A method, system, device and medium for obtaining identity based on documents

Abstract

The present invention provides a method, system, device and medium for obtaining identity based on documents, including: by analyzing and processing the text content in the documents, using natural language processing and image recognition processing to extract the identity information of the document issuer or the document recipient. The present invention utilizes natural language and image recognition technologies, which can accurately identify the information on the documents and reduce the error rate. At the same time, through the automated recognition technology, the present invention can save labor costs and improve work efficiency. In addition, the present invention can quickly and accurately obtain the document identity information, provide convenient and efficient services for users, and enhance the user experience. Therefore, compared with the traditional method of manually searching for document identity information, the present invention has higher efficiency, lower error rate, lower cost and better user experience, showing obvious superiority.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to a method, system, device and medium for obtaining identity based on documents. Background Art

[0002] Currently, some enterprises use different documents associated with document management software to assist the enterprises in controlling the transaction business flow process. For example, the upstream documents and downstream documents in the transaction business flow process are associated through the document management software. Basically, the identity information of the document issuer and the recipient is recorded in the upstream documents and downstream documents. When an enterprise needs to know the identity information of a supplier in the business flow process for some reasons, it can complete this by searching for the identity information of the document. Currently, the identity information of the document issuer or the recipient is usually searched manually, which may require browsing each document one by one for query, and then when the corresponding document is found, the identity information on the document is consulted.

[0003] However, when the business volume of an enterprise is too large, a large number of documents will be associated. If, under the condition of a large number of documents, the identity of each document is still searched manually, it will invisibly increase the workload of the corresponding personnel, and the manual search method has low efficiency, long time consumption and high error rate. Summary of the Invention

[0004] In view of the above-mentioned disadvantages of the prior art, the purpose of the present invention is to provide a method, system, device and medium for obtaining identity based on documents, which are used to solve the technical problems existing in the prior art.

[0005] To achieve the above object and other related objects, the present invention provides a method for obtaining identity based on documents, including the following steps:

[0006] Receiving a first identity acquisition character input by a first target object on a first display interface;

[0007] Parsing the first identity acquisition character, and extracting a keyword field for obtaining the identity information of a second target object based on the parsing result of the first identity acquisition character, denoted as the first keyword field; wherein, there is a transaction document record between the first target object and the second target object;

[0008] Using the first keyword field as the identity index of the first identity acquisition character, and calculating the similarity between the first keyword field and a second keyword field as the first similarity; and comparing the first similarity value with a preset similarity value; wherein, the second keyword field is pre-stored in an identity database and is generated based on the transaction document record between the first target object and the second target object;

[0009] If the first similarity value is greater than or equal to the preset similarity value, it is determined that there is a second target object in the identity database, and the identity information corresponding to the second keyword field is matched from the identity database as the first query result corresponding to the first identity acquisition character, and the first query result is displayed on the second display interface;

[0010] If the first similarity value is less than the preset similarity value, it is determined that there is no second target object in the identity database, and a third display interface is fed back to the first target object, and the first identity acquisition character is redisplayed on the third display interface to remind the first target object to check the first identity acquisition character again.

[0011] In an embodiment of the present invention, after the identity acquisition character is redisplayed on the third display interface, the method further includes:

[0012] Detect whether the first target object modifies the first identity acquisition character on the third display interface;

[0013] If the first target object does not modify the first identity acquisition character on the third display interface, the first identity acquisition character is re-analyzed, and it is determined whether there is a second target object in the identity database based on the analysis result of the first identity acquisition character;

[0014] If the second target object modifies the first identity acquisition character on the third display interface, the first identity acquisition character after the modification is determined as the second identity acquisition character; and, the second identity acquisition character is analyzed, and it is determined whether there is a second target object in the identity database based on the analysis result of the second identity acquisition character.

[0015] In an embodiment of the present invention, the process of analyzing the second identity acquisition character and determining whether there is a second target object in the identity database based on the analysis result of the second identity acquisition character includes:

[0016] Analyze the second identity acquisition character, and extract the keyword field for obtaining the identity information of the second target object based on the analysis result of the second identity acquisition character, denoted as the third keyword field;

[0017] Use the third keyword field as the identity index of the second identity acquisition character, calculate the similarity between the third keyword field and the second keyword field as the second similarity; and compare the second similarity value with the preset similarity value;

[0018] If the second similarity value is greater than or equal to a preset similarity value, it is determined that there is a second target object in the identity database, and the identity information corresponding to the second keyword field is matched from the identity database as the second query result corresponding to the second identity acquisition character, and the second query result is displayed on the second display interface;

[0019] If the second similarity value is less than the preset similarity value, it is determined that there is no second target object in the identity database, and a fourth display interface is fed back to the first target object, and a preset prompt message is displayed on the fourth display interface; wherein, the preset prompt message is used to prompt that there is no second target object in the identity database.

[0020] In an embodiment of the present invention, the process of generating the second keyword field in the identity database includes:

[0021] Use an image capturing device to capture a pre-acquired or real-time paper transaction document to form an electronic image of the transaction document;

[0022] Perform optical character recognition processing on the electronic image of the transaction document to obtain the text information in the electronic image of the transaction document;

[0023] Perform semantic recognition on the text information, and extract from the text information the keyword field for characterizing the identity information of the second target object in the paper transaction document, denoted as the second keyword field;

[0024] Use the second keyword field as the identity index of the second target object and add it to the pre-established or real-time identity database.

