CN117671701A

CN117671701A - Financial account opening method and device, storage medium and electronic equipment

Info

Publication number: CN117671701A
Application number: CN202311607599.0A
Authority: CN
Inventors: 王金; 龚嘉炜; 周炜; 李林蔚
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2023-11-28
Filing date: 2023-11-28
Publication date: 2024-03-08

Abstract

The application discloses a financial account opening method and device, a storage medium and electronic equipment, and relates to the field of artificial intelligence, financial science and technology or other related fields. The method comprises the following steps: receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object; inputting a plurality of target images into a target network model, identifying a first target area of the plurality of target images through an area identification module in the target network model to obtain an image corresponding to the first target area, and carrying out character identification through a character identification module in the target network model according to the image corresponding to the first target area to obtain target character information corresponding to each target image; and processing the opening request according to the target text information to obtain a processing result. By the method and the device, the problem that the efficiency of opening the financial account for the client is low in the related technology is solved.

Description

Financial account opening method and device, storage medium and electronic equipment

Technical Field

The present disclosure relates to the field of artificial intelligence, financial science and technology, or other related fields, and in particular, to a method and apparatus for opening a financial account, a storage medium, and an electronic device.

Background

The financial institution can open the financial account for the customer only when the customer meets the transacting requirement of the financial account, so that the customer is required to provide relevant proving materials to prove that the financial account meets the opening requirement of the financial account when transacting the business of opening the financial account for the customer. After the financial institution receives the certification materials provided by the clients, the certification materials are manually checked by staff to judge whether the certification material information has problems. The verification of the proving material of the customer is performed manually, so that the problem of low verification material efficiency is caused, and the problem of low financial account opening efficiency is caused.

Aiming at the problem that in the related art, when a financial account is opened for a client, an artificial mode is adopted to audit the image material of the client, so that the efficiency of opening the financial account for the client is lower, no effective solution is proposed at present.

Disclosure of Invention

The main purpose of the application is to provide a method and a device for opening a financial account, a storage medium and electronic equipment, so as to solve the problem that in the related art, when the financial account is opened for a customer, the image material of the customer is audited in a manual mode, so that the efficiency of opening the financial account for the customer is lower.

In order to achieve the above object, according to one aspect of the present application, there is provided a method for opening a financial account. The method comprises the following steps: receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object; inputting the plurality of target images into a target network model, and identifying a first target area of the plurality of target images through an area identification module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, and the target text information is proving information required for processing the account opening request, wherein the target network model is obtained by training based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model; performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image; and processing the account opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object.

Further, identifying, by the region identification module in the target network model, a first target region of the plurality of target images, where obtaining an image corresponding to the first target region includes: identifying a first target area of the plurality of target images through the area identification module to obtain target position information corresponding to the first target area; and dividing the plurality of target images according to the target position information to obtain images corresponding to the first target area.

Further, the account opening request is processed according to the target text information, and the processing result comprises: matching the target text information according to a preset first keyword to obtain a first matching result, wherein the first matching result is used for representing whether first text information exists in the target text information, and the first text information is used for representing identification information of the target object; if the first matching result represents that the first text information exists in the target text information, matching the target text information according to a preset second keyword to obtain a second matching result, wherein the second matching result is used for representing whether the second text information exists in the target text information or not, and the second text information is used for representing insurance payment proving information of the target object; and processing the account opening request according to the second matching result to obtain the processing result.

Further, before acquiring the account opening request of the target object and the plurality of target images corresponding to the target object, the method further includes: acquiring a training sample set, wherein the training sample set at least comprises a plurality of sample images, sample text information corresponding to each sample image and position information of the sample text information in each sample image; and training the initial network model according to the training sample set to obtain the target network model.

Further, training the initial network model according to the training sample set, and obtaining the target network model includes: dividing the training sample set to obtain a target training set and a target test set; training the initial network model according to the target training set to obtain a trained initial network model; testing the trained initial network model according to the target test set to obtain the test score of the trained initial network model; and if the test score is larger than a preset value, determining the initial network model after training as the target network model.

