current user's login days, and the number of days corresponding to the preset time period. The calculation formula is as follows:

[0331]

[0332] Among them, H k is the activity of the kth user in the big data platform, D kis the number of days the kth user logged in; D0 is the number of days corresponding to the preset time period; B k is the effective time of the kth user; M is the number of users on the big data platform; ε1 and ε2 are preset weights;

[0333] Points are allocated to user accounts based on activity and preset activity allocation rules.

[0334] The working principle and beneficial effects of the above technical solution are:

[0335] The user activity is calculated based on the number of login days and the effective login time of each day to ensure the accuracy of the activity. In addition, the effective time is introduced to correct the login time through the data called each time the user logs in, eliminating the impact of the user's inactivity time on the grading, making the activity more accurate and objective. Points are allocated based on activity to increase users' enthusiasm for using the platform.

[0336] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A user account management method based on stroke order, OCR and artificial intelligence, characterized in that: include: When receiving a registration instruction from the first user, entering a registration mode; Obtaining user information of the first user, the user information including: one or more combinations of user name, registered residence, and school registration place; generating a login name for a user account based on the user information; Acquire the first handwriting of the user inputting the login name multiple times; Based on the first handwriting, determining a first model as a login verification for the user account; Calculate the hash value of the first model and save the hash value on the chain; Also includes: Obtain a user list input by a second user with high-level permissions; Based on the user list, a plurality of initial accounts are opened; the initial accounts use the names of the user list and the names of the users in the user list as login names; Receiving a first login operation performed by a third user corresponding to the initial account using the initial account; When the third user performs the first login operation, outputting the handwriting collection interface multiple times; collecting the second handwriting of the third user inputting the login name multiple times through the handwriting collection interface; Based on the second handwriting, determining a second model as login verification for the initial account; Receiving a processing operation of the third user on the user information of the initial account, wherein the processing operation includes: modifying, inputting, deleting, or one or more combinations thereof; receiving an operation of the third user logging out of the first login, and saving the initial account as the user account; Wherein, receiving a first login operation performed by a third user corresponding to the initial account using the initial account includes: Acquire a second handwriting input by a third user; Performing OCR recognition on the second handwriting to obtain recognition information; Matching the identification information with the login name of the initial account, and when a match is found, completing the first login of the initial account; Also includes: When obtaining first account transfer instruction information input by the third user corresponding to the user account of the third user, sending a first inquiry to the second user associated with the third user; receiving a first feedback from the second user regarding the first inquiry; When the first feedback is passed, parsing the first account transfer instruction information, and acquiring a source data platform for the account transfer; Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission; adding the first feedback to the end of the first account transfer instruction information to generate a second inquiry, and sending the second inquiry to the fourth user; receiving a second feedback from the fourth user regarding the second inquiry; When the second feedback is passed, parsing the account transfer instruction information again to obtain the target data platform for the account transfer; Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission; adding the second feedback to the end of the first account transfer instruction information to generate a third inquiry, and sending the third inquiry to the fifth user; receiving a third feedback from the fifth user regarding the third inquiry; When the third feedback is passed, transferring the user account corresponding to the account transfer instruction information to the account management database of the target data platform; or, When obtaining the second account transfer instruction information input by the second user corresponding to the user account of the third user, parsing the first account transfer instruction information to obtain the source data platform for the account transfer; Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission; generating the second inquiry based on the second account transfer instruction information, and sending the second inquiry to the fourth user; receiving a second feedback from the fourth user regarding the second inquiry; When the second feedback is passed, parsing the account transfer instruction information again to obtain the target data platform for the account transfer; Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission; adding the second feedback to the end of the first account transfer instruction information to generate a third inquiry, and sending the third inquiry to the fifth user; receiving a third feedback from the fifth user regarding the third inquiry; When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform.