[0025] In an embodiment of the present invention, the process of performing optical character recognition processing on the electronic image of the transaction document includes:

[0026] Label multiple electronic images of transaction documents to form a sample data set; and, select multiple electronic images of transaction documents from the sample data set according to a random ratio as the training data set, and use the remaining electronic images of transaction documents as the test set;

[0027] Input the training data set into a pre-trained optical character recognition model for training, and when the number of training times reaches a preset number of times, output the trained optical character recognition model;

[0028] Use the test set to detect the recognition accuracy of the trained optical character recognition model;

[0029] When the recognition accuracy is greater than or equal to a preset accuracy value, use the trained optical character recognition model at the current moment to perform optical character recognition processing on the electronic image of the transaction document;

[0030] When the recognition accuracy is less than the preset accuracy value, multiple electronic images of transaction documents are randomly selected from the sample dataset again as a new training dataset, and the remaining electronic images of transaction documents are used as a new test set; and the optically character recognition model trained at the current moment is iteratively trained using the new training dataset, and the recognition accuracy of the optically character recognition model after iterative training is detected using the new test set, until the recognition accuracy corresponding to the optically character recognition model after iterative training is greater than or equal to the preset accuracy value, the iterative training is ended, and the optically character recognition model at the end of the iterative training is used to perform optically character recognition processing on the electronic images of the transaction documents.

[0031] In an embodiment of the present invention, when multiple electronic images of transaction documents are randomly selected from the sample dataset as a training dataset, the method further includes:

[0032] Adding a perturbation factor to one or more selected electronic images of transaction documents for image enhancement, and adding the transaction document electronic images after image enhancement to the training dataset to update the training dataset; and training the pre-trained optically character recognition model using the updated training dataset.

[0033] Among them, the parameters for adding the perturbation factor include at least one of the following: character rule, character length, dictionary range, number of characters, text line, text box.

[0034] In an embodiment of the present invention, the second keyword field includes at least one of the following: name of the document issuer, name of the document recipient, address of the document issuer, address of the document recipient.

[0035] The present invention also provides an identity acquisition system based on documents, and the system includes:

[0036] A character acquisition module, configured to receive a first identity acquisition character input by a first target object on a first display interface;

[0037] A field extraction module, configured to parse the first identity acquisition character, and extract a keyword field for obtaining the identity information of a second target object based on the parsing result of the first identity acquisition character, denoted as the first keyword field; wherein, there is a transaction document record between the first target object and the second target object.

[0038] A similarity calculation and comparison module, configured to use the first keyword field as the identity index of the first identity acquisition character, calculate the similarity between the first keyword field and the second keyword field as the first similarity; and compare the first similarity value with a preset similarity value; wherein, the second keyword field is pre-stored in an identity database and is generated based on the transaction document records of the first target object and the second target object.

[0039] A first display module, configured to determine that there is a second target object in the identity database when the first similarity value is greater than or equal to the preset similarity value, match the identity information corresponding to the second keyword field from the identity database as the first query result corresponding to the first identity acquisition character, and display the first query result on a second display interface.

[0040] A second display module, configured to determine that there is no second target object in the identity database when the first similarity value is less than the preset similarity value, feedback a third display interface to the first target object, and redisplay the first identity acquisition character on the third display interface to remind the first target object to check the first identity acquisition character again.

[0041] The present invention also provides a document-based identity acquisition device, including:

[0042] A processor; and,

[0043] A computer-readable medium storing instructions, when the processor executes the instructions, enabling the device to execute the document-based identity acquisition method as described in any one of the above.

[0044] The present invention also provides a computer-readable medium, on which instructions are stored, and the instructions are loaded and executed by a processor to execute the document-based identity acquisition method as described in any one of the above.

[0045] As described above, the present invention provides a document-based identity acquisition method, system, device and medium, having the following beneficial effects: Through natural language or image recognition technology, the present invention can quickly and accurately find identity information, greatly improving the search efficiency; moreover, the present invention uses natural language and image recognition technology to accurately identify the information on the document, reducing the error rate; at the same time, through automated recognition technology, the present invention can save labor costs and improve work efficiency. In addition, the present invention can quickly and accurately obtain the document identity information, providing users with convenient and efficient services and enhancing the user experience. In summary, compared with the traditional method of manually searching for document identity information, the present invention has higher efficiency, lower error rate, lower cost and better user experience, and has obvious advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of an exemplary system architecture for applying the technical solutions in one or more embodiments of the present invention;

[0047] Figure 2 A schematic flowchart of a document-based identity acquisition method provided in an embodiment of the present invention;

[0048] Figure 3 A schematic flowchart of the hardware structure of a document-based identity acquisition system provided in an embodiment of the present invention;

[0049] Figure 4 A schematic diagram of the hardware structure of a document-based identity acquisition device suitable for implementing one or more embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] The following uses specific specific examples to illustrate the embodiments of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0051] It should be noted that the diagrams provided in this embodiment only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the components related to the present invention are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0052] Figure 1 A schematic diagram showing an exemplary system architecture to which the technical solutions in one or more embodiments of the present invention can be applied. As Figure 1 shown, the system architecture 100 may include a terminal device 110, a network 120, and a server 130. The terminal device 110 may include various electronic devices such as smartphones, tablets, laptops, and desktop computers. The server 130 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The network 120 may be a communication medium of various connection types capable of providing a communication link between the terminal device 110 and the server 130. For example, it may be a wired communication link or a wireless communication link.