Further, training the initial network model according to the target training set, and obtaining the trained initial network model includes: identifying a second target area in the sample image through the initial network model to obtain an image corresponding to the second target area; performing text recognition on the image corresponding to the second target area through the initial network model to obtain predicted text information corresponding to the sample image; and training the initial network model according to the predicted text information and the image corresponding to the second target area to obtain the trained initial network model.

Further, training the initial network model according to the predicted text information and the image corresponding to the second target area, and obtaining the trained initial network model includes: calculating according to the predicted text information and the sample text information to obtain a first loss function; calculating according to the position information corresponding to the image corresponding to the second target area and the position information of the sample text information to obtain a second loss function; determining the target loss function according to the first loss function and the second loss function; and training the initial network model according to the target loss function to obtain the trained initial network model.

In order to achieve the above object, according to another aspect of the present application, there is provided an opening device of a financial account. The device comprises: the receiving unit is used for receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object; the first recognition unit is used for inputting the plurality of target images into a target network model, recognizing a first target area of the plurality of target images through an area recognition module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, the target text information is proving information required for processing the account opening request, the target network model is obtained through training based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model; the second recognition unit is used for performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image; the processing unit is used for processing the account opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object.

Further, the first identifying unit includes: the identification module is used for identifying a first target area of the plurality of target images through the area identification module to obtain target position information corresponding to the first target area; the segmentation module is used for segmenting the plurality of target images according to the target position information to obtain images corresponding to the first target area.

Further, the processing unit includes: the first matching module is used for matching the target text information according to a preset first keyword to obtain a first matching result, wherein the first matching result is used for representing whether first text information exists in the target text information, and the first text information is used for representing identification information of the target object; the second matching module is used for matching the target text information according to a preset second keyword if the first matching result represents that the first text information exists in the target text information, so as to obtain a second matching result, wherein the second matching result is used for representing whether the second text information exists in the target text information or not, and the second text information is used for representing insurance payment proving information of the target object; and the processing module is used for processing the account opening request according to the second matching result to obtain the processing result.

Further, the apparatus further comprises: the acquisition unit is used for acquiring a training sample set before acquiring an account opening request of a target object and a plurality of target images corresponding to the target object, wherein the training sample set at least comprises a plurality of sample images, sample text information corresponding to each sample image and position information of the sample text information in each sample image; and the training unit is used for training the initial network model according to the training sample set to obtain the target network model.

Further, the training unit includes: the dividing module is used for dividing the training sample set to obtain a target training set and a target testing set; the training module is used for training the initial network model according to the target training set to obtain a trained initial network model; the test module is used for testing the initial network model after training according to the target test set to obtain the test score of the initial network model after training; and the determining module is used for determining the obtained trained initial network model as the target network model if the test score is larger than a preset value.

Further, the test module includes: the first identification sub-module is used for identifying a second target area in the sample image through the initial network model to obtain an image corresponding to the second target area; the second recognition sub-module is used for carrying out character recognition on the image corresponding to the second target area through the initial network model to obtain the predicted character information corresponding to the sample image; and the training sub-module is used for training the initial network model according to the predicted text information and the image corresponding to the second target area to obtain the trained initial network model.

Further, the training submodule includes: the first calculation sub-module is used for calculating according to the predicted text information and the sample text information to obtain a first loss function; a second calculation sub-module, configured to calculate according to the position information corresponding to the image corresponding to the second target area and the position information of the sample text information, to obtain a second loss function; a first determining sub-module, configured to determine the target loss function according to the first loss function and the second loss function; and the second determining sub-module is used for training the initial network model according to the target loss function to obtain the trained initial network model.

In order to achieve the above object, according to an aspect of the present application, there is provided a computer-readable storage medium storing a program, wherein the program, when run, controls a device in which the storage medium is located to execute the opening method of any one of the above financial accounts.