2. The user account management method based on stroke order, OCR and artificial intelligence as claimed in claim 1, characterized in that: Also includes: Obtaining login sign-in information of the user account; Allocate a first point for the user account according to the login sign-in information and a preset login sign-in point accumulation rule; and / or, Obtaining original resource information published by the user account; Allocating second points for the user account according to the original resource information and a preset original resource point accumulation rule; and / or, Obtaining resource processing information of the user account; Allocate third points for the user account according to the resource processing information and preset resource processing point accumulation rules; and / or, Obtain resource sharing information of the user account; Allocating fourth points for the user account according to the resource sharing information and a preset resource sharing points accumulation rule; and / or, Obtain information comment information of the user account; The fifth point is configured for the user account according to the information comment information and the preset information comment point accumulation rule.

3. The user account management method based on stroke order, OCR and artificial intelligence as claimed in claim 1, characterized in that: Determining a first model for login verification of the user account based on the first handwriting includes: Parse the first handwriting to obtain stroke order information; the stroke order information includes: a stroke type, an order between stroke types, and parameter information of each stroke, or one or more combinations thereof; the parameter information of each stroke includes: a position relationship parameter between a starting point, an end point, and a preset number of sampling points between the starting point and the end point of each stroke, and / or a handwriting strength parameter of a starting point, an end point, and a preset number of sampling points between the starting point and the end point of each stroke; Extracting feature values ​​of the stroke order information, inputting the feature values ​​into a preset initial neural network model for training, and obtaining a first model for login verification of a user account after the initial neural network model converges; or, Extracting characteristic values ​​of the stroke order information, and constructing a verification vector based on the characteristic values; Building a login verification library based on the verification vector, and using the login verification library as a first model; When the login verification is performed using the login verification library as the first model, the login verification steps are as follows: Obtaining the third handwriting to be verified; Extracting a feature value of the third handwriting, and constructing a handwriting vector based on the extracted feature value; The similarity between the handwriting vector and the verification vector is calculated using the following formula: ; in, The handwriting vector and the first Similarity of the verification vectors; is the dimension of the handwriting vector and the verification vector; For the The verification vector Dimension data value; is the first Dimension data value; When the similarity between the handwriting vector and all verification vectors in the login verification library is greater than or equal to a preset first threshold and the maximum value of the similarities is greater than a preset second threshold, the login verification is passed.

4. A user account management system based on stroke order, OCR and artificial intelligence, characterized in that: include: A registration module, wherein the registration module performs the following operations: When receiving a registration instruction from the first user, entering a registration mode; Obtaining user information of the first user, the user information including: one or more combinations of user name, registered residence, and school registration place; generating a login name for a user account based on the user information; Acquire the first handwriting of the user inputting the login name multiple times; Based on the first handwriting, determining a first model as a login verification for the user account; Calculate the hash value of the first model and save the hash value on the chain; The registration module also performs the following operations: Obtain a user list input by a second user with high-level permissions; Based on the user list, a plurality of initial accounts are opened; the initial accounts use the names of the user list and the names of the users in the user list as login names; Receiving a first login operation performed by a third user corresponding to the initial account using the initial account; When the third user performs the first login operation, outputting the handwriting collection interface multiple times; collecting the second handwriting of the third user inputting the login name multiple times through the handwriting collection interface; Based on the second handwriting, determining a second model as login verification for the initial account; Receiving a processing operation of the third user on the user information of the initial account, wherein the processing operation includes: modifying, inputting, deleting, or one or more combinations thereof; receiving an operation of the third user logging out of the first login, and saving the initial account as the user account; Wherein, receiving a first login operation performed by a third user corresponding to the initial account using the initial account includes: Acquire a second handwriting input by a third user; Performing OCR recognition on the second handwriting to obtain recognition information; Matching the identification information with the login name of the initial account, and when a match is found, completing the first login of the initial account; The invention also includes: an account transfer module, wherein the account transfer module performs the following operations: When obtaining first account transfer instruction information input by the third user corresponding to the user account of the third user, sending a first inquiry to the second user associated with the third user; receiving a first feedback from the second user regarding the first inquiry; When the first feedback is passed, parsing the first account transfer instruction information, and acquiring a source data platform for the account transfer; Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission; adding the first feedback to the end of the first account transfer instruction information to generate a second inquiry, and sending the second inquiry to the fourth user; receiving a second feedback from the fourth user regarding the second inquiry; When the second feedback is passed, parsing the account transfer instruction information again to obtain the target data platform for the account transfer; Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission; adding the second feedback to the end of the first account transfer instruction information to generate a third inquiry, and sending the third inquiry to the fifth user; receiving a third feedback from the fifth user regarding the third inquiry; When the third feedback is passed, transferring the user account corresponding to the account transfer instruction information to the account management database of the target data platform; or, When obtaining the second account transfer instruction information input by the second user corresponding to the user account of the third user, parsing the first account transfer instruction information to obtain the source data platform for the account transfer; Obtain the first permission information list of the source data platform, and obtain the fourth user with the first level permission; generating the second inquiry based on the second account transfer instruction information, and sending the second inquiry to the fourth user; receiving a second feedback from the fourth user regarding the second inquiry; When the second feedback is passed, parsing the account transfer instruction information again to obtain the target data platform for the account transfer; Obtain a second permission information list of the target data platform, and obtain a fifth user with first-level permission; adding the second feedback to the end of the first account transfer instruction information to generate a third inquiry, and sending the third inquiry to the fifth user; receiving a third feedback from the fifth user regarding the third inquiry; When the third feedback is passed, the user account corresponding to the account transfer instruction information is transferred to the account management database of the target data platform.