[0053] According to implementation requirements, the system architecture in the embodiments of the present invention may have any number of terminal devices, networks, and servers. For example, the server 130 may be a server group composed of multiple server devices. In addition, the technical solutions provided in the embodiments of the present invention may be applied to the terminal device 110, may also be applied to the server 130, or may be jointly implemented by the terminal device 110 and the server 130. The present invention does not make special limitations on this.

[0054] In an embodiment of the present invention, the terminal device 110 or the server 130 of the present invention may receive a first identity acquisition character input by a first target object on a first display interface, parse the first identity acquisition character, and extract a keyword field for obtaining the identity information of a second target object based on the parsing result of the first identity acquisition character, denoted as the first keyword field; wherein, there is a transaction document record between the first target object and the second target object; then use the first keyword field as the identity index of the first identity acquisition character, and calculate the similarity between the first keyword field and a second keyword field as the first similarity; and compare the first similarity value with a preset similarity value; wherein, the second keyword field is pre-stored in the identity database and generated based on the transaction document record between the first target object and the second target object; if the first similarity value is greater than or equal to the preset similarity value, it is determined that there is a second target object in the identity database, and the identity information corresponding to the second keyword field is matched from the identity database as the first query result corresponding to the first identity acquisition character, and the first query result is displayed on a second display interface; if the first similarity value is less than the preset similarity value, it is determined that there is no second target object in the identity database, and a third display interface is fed back to the first target object, and the first identity acquisition character is redisplayed on the third display interface to remind the first target object to check the first identity acquisition character again. By using the terminal device 110 or the server 130 to execute the document-based identity acquisition method, through natural language or image recognition technology, the identity information can be quickly and accurately found, greatly improving the search efficiency; moreover, by using natural language and image recognition technology, the information on the document can be accurately identified, reducing the error rate; at the same time, through automated recognition technology, the labor cost can be saved and the work efficiency can be improved. In addition, the present invention can quickly and accurately obtain the document identity information, provide convenient and efficient services for users, and enhance the user experience. In summary, compared with the traditional method of manually searching for document identity information, the present invention has higher efficiency, lower error rate, lower cost, and better user experience, and has obvious advantages.

[0055] The above part introduced the content of the exemplary system architecture applying the technical solution of the present invention. Next, the document-based identity acquisition method of the present invention will be continued to be introduced.

[0056] Figure 2 shows a schematic flowchart of a document-based identity acquisition method provided by an embodiment of the present invention. Specifically, in an exemplary embodiment, as Figure 2 shown, this embodiment provides a document-based identity acquisition method, which includes the following steps:

[0057] S210, receive the first identity acquisition character input by the first target object on the first display interface;

[0058] S220, parse the first identity acquisition character, and extract a keyword field for obtaining the identity information of the second target object based on the parsing result of the first identity acquisition character, denoted as the first keyword field; wherein, there is a transaction document record between the first target object and the second target object;

[0059] S230, use the first keyword field as the identity index of the first identity acquisition character, and calculate the similarity between the first keyword field and the second keyword field as the first similarity; and compare the first similarity value with a preset similarity value; wherein, the second keyword field is pre-stored in the identity database and generated based on the transaction document record between the first target object and the second target object. As an example, in this embodiment, the second keyword field includes at least one of the following: the name of the document issuer, the name of the document recipient, the address of the document issuer, the address of the document recipient. In this embodiment, the preset similarity value can be set according to the actual situation. For example, the similarity value 0.95 can be used as the preset similarity value, or other numerical similarity values can be used as the preset similarity value, which will not be elaborated here in this embodiment.

[0060] S240-1, if the first similarity value is greater than or equal to the preset similarity value, determine that there is a second target object in the identity database, match the identity information corresponding to the second keyword field from the identity database as the first query result corresponding to the first identity acquisition character, and display the first query result on the second display interface;

[0061] S240-2, if the first similarity value is less than the preset similarity value, determine that there is no second target object in the identity database, feedback the third display interface to the first target object, and redisplay the first identity acquisition character on the third display interface to remind the first target object to check the first identity acquisition character again.

[0062] It can be seen from this that in this embodiment, through natural language or image recognition technology, the identity information can be quickly and accurately found, greatly improving the search efficiency; moreover, in this embodiment, by using natural language and image recognition technology, the information on the document can be accurately recognized, reducing the error rate; at the same time, in this embodiment, through automated recognition technology, the labor cost can be saved and the work efficiency can be improved. In addition, this embodiment can quickly and accurately obtain the document identity information, providing a convenient and efficient service for users and enhancing the user experience. In summary, compared with the traditional method of manually searching for document identity information, this embodiment has higher efficiency, lower error rate, lower cost and better user experience, showing obvious superiority. In addition, in this embodiment, the identity acquisition characters input by the first target object on the first display interface may be incorrect or incomplete. Therefore, when the second target object cannot be obtained through the identity acquisition characters input on the first display interface, the first identity acquisition characters will be redisplayed on the third display interface again, so that the first target object can check the input first identity acquisition characters again, avoiding incorrect input or incomplete input, resulting in the inability to obtain the identity result corresponding to the document. As an example, in this embodiment, the first target object and the second target object can be the issuer of the paper transaction document or the recipient of the paper transaction document. If the first target object is the issuer of the paper transaction document, the second target object is the recipient of the paper transaction document; if the second target object is the issuer of the paper transaction document, the first target object is the recipient of the paper transaction document.