In order to achieve the above object, according to another aspect of the present application, there is further provided an electronic device, where the electronic device includes one or more processors and a memory, and the memory is configured to store one or more processors to implement the method for opening a financial account according to any one of the above.

Through the application, the following steps are adopted: acquiring an account opening request of a target object and a plurality of target images corresponding to the target object; receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object; inputting a plurality of target images into a target network model, and identifying a first target area of the plurality of target images through an area identification module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, and the target text information is proving information required for processing an account opening request, wherein the target network model is obtained by training based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model; performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image; processing the opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object, and the problem that the efficiency of opening the financial account for the client is lower due to the fact that the image material of the client is audited in a manual mode when the financial account is opened for the client in the related technology is solved. According to the method and the device, information extraction is carried out on the proving material image of the customer according to the account opening request of the financial account provided by the customer, and the account opening request is processed according to the extracted target text information, so that the problem that the efficiency of opening the financial account for the customer is low due to the fact that the proving material image of the customer is audited manually is avoided, the proving material image of the customer is audited automatically, the auditing efficiency of the proving material image is improved, and the effect of improving the efficiency of opening the financial account for the customer is achieved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application, illustrate and explain the application and are not to be construed as limiting the application. In the drawings:

FIG. 1 is a flowchart I of a method for provisioning a financial account according to an embodiment of the present application;

FIG. 2 is a second flowchart of a method for provisioning a financial account according to an embodiment of the present application;

FIG. 3 is a schematic diagram of an opening device for a financial account provided in accordance with an embodiment of the present application;

fig. 4 is a schematic diagram of an electronic device provided according to an embodiment of the present application.

Detailed Description

It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the present application described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

For convenience of description, the following will describe some terms or terms related to the embodiments of the present application:

smooth L1 loss function: the L1 loss function is smooth, and the influence of abnormal values can be restrained, so that the robustness of the model is improved. In the text extraction task, the position of the text can be used as a prediction result, the real position of the text can be used as a real result, and then the model is trained by using a Smooth L1 loss function.

PaddleOCR open source suite is an open source OCR (Optical Character Recognition ) suite based on the PaddlePaddle deep learning framework. It provides rich OCR models and tools that can be used for text detection, text recognition, and text direction detection tasks.

LayoutxLM algorithm: the LayoutXLM can effectively extract text information in an image and perform layout analysis and semantic understanding. The algorithm has wide application prospect in the fields of image character recognition, image retrieval, document understanding and the like.

It should be noted that, related information (including, but not limited to, user insurance payment information, user personal information, etc.) and data (including, but not limited to, data for presentation, analyzed data, etc.) related to the present disclosure are information and data authorized by the user or sufficiently authorized by each party. For example, an interface is provided between the system and the relevant user or institution, before acquiring the relevant information, the system needs to send an acquisition request to the user or institution through the interface, and acquire the relevant information after receiving the consent information fed back by the user or institution.

The present invention will be described with reference to preferred implementation steps, and fig. 1 is a flowchart of a method for opening a financial account according to an embodiment of the present application, as shown in fig. 1, and the method includes the following steps:

Step S101, receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object.

For example, an account opening request of a customer (i.e., the target object described above) and a plurality of certification material images (i.e., the target images described above) provided by the customer may be received through a transaction platform of a financial service. For example, an opening request of a customer to open a personal pension account and an image of a proof material required to open the personal pension account are received.

Step S102, inputting a plurality of target images into a target network model, and identifying a first target area of the plurality of target images through an area identification module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, and the target text information is proving information required for processing an account opening request, wherein the target network model is trained based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model.

For example, after receiving the above-mentioned account opening request and the multiple target images, inputting the multiple target images into the target network model, and then identifying the region (i.e. the first target region) containing the target text information in the multiple target images provided by the client through the region identification module in the target network model, so as to obtain the image corresponding to the first target region. For example, an area including personal information in the document image is located. The target text information includes all key information required for opening the financial account, such as name, gender, certificate number, etc.

Step S103, performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image.