5. The user account management system based on stroke order, OCR and artificial intelligence as claimed in claim 4, characterized in that: Also includes: Point management module, the point management module performs the following operations: Obtaining login sign-in information of the user account; Allocate a first point for the user account according to the login sign-in information and a preset login sign-in point accumulation rule; and / or, Obtaining original resource information published by the user account; Allocating second points for the user account according to the original resource information and a preset original resource point accumulation rule; and / or, Obtaining resource processing information of the user account; Allocate third points for the user account according to the resource processing information and preset resource processing point accumulation rules; and / or, Obtain resource sharing information of the user account; Allocating fourth points for the user account according to the resource sharing information and a preset resource sharing points accumulation rule; and / or, Obtain information comment information of the user account; The fifth point is configured for the user account according to the information comment information and the preset information comment point accumulation rule.

6. The user account management system based on stroke order, OCR and artificial intelligence as claimed in claim 4, characterized in that: Determining a first model for login verification of the user account based on the first handwriting includes: Parse the first handwriting to obtain stroke order information; the stroke order information includes: a stroke type, an order between stroke types, and parameter information of each stroke, or one or more combinations thereof; the parameter information of each stroke includes: a position relationship parameter between a starting point, an end point, and a preset number of sampling points between the starting point and the end point of each stroke, and / or a handwriting strength parameter of a starting point, an end point, and a preset number of sampling points between the starting point and the end point of each stroke; Extracting feature values ​​of the stroke order information, inputting the feature values ​​into a preset initial neural network model for training, and obtaining a first model for login verification of a user account after the initial neural network model converges; or, Extracting characteristic values ​​of the stroke order information, and constructing a verification vector based on the characteristic values; Building a login verification library based on the verification vector, and using the login verification library as a first model; When the login verification is performed using the login verification library as the first model, the login verification steps are as follows: Obtaining the third handwriting to be verified; Extracting a feature value of the third handwriting, and constructing a handwriting vector based on the extracted feature value; The similarity between the handwriting vector and the verification vector is calculated using the following formula: ; in, The handwriting vector and the first Similarity of the verification vectors; is the dimension of the handwriting vector and the verification vector; For the The verification vector Dimension data value; is the first Dimension data value; When the similarity between the handwriting vector and all verification vectors in the login verification library is greater than or equal to a preset first threshold and the maximum value of the similarities is greater than a preset second threshold, the login verification is passed.

Images