[0063] In an exemplary embodiment, after the identity acquisition character is redisplayed on the third display interface, this embodiment may further include: detecting whether a first target object modifies the first identity acquisition character on the third display interface; if the first target object does not modify the first identity acquisition character on the third display interface, re-parsing the first identity acquisition character, and determining whether a second target object exists in the identity database based on the parsing result of the first identity acquisition character; if the second target object modifies the first identity acquisition character on the third display interface, determining the modified first identity acquisition character as the second identity acquisition character; and, parsing the second identity acquisition character, and determining whether a second target object exists in the identity database based on the parsing result of the second identity acquisition character. It can be seen from this that after the identity acquisition character is redisplayed on the third display interface in this embodiment, it will continue to identify whether the first target object modifies the first identity acquisition character on the third display interface. If not, re-parse based on the first identity acquisition character again, and determine whether a second target object exists in the identity database based on the parsing result of the first identity acquisition character; if the first identity acquisition character has been modified on the third display interface, parse based on the modified first identity acquisition character, that is, parse the second identity acquisition character, and then determine whether a second target object exists in the identity database based on the parsing result of the second identity acquisition character. As an example, the first display interface and the third display interface in this embodiment may be the same display interface at different times. As another example, the first display interface and the third display interface in this embodiment may be two different display interfaces.

[0064] According to the above description, in an exemplary embodiment, the process of parsing the second identity acquisition character and determining whether a second target object exists in the identity database based on the parsing result of the second identity acquisition character includes:

[0065] Parsing the second identity acquisition character, and extracting a keyword field for obtaining the identity information of the second target object based on the parsing result of the second identity acquisition character, denoted as the third keyword field;

[0066] Using the third keyword field as the identity index of the second identity acquisition character, and calculating the similarity between the third keyword field and the second keyword field as the second similarity; and, comparing the second similarity value with a preset similarity value. In this embodiment, the preset similarity value can be set according to the actual situation. For example, the similarity value 0.95 can be used as the preset similarity value, or other numerical similarity values can be used as the preset similarity value, which will not be elaborated in this embodiment.

[0067] If the second similarity value is greater than or equal to the preset similarity value, it is determined that there is a second target object in the identity database, and the identity information corresponding to the second keyword field is matched from the identity database as the second identity, the second query result corresponding to the obtained character is acquired, and the second query result is displayed on the second display interface;

[0068] If the second similarity value is less than the preset similarity value, it is determined that there is no second target object in the identity database, the fourth display interface is fed back to the first target object, and a preset prompt message is displayed on the fourth display interface; wherein, the preset prompt message is used to prompt that there is no second target object in the identity database. As an example, in this embodiment, the preset prompt message can be set according to the actual situation. For example, the preset prompt message can be "Hello, the identity of the document you want to query does not exist"; the preset prompt message can also be "Hello, the object you want to query does not exist".

[0069] In an exemplary embodiment, the process of generating the second keyword field in the identity database includes: using an image capturing device to capture a pre-acquired or real-time paper transaction document to form an electronic image of the transaction document; performing optical character recognition processing on the electronic image of the transaction document to obtain the text information in the electronic image of the transaction document; performing semantic recognition on the text information, and extracting from the text information the keyword field for characterizing the identity information of the second target object in the paper transaction document, denoted as the second keyword field; using the second keyword field as the identity index of the second target object and adding it to the pre-established or real-time identity database. Among them, the image capturing device can be a camera, such as a camera, a mobile phone with a camera function, a tablet with a camera function, a computer with a camera function, a printer with a camera function or a scanning function, etc. As an example, in this embodiment, the second keyword field includes at least one of the following: the name of the document issuer, the name of the document recipient, the address of the document issuer, the address of the document recipient. Specifically, the second keyword fields on a certain document are shown in Table 1 below.

[0070] Table 1 Second keyword fields on a certain document

[0071]

[0072]

[0073] In an exemplary embodiment, the process of performing optical character recognition processing on the electronic image of the transaction document includes:

[0074] Labeling multiple electronic images of transaction documents to form a sample data set; and, selecting multiple electronic images of transaction documents from the sample data set according to a random ratio as the training data set, and using the remaining electronic images of transaction documents as the test set;

[0075] Input the training dataset into the pre-trained optical character recognition model for training, and when the number of training times reaches the preset number of times, output the trained optical character recognition model;

[0076] Use the test set to detect the recognition accuracy of the trained optical character recognition model;

[0077] When the recognition accuracy is greater than or equal to the preset accuracy value, use the optical character recognition model trained at the current moment to perform optical character recognition processing on the electronic image of the transaction document;

[0078] When the recognition accuracy is less than the preset accuracy value, randomly select multiple electronic images of transaction documents from the sample dataset again according to a random ratio as the new training dataset, and use the remaining electronic images of transaction documents as the new test set; and use the new training dataset to perform iterative training on the optical character recognition model trained at the current moment, and use the new test set to detect the recognition accuracy of the optical character recognition model after iterative training until the recognition accuracy corresponding to the optical character recognition model after iterative training is greater than or equal to the preset accuracy value, end the iterative training, and use the optical character recognition model at the end of the iterative training to perform optical character recognition processing on the electronic image of the transaction document.