For example, the text information in the image corresponding to the first target area is identified by a text identification module in the target network model, so as to obtain the target text information corresponding to each client. For example, by means of a character recognition module, character recognition is performed on the area including the personal information in the certificate image, and the personal information of the client is extracted, so that character information describing the identity of the client is obtained. It should be noted that, the target network model may use the target text information output in the form of key value pairs, for example, the target text information may be "name: xiaoming.

Step S104, processing the opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object.

For example, according to the target text information obtained by extracting the information from the target image, the integrity and the authenticity of the information provided by the user are confirmed so as to process the opening request to obtain a processing result, and it is to be noted that the processing result is used for representing whether the financial account has been opened for the target object.

In summary, in the application, according to the account opening request of the financial account provided by the customer, information extraction is performed on the proving material image of the customer through the target network model, and the account opening request is processed according to the extracted target text information, so that the problem that the efficiency of opening the financial account for the customer is relatively low due to the fact that the proving material image of the customer is audited manually is avoided, the proving material image of the customer is audited automatically, the auditing efficiency of the proving material image is improved, and the effect of improving the efficiency of opening the financial account for the customer is achieved.

Optionally, in the method for opening a financial account provided in the embodiment of the present application, identifying, by an area identification module in a target network model, a first target area of a plurality of target images, and obtaining an image corresponding to the first target area includes: identifying first target areas of the plurality of target images through an area identification module to obtain target position information corresponding to the first target areas; and dividing the plurality of target images according to the target position information to obtain images corresponding to the first target area.

For example, the region recognition module is used for recognizing the first target region of the plurality of target images to obtain target position information corresponding to the first target region, and then the plurality of target images are segmented according to the target position information to obtain the image corresponding to the first target region. In an optional embodiment, the region recognition module may delineate the region of the first target region in the multiple target images, so as to obtain the image corresponding to the first target region.

Optionally, in the method for opening a financial account provided in the embodiment of the present application, processing the opening request according to the target text information, where the processing result includes: matching the target text information according to a preset first keyword to obtain a first matching result, wherein the first matching result is used for representing whether the first text information exists in the target text information, and the first text information is used for representing identification information of a target object; if the first matching result represents that the first text information exists in the target text information, matching the target text information according to a preset second keyword to obtain a second matching result, wherein the second matching result is used for representing whether the second text information exists in the target text information or not, and the second text information is used for representing insurance payment proving information of the target object; and processing the opening request according to the second matching result to obtain a processing result.

For example, matching is performed according to a preset first keyword, and whether the first text information exists in the target text information is judged, so that a first judgment result is obtained. The first text information may be text information in a certificate of the client, for example, information such as name, gender, and birth year and month.

In an alternative embodiment, the map function and lambda anonymization function in python may be used to determine whether the first text information exists in the target text information.

If the first judging result indicates that the first text information exists in the target text information, judging whether the second text information exists in the target text information according to a preset second keyword to obtain a second judging result, and processing the opening request according to the target text information of the second judging result to obtain a processing result. It should be noted that, if the first text information exists in the target text information, the proving material of the customer identity information is complete, and under the condition that the proving material of the customer identity information is complete, the relevant qualification examination of the financial account is performed on the customer, and whether the customer meets the account opening requirement of the financial account is judged according to the qualification examination result. It should be noted that the second text information may include insurance payment proving information of the customer.

By matching the keywords of the target text information, whether the proving material image provided by the customer has a problem or not is judged, and whether the customer accords with an account opening request of a financial account or not is further judged, so that the target text information is automatically checked, the efficiency of checking the target text information is improved, and the effect of improving the checking efficiency of the target image is further achieved.

For example, if the second judgment result indicates that the second text information exists in the target text information, determining that the client meets the account opening requirement of the financial account; if the second judging result indicates that the second text information does not exist in the target text information, determining that the client does not meet the account opening requirement of the financial account.

And determining whether the customer meets the account opening requirement of the financial account according to the second judging result, so that the efficiency of whether the customer meets the account opening requirement of the financial account is improved, and the efficiency of opening the financial account for the customer is further improved.