[0079] In this embodiment, the training framework of the pre-trained optical character recognition model includes but is not limited to: convolutional neural network, recurrent neural network, connectionist sequence classification. As an example, the training framework selected in the embodiment of the present invention is: convolutional neural network + recurrent neural network + connectionist sequence classification. Obtain the hyperparameters of the training sample set data, and set the training probability according to the hyperparameters; randomly select one or more training sample set data according to the set training probability to form a batch for training, and generate one or more pre-trained optical character recognition models. After one or multiple trainings, use the Adaptive Moment Estimation Optimizer (Adam Optimizer) to evaluate and verify the generated pre-trained optical character recognition model, and save the best pre-trained optical character recognition model in the evaluation and verification results. When inputting the training dataset into the pre-trained optical character recognition model for training, if there is a pre-trained optical character recognition model, use the pre-trained optical character recognition model for transfer learning; if there is no pre-trained optical character recognition model, first train to generate a pre-trained optical character recognition model, and then input the training dataset into the pre-trained optical character recognition model for training.

[0080] In an exemplary embodiment, when selecting multiple electronic images of transaction documents from a sample dataset according to a random ratio, this embodiment may further include: adding disturbance factors to one or more selected electronic images of transaction documents for image enhancement, and adding the enhanced electronic images of transaction documents to the training dataset to update the training dataset; and training the pre-trained optical character recognition model using the updated training dataset; wherein the parameters for adding disturbance factors include at least one of the following: character rules, character length, dictionary range, number of characters, text line, text box. Specifically, this embodiment can improve the accuracy and generalization ability of a specific character set for one or more selected electronic images of transaction documents through various sample auto-augmentation methods. This includes enhancing the generalization ability of the basic model through a sample augmentation algorithm, especially enhancing the generalization ability of small character sets. Disturbance factors are added to parameters such as character rules, character length, dictionary range, number of characters, text line, text box, while excluding parameters with little impact, such as font, character set, background image, texture image. To avoid the influence between parameters, the effects generated by specific parameters are automatically tested and iterated. Through transfer learning in a specific scenario, the accuracy requirement is met.

[0081] In summary, the present invention provides a method for obtaining an identity based on a document. By receiving a first identity acquisition character input by a first target object on a first display interface, parsing the first identity acquisition character, and extracting a keyword field for obtaining the identity information of a second target object based on the parsing result of the first identity acquisition character, which is denoted as the first keyword field. Herein, there is a transaction document record between the first target object and the second target object. Then, using the first keyword field as the identity index of the first identity acquisition character, calculating the similarity between the first keyword field and a second keyword field as the first similarity. And comparing the first similarity value with a preset similarity value. The second keyword field is pre-stored in an identity database and generated based on the transaction document record between the first target object and the second target object. If the first similarity value is greater than or equal to the preset similarity value, it is determined that there is a second target object in the identity database, and the identity information corresponding to the second keyword field is matched from the identity database as the first query result corresponding to the first identity acquisition character, and the first query result is displayed on a second display interface. If the first similarity value is less than the preset similarity value, it is determined that there is no second target object in the identity database, and a third display interface is fed back to the first target object, and the first identity acquisition character is redisplayed on the third display interface to remind the first target object to check the first identity acquisition character again. Thus, it can be seen that this method can quickly and accurately find identity information through natural language or image recognition technology, greatly improving the search efficiency. Moreover, this method can accurately identify the information on the document by using natural language and image recognition technology, reducing the error rate. At the same time, this method can save labor costs and improve work efficiency through automated recognition technology. In addition, this method can quickly and accurately obtain the document identity information, providing a convenient and efficient service for users and enhancing the user experience. Therefore, compared with the traditional method of manually searching for document identity information, this method has higher efficiency, lower error rate, lower cost and better user experience, showing obvious superiority. In terms of enterprise financial management, enterprises need to manage and sort a large number of documents such as invoices and receipts. The application of this method can enable enterprise financial personnel to quickly find the relevant information of specific documents, facilitating the preparation and review of financial statements. At the same time, it can improve the audit efficiency, reduce the search time of auditors for documents, and improve the audit accuracy.

[0082] As Figure 3 shown, the present invention also provides a document-based identity acquisition system, including:

[0083] A character acquisition module 310, configured to receive a first identity acquisition character input by a first target object on a first display interface;

[0084] A field extraction module 320 is configured to parse the first identity acquisition character and extract, based on the parsing result of the first identity acquisition character, a keyword field for obtaining the identity information of the second target object, denoted as the first keyword field; wherein, there is a transaction document record between the first target object and the second target object.

[0085] A similarity calculation and comparison module 330 is configured to use the first keyword field as the identity index of the first identity acquisition character, calculate the similarity between the first keyword field and the second keyword field as the first similarity; and compare the first similarity value with a preset similarity value; wherein, the second keyword field is pre-stored in the identity database and generated based on the transaction document record between the first target object and the second target object; as an example, in this embodiment, the second keyword field includes at least one of the following: the name of the document issuer, the name of the document recipient, the address of the document issuer, the address of the document recipient. In this embodiment, the preset similarity value can be set according to the actual situation. For example, the similarity value 0.95 can be used as the preset similarity value, or other numerical similarity values can be used as the preset similarity value, which will not be elaborated in this embodiment.

[0086] A first display module 340 is configured to, when the first similarity value is greater than or equal to the preset similarity value, determine that there is a second target object in the identity database, match the identity information corresponding to the second keyword field from the identity database as the first query result corresponding to the first identity acquisition character, and display the first query result on the second display interface.

[0087] A second display module 350 is configured to, when the first similarity value is less than the preset similarity value, determine that there is no second target object in the identity database, feedback a third display interface to the first target object, and redisplay the first identity acquisition character on the third display interface to remind the first target object to check the first identity acquisition character again.

[0088] It can be seen from this that in this embodiment, through natural language or image recognition technology, the identity information can be quickly and accurately found, greatly improving the search efficiency; moreover, in this embodiment, natural language and image recognition technology are used to accurately identify the information on the document, reducing the error rate; at the same time, in this embodiment, through automated recognition technology, the labor cost can be saved and the work efficiency can be improved. In addition, this embodiment can quickly and accurately obtain the identity information of the document, providing a convenient and efficient service for users and enhancing the user experience. In summary, compared with the traditional method of manually searching for the identity information of the document, this embodiment has higher efficiency, lower error rate, lower cost and better user experience, with obvious advantages. In addition, in this embodiment, the identity acquisition character input by the first target object on the first display interface may be incorrect or incomplete. Therefore, when the second target object cannot be obtained through the identity acquisition character input on the first display interface, the first identity acquisition character will be redisplayed on the third display interface again, so that the first target object can check the input first identity acquisition character again to avoid incorrect or incomplete input, resulting in the inability to obtain the identity result corresponding to the document. As an example, in this embodiment, the first target object and the second target object can be the issuer of the paper transaction document or the recipient of the paper transaction document. If the first target object is the issuer of the paper transaction document, the second target object is the recipient of the paper transaction document; if the second target object is the issuer of the paper transaction document, the first target object is the recipient of the paper transaction document.

[0089] It should be noted that the document-based identity acquisition system provided in the above embodiment and the document-based identity acquisition method provided in the above embodiment belong to the same concept. The specific ways in which each module performs operations have been described in detail in the method embodiment and will not be repeated here. In actual application, the document-based identity acquisition system provided in the above embodiment can, according to needs, allocate the above functions to different functional modules, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above. This will not be limited here either. For example, the character acquisition module 310 is used to implement step S210, the field extraction module 320 is used to implement step S220, the similarity calculation and comparison module 330 is used to implement step S230, the first display module 340 is used to implement step S240-1, and the second display module 350 is used to implement step S240-2.

[0090] It should be noted that when the above embodiments collect, store, use, process, transmit, provide, disclose, delete and other processing of relevant data (such as the first identity acquisition character, the second identity acquisition character, etc.), it is completed with the consent of the user or after obtaining the consent of the user. For example, the first identity acquisition character, the second identity acquisition character, etc. are authorized with the knowledge and consent of the user; or are actively provided by the user after reading the relevant instructions, or are actively authorized / provided / uploaded by the user when using some or all of the functions described in the above embodiments, or obtained through other means / ways with the consent of the user or after obtaining the consent of the user.

[0091] An embodiment of the present invention also provides an identity acquisition device based on a document. The device may include: one or more processors; and one or more machine-readable media storing instructions thereon, which when executed by the one or more processors, cause the device to execute Figure 2 the identity acquisition method based on the document as described above. Figure 4 A schematic structural diagram of an identity acquisition device 1000 based on a document is shown. Refer to Figure 4 As shown, the identity acquisition device 1000 based on a document includes: a processor 1010, a memory 1020, a power supply 1030, a display unit 1040, and an input unit 1060.

[0092] The processor 1010 is the control center of the identity acquisition device 1000 based on a document. It connects various components through various interfaces and lines, and executes various functions of the identity acquisition device 1000 by running or executing software programs and / or data stored in the memory 1020, so as to perform overall monitoring of the identity acquisition device 1000 based on a document. In an embodiment of the present invention, when the processor 1010 calls the computer program stored in the memory 1020, it executes as Figure 2 the identity acquisition method based on the document as described above. Optionally, the processor 1010 may include one or more processing units; preferably, the processor 1010 may integrate an application processor and a modulation and demodulation processor, where the application processor mainly processes the operating system, user interface, applications, etc., and the modulation and demodulation processor mainly processes wireless communication. In some embodiments, the processor and the memory can be implemented on a single chip, and in some embodiments, they can also be separately implemented on independent chips.

[0093] The memory 1020 may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system, various applications, etc.; the data storage area may store data created according to the use of the identity acquisition device 1000 based on a document, etc. In addition, the memory 1020 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices, etc.

[0094] The document-based identity acquisition device 1000 further includes a power supply 1030 (such as a battery) for powering each component. The power supply can be logically connected to the processor 1010 through a power management system, so as to manage functions such as charging, discharging, and power consumption through the power management system.

[0095] The display unit 1040 can be used to display information input by the user or information provided to the user, as well as various menus of the document-based identity acquisition device 1000. In the embodiments of the present invention, it is mainly used to display the display interfaces of various applications in the document-based identity acquisition device 1000 and objects such as text and pictures displayed in the display interfaces. The display unit 1040 may include a display panel 1050. The display panel 1050 can be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc.