Optionally, in the method for opening a financial account provided in the embodiment of the present application, before acquiring an account opening request of a target object and a plurality of target images corresponding to the target object, the method further includes: acquiring a training sample set, wherein the training sample set at least comprises a plurality of sample images, sample text information corresponding to each sample image and position information of the sample text information in each sample image; and training the initial network model according to the training sample set to obtain a target network model.

For example, a training sample set is obtained first, and then the initial network model is trained according to the training sample set, so as to obtain the target network model. The training sample set may include a plurality of sample images, sample text information corresponding to each sample image, and training sample sets of position information of the sample text information in each sample image. It should be noted that, the purpose of determining the position of the sample text information may be achieved by labeling the region where the sample text information is located in each sample image.

The initial network model is trained through the training sample set, so that a target network model is obtained, the accuracy of information extraction of the target network model on the image is improved, and the accuracy of auditing the target image is further improved.

Optionally, in the method for opening a financial account provided in the embodiment of the present application, training the initial network model according to the training sample set, to obtain the target network model includes: dividing a training sample set to obtain a target training set and a target test set; training the initial network model according to the target training set to obtain a trained initial network model; testing the initial network model after training according to the target test set to obtain the test score of the initial network model after training; if the test score is greater than the preset value, determining the obtained trained initial network model as a target network model.

For example, the training sample set is first divided into a target training set and a target testing set according to a ratio (e.g., 8:2), then the initial network model is trained according to the target training set to obtain a trained initial network model, then the trained initial network model is tested according to the target testing set to obtain a test score of the trained initial network model, and if the test score is greater than a preset value (e.g., 0.9), the trained initial network model is determined to be the target network model. It should be noted that, if the test score is less than or equal to the preset value, the step of training the initial network model after training by the target training sample set needs to be repeatedly performed until the test score is greater than the preset score.

In an alternative embodiment, the initial network model may employ a network model constructed based on the LayoutXLM algorithm.

The initial network model is trained through the target training set, the trained initial network model is obtained, the trained initial network model is tested through the target testing set, the accuracy of information extraction of the image by the trained initial network model is guaranteed, the accuracy of information extraction of the target image by the target network model is effectively improved, and further the handling accuracy of opening financial account business is improved.

Optionally, in the method for opening a financial account provided in the embodiment of the present application, training the initial network model according to the target training set, and obtaining the trained initial network model includes: identifying a second target area in the sample image through the initial network model to obtain an image corresponding to the second target area; performing character recognition on the image corresponding to the second target area through the initial network model to obtain predicted character information corresponding to the sample image; and training the initial network model according to the predicted text information and the image corresponding to the second target area to obtain a trained initial network model.

For example, the area where the sample text information is located (i.e., the second target area) in the sample image is located through the initial network model, so as to obtain the predicted position information and an image corresponding to the second target area, then the text information in the second area is identified, so as to obtain the predicted text information corresponding to the sample image, then the error between the predicted text information and the sample information is calculated, and the error between the predicted position information and the real position information is calculated, and the purpose of training the initial network model is realized by minimizing the error, so as to obtain the trained initial network model.

Optionally, in the method for opening a financial account provided in the embodiment of the present application, training the initial network model according to the predicted text information and the image corresponding to the second target area, where obtaining the trained initial network model includes: calculating according to the predicted text information and the sample text information to obtain a first loss function; calculating according to the position information corresponding to the image corresponding to the second target area and the position information of the sample text information to obtain a second loss function; determining a target loss function according to the first loss function and the second loss function; and training the initial network model according to the target loss function to obtain a trained initial network model.

For example, a first loss function is obtained by performing a first loss calculation on the predicted text information and the sample text information, then a second loss calculation is performed on the position information corresponding to the image corresponding to the second target area and the position information of the sample text information, so as to obtain the second loss function, the target loss function is obtained by calculating according to the first loss function and the second loss function, and finally the initial network model is trained according to the target loss function, so that the trained initial network model is obtained.