[0096] The input unit 1060 can be used to receive information such as numbers or characters input by the user. The input unit 1060 may include a touch panel 1070 and other input devices 1080. Among them, the touch panel 1070, also known as a touch screen, can collect touch operations of the user on or near it (such as operations of the user using a finger, a stylus, or any suitable object or accessory on or near the touch panel 1070).

[0097] Specifically, the touch panel 1070 can detect the touch operation of the user, detect the signals brought by the touch operation, convert these signals into contact coordinates, send them to the processor 1010, and receive and execute the commands sent by the processor 1010. In addition, the touch panel 1070 can be implemented in multiple types such as resistive, capacitive, infrared, and surface acoustic wave. The other input devices 1080 may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control keys, power on / off keys, etc.), a trackball, a mouse, a joystick, etc.

[0098] Of course, the touch panel 1070 can cover the display panel 1050. After the touch panel 1070 detects a touch operation on or near it, it transmits the operation to the processor 1010 to determine the type of touch event. Subsequently, the processor 1010 provides a corresponding visual output on the display panel 1050 according to the type of touch event. Although Figure 4In [the device], the touch panel 1070 and the display panel 1050 are implemented as two independent components to perform the input and output functions of the document-based identity acquisition device 1000. However, in some embodiments, the touch panel 1070 and the display panel 1050 can be integrated to perform the input and output functions of the document-based identity acquisition device 1000.

[0099] The document-based identity acquisition device 1000 may further include one or more sensors, such as a pressure sensor, a gravitational acceleration sensor, a proximity light sensor, etc. Of course, according to the needs in specific applications, the above-mentioned document-based identity acquisition device 1000 may further include other components such as a camera.

[0100] An embodiment of the present invention also provides a computer-readable storage medium. Instructions are stored in the storage medium. When one or more processors execute the instructions, the above-mentioned device can execute the document-based identity acquisition method as Figure 2 described in the present invention.

[0101] Those skilled in the art can understand that Figure 4 merely examples of the document-based identity acquisition device are provided, which do not constitute a limitation to the device. The device may include more or fewer components than those shown in the figure, or combine certain components, or different components. For the convenience of description, the above parts are divided into various modules (or units) according to their functions and described separately. Of course, when implementing the present invention, the functions of the various modules (or units) can be implemented in the same or multiple software or hardware.

[0102] Those skilled in the art should understand that the present invention can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be applied to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, so that the instructions executed by the processors of the computer or other programmable data processing devices generate for implementing in the process Figure 1 each process or multiple processes and / or blocks Figure 1a device for the functions specified in one or more boxes. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device that implements the process Figure 1 one process or more processes and / or boxes Figure 1 a device for the functions specified in one or more boxes. These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process Figure 1 one process or more processes and / or boxes Figure 1 steps for the functions specified in one or more boxes.

[0103] It should be understood that although terms such as first, second, and third may be used in the embodiments of the present invention to describe preset ranges, etc., these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range can also be referred to as the second preset range, and similarly, the second preset range can also be referred to as the first preset range.

[0104] The above embodiments merely illustrate the principles and effects of the present invention and are not used to limit the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. A method for obtaining identity based on documents, characterized in that, The method includes the following steps: Receiving a first identity acquisition character input by a first target object on a first display interface; Parsing the first identity acquisition character, and extracting a keyword field for obtaining the identity information of a second target object based on the parsing result of the first identity acquisition character, denoted as the first keyword field; wherein, there is a transaction document record between the first target object and the second target object; Using the first keyword field as the identity index of the first identity acquisition character, and calculating the similarity between the first keyword field and a second keyword field as the first similarity; and, comparing the first similarity value with a preset similarity value; wherein, the second keyword field is pre-stored in an identity database and is generated based on the transaction document record between the first target object and the second target object; the process of generating the second keyword field in the identity database includes: using an image capturing device to capture a pre-acquired or real-time paper transaction document to form an electronic image of the transaction document; performing optical character recognition processing on the electronic image of the transaction document to obtain the text information in the electronic image of the transaction document; performing semantic recognition on the text information, and extracting from the text information a keyword field for characterizing the identity information of the second target object in the paper transaction document, denoted as the second keyword field; using the second keyword field as the identity index of the second target object and adding it to the pre-established or real-time identity database; If the first similarity value is greater than or equal to the preset similarity value, it is determined that there is a second target object in the identity database, and the identity information corresponding to the second keyword field is matched from the identity database as the first query result corresponding to the first identity acquisition character, and the first query result is displayed on a second display interface; If the first similarity value is less than the preset similarity value, it is determined that there is no second target object in the identity database, and a third display interface is fed back to the first target object, and the first identity acquisition character is redisplayed on the third display interface to remind the first target object to check the first identity acquisition character again.

2. The method for obtaining identity based on documents according to claim 1, characterized in that, After the identity acquisition character is redisplayed on the third display interface, the method further includes: Detecting whether the first target object modifies the first identity acquisition character on the third display interface; If the first target object does not modify the first identity acquisition character on the third display interface, re-parsing the first identity acquisition character, and determining whether there is a second target object in the identity database based on the parsing result of the first identity acquisition character; If the second target object modifies the first identity acquisition character on the third display interface, determining the first identity acquisition character after the modification is completed as a second identity acquisition character; and, parsing the second identity acquisition character, and determining whether there is a second target object in the identity database based on the parsing result of the second identity acquisition character.