In an alternative embodiment, the error calculation described above may be calculated using a Smooth L1 loss function.

In an alternative embodiment, the weight values of the first loss function and the second loss function may be set separately, and then the target loss function is calculated by the weight values of the first loss function and the second loss function.

Training the initial network model according to the target training set to obtain a trained initial network model, and effectively improving the accuracy of the obtained trained initial network model on text information extraction.

In an alternative embodiment, the method for opening the financial account may be implemented by using the flow shown in fig. 2, which is specifically as follows: step 201, preparing a training sample set and a training environment, loading image proving materials of a plurality of clients to a server, and installing and deploying a PaddleOCR open source kit and a python programming language environment on the server; step 202, continuing multi-mode information labeling of sample images in a training sample set, labeling the area where the characters in each sample image are located, and forming multi-mode information by the visual information of the sample images, the position information of a detection frame and corresponding character information; step 203, training the initial model according to the training samples, and setting the training sample set to 7:2:1 is divided into a training set, a verification set and a test set, a model constructed based on a LayoutXLM algorithm is trained through the training set, the model is evaluated through the verification set, parameters of the model are adjusted according to an evaluation result, the model is scored through the test set, and the model with the score larger than 0.9 is determined as a target model; 204, extracting information from image materials provided by customers by using a target model to obtain text information; step 205, keyword rule matching is performed according to the text information, and specifically, map function and lambda function in python can be adopted to perform keyword matching on the text information; step 206, judging whether the customer provides the certificate image or not according to the keyword matching result; step 207, if the customer has uploaded the evidence image, judging whether the customer provides the endowment insurance payment evidence image according to the keyword matching result; and step 208, if the customer has provided the pension insurance payment proving image, determining that the customer meets the pension account opening requirement, and opening the pension account for the customer.

According to the method for opening the financial account, which is provided by the embodiment of the application, the account opening request of the target object and a plurality of target images corresponding to the target object are acquired; receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object; inputting a plurality of target images into a target network model, and identifying a first target area of the plurality of target images through an area identification module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, and the target text information is proving information required for processing an account opening request, wherein the target network model is obtained by training based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model; performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image; processing the opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object, and the problem that the efficiency of opening the financial account for the client is lower due to the fact that the image material of the client is audited in a manual mode when the financial account is opened for the client in the related technology is solved. According to the method and the device, information extraction is carried out on the proving material image of the customer according to the account opening request of the financial account provided by the customer, and the account opening request is processed according to the extracted target text information, so that the problem that the efficiency of opening the financial account for the customer is low due to the fact that the proving material image of the customer is audited manually is avoided, the proving material image of the customer is audited automatically, the auditing efficiency of the proving material image is improved, and the effect of improving the efficiency of opening the financial account for the customer is achieved.

It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.

The embodiment of the application also provides a device for opening the financial account, and it should be noted that the device for opening the financial account in the embodiment of the application can be used for executing the method for opening the financial account provided in the embodiment of the application. The following describes a financial account opening device provided in the embodiments of the present application.

Fig. 3 is a schematic diagram of an opening device of a financial account according to an embodiment of the present application. As shown in fig. 3, the apparatus includes: a receiving unit 301, a first identifying unit 302, a second identifying unit 303, and a processing unit 304.

A receiving unit 301, configured to receive an account opening request of a target object for a financial account and receive a plurality of target images corresponding to the target object;

the first identifying unit 302 is configured to input a plurality of target images into a target network model, identify a first target area of the plurality of target images through an area identifying module in the target network model, and obtain an image corresponding to the first target area, where the first target area is an area where target text information exists, and the target text information is proof information required for processing an account opening request, where the target network model is obtained based on training of a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model;

The second recognition unit 303 is configured to perform text recognition according to the image corresponding to the first target area through a text recognition module in the target network model, so as to obtain target text information corresponding to each target image;

the processing unit 304 is configured to process the opening request according to the target text information to obtain a processing result, where the processing result is used to characterize whether the financial account has been opened for the target object.