3. The method for obtaining identity based on documents according to claim 2, characterized in that, The process of parsing the second identity acquisition character and determining whether there is a second target object in the identity database based on the parsing result of the second identity acquisition character includes: Parsing the second identity acquisition character, and extracting a keyword field for obtaining the identity information of the second target object based on the parsing result of the second identity acquisition character, denoted as the third keyword field; Using the third keyword field as the identity index of the second identity acquisition character, calculating the similarity between the third keyword field and the second keyword field as the second similarity; and comparing the second similarity value with a preset similarity value; If the second similarity value is greater than or equal to the preset similarity value, it is determined that there is a second target object in the identity database, and the identity information corresponding to the second keyword field is matched from the identity database as the second query result corresponding to the second identity acquisition character, and the second query result is displayed on the second display interface; If the second similarity value is less than the preset similarity value, it is determined that there is no second target object in the identity database, and a fourth display interface is fed back to the first target object, and a preset prompt message is displayed on the fourth display interface; wherein, the preset prompt message is used to prompt that there is no second target object in the identity database.

4. The method for obtaining identity based on documents according to claim 1, characterized in that, The process of performing optical character recognition processing on the electronic image of the transaction document includes: Labeling multiple electronic images of transaction documents to form a sample data set; and randomly selecting multiple electronic images of transaction documents from the sample data set as a training data set according to a random ratio, and using the remaining electronic images of transaction documents as a test set; Inputting the training data set into a pre-trained optical character recognition model for training, and outputting the trained optical character recognition model when the number of training times reaches a preset number of times; Using the test set to detect the recognition accuracy of the trained optical character recognition model; When the recognition accuracy is greater than or equal to a preset accuracy value, using the trained optical character recognition model at the current moment to perform optical character recognition processing on the electronic image of the transaction document; When the recognition accuracy is less than the preset accuracy value, randomly selecting multiple electronic images of transaction documents from the sample data set as a new training data set according to a random ratio again, and using the remaining electronic images of transaction documents as a new test set; and using the new training data set to perform iterative training on the trained optical character recognition model at the current moment, and using the new test set to detect the recognition accuracy of the iteratively trained optical character recognition model until the recognition accuracy corresponding to the iteratively trained optical character recognition model is greater than or equal to the preset accuracy value, ending the iterative training, and using the optical character recognition model at the end of the iterative training to perform optical character recognition processing on the electronic image of the transaction document.

5. The method for obtaining identity based on documents according to claim 4, characterized in that, When randomly selecting multiple electronic images of transaction documents from the sample data set as a training data set, the method further includes: Adding disturbance factors to the electronic images of one or more transaction documents for image enhancement, and adding the enhanced transaction document electronic images to the training data set to update the training data set; and training the pre-trained optical character recognition model using the updated training data set; Among them, the parameters for adding disturbance factors include at least one of the following: character rules, character length, dictionary range, number of characters, text lines, and text boxes.

6. The method for obtaining identity based on documents according to any one of claims 1 to 4, characterized in that, The second keyword field includes at least one of the following: name of the document issuer, name of the document recipient, address of the document issuer, and address of the document recipient.

7. An identity acquisition system based on documents, characterized in that, The system includes: A character acquisition module for receiving the first identity acquisition character input by the first target object on the first display interface; A field extraction module for parsing the first identity acquisition character and extracting, based on the parsing result of the first identity acquisition character, a keyword field for obtaining the identity information of the second target object, denoted as the first keyword field; where there is a transaction document record between the first target object and the second target object; A similarity calculation and comparison module for using the first keyword field as the identity index of the first identity acquisition character, calculating the similarity between the first keyword field and the second keyword field as the first similarity; and comparing the first similarity value with a preset similarity value; where the second keyword field is pre-stored in the identity database and generated based on the transaction document record between the first target object and the second target object; the process of generating the second keyword field in the identity database includes: using an image capture device to capture a paper transaction document obtained in advance or in real time to form an electronic image of the transaction document; performing optical character recognition processing on the electronic image of the transaction document to obtain the text information in the electronic image of the transaction document; performing semantic recognition on the text information and extracting from the text information a keyword field for characterizing the identity information of the second target object in the paper transaction document, denoted as the second keyword field; using the second keyword field as the identity index of the second target object and adding it to the identity database established in advance or in real time; A first display module for determining that the second target object exists in the identity database when the first similarity value is greater than or equal to the preset similarity value, matching the identity information corresponding to the second keyword field from the identity database as the first query result corresponding to the first identity acquisition character, and displaying the first query result on the second display interface; A second display module for determining that the second target object does not exist in the identity database when the first similarity value is less than the preset similarity value, feeding back a third display interface to the first target object, and redisplaying the first identity acquisition character on the third display interface to remind the first target object to check the first identity acquisition character again.

8. An identity acquisition device based on documents, characterized in that Including: A processor; And, A computer-readable medium storing instructions that, when executed by the processor, cause the device to perform the document-based identity acquisition method according to any one of claims 1 to 6.

9. A computer-readable medium, characterized in that Instructions are stored thereon, and the instructions are loaded and executed by the processor to perform the document-based identity acquisition method according to any one of claims 1 to 6.

Images