According to the opening device of the financial account, the receiving unit 301 receives an account opening request of a target object to the financial account and receives a plurality of target images corresponding to the target object; the first recognition unit 302 inputs a plurality of target images into a target network model, recognizes a first target area of the plurality of target images through an area recognition module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, the target text information is proving information required for processing an account opening request, the target network model is obtained based on training of a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model; the second recognition unit 303 performs text recognition according to the image corresponding to the first target area through a text recognition module in the target network model to obtain target text information corresponding to each target image; the processing unit 304 processes the opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object, so that the problem that in the related art, when the financial account is opened for the client, the efficiency of opening the financial account for the client is lower due to the fact that the image material of the client is audited in a manual mode is solved. According to the method and the device, information extraction is carried out on the proving material image of the customer according to the account opening request of the financial account provided by the customer, and the account opening request is processed according to the extracted target text information, so that the problem that the efficiency of opening the financial account for the customer is low due to the fact that the proving material image of the customer is audited manually is avoided, the proving material image of the customer is audited automatically, the auditing efficiency of the proving material image is improved, and the effect of improving the efficiency of opening the financial account for the customer is achieved.

Optionally, in the financial account opening device provided in the embodiment of the present application, the first identifying unit includes: the identification module is used for identifying first target areas of the plurality of target images through the area identification module to obtain target position information corresponding to the first target areas; the segmentation module is used for segmenting the plurality of target images according to the target position information to obtain images corresponding to the first target area.

Optionally, in the financial account opening device provided in the embodiment of the present application, the processing unit includes: the first matching module is used for matching the target text information according to a preset first keyword to obtain a first matching result, wherein the first matching result is used for representing whether the first text information exists in the target text information, and the first text information is used for representing identification information of a target object; the second matching module is used for matching the target text information according to a preset second keyword if the first matching result represents that the first text information exists in the target text information, so as to obtain a second matching result, wherein the second matching result is used for representing whether the second text information exists in the target text information, and the second text information is used for representing insurance payment proving information of the target object; and the processing module is used for processing the opening request according to the second matching result to obtain a processing result.

Optionally, in the financial account opening device provided in the embodiment of the present application, the device further includes: the acquisition unit is used for acquiring a training sample set before acquiring an account opening request of a target object and a plurality of target images corresponding to the target object, wherein the training sample set at least comprises a plurality of sample images, sample text information corresponding to each sample image and position information of the sample text information in each sample image; the training unit is used for training the initial network model according to the training sample set to obtain a target network model.

Optionally, in the financial account opening device provided in the embodiment of the present application, the training unit includes: the dividing module is used for dividing the training sample set to obtain a target training set and a target testing set; the training module is used for training the initial network model according to the target training set to obtain a trained initial network model; the test module is used for testing the initial network model after training according to the target test set to obtain the test score of the initial network model after training; and the determining module is used for determining the initial network model after training as a target network model if the test score is larger than a preset value.

Optionally, in the financial account opening device provided in the embodiment of the present application, the test module includes: the first identification sub-module is used for identifying a second target area in the sample image through the initial network model to obtain an image corresponding to the second target area; the second recognition sub-module is used for carrying out character recognition on the image corresponding to the second target area through the initial network model to obtain predicted character information corresponding to the sample image; and the training sub-module is used for training the initial network model according to the predicted text information and the image corresponding to the second target area to obtain a trained initial network model.

Optionally, in the financial account opening device provided in the embodiment of the present application, the training submodule includes: the first calculation sub-module is used for calculating according to the predicted text information and the sample text information to obtain a first loss function; the second calculation sub-module is used for calculating according to the position information corresponding to the image corresponding to the second target area and the position information of the sample text information to obtain a second loss function; a first determining sub-module for determining a target loss function according to the first loss function and the second loss function; and the second determining sub-module is used for training the initial network model according to the target loss function to obtain a trained initial network model.

The opening device of the financial account includes a processor and a memory, where the above-mentioned units such as the receiving unit 301, the first identifying unit 302, the second identifying unit 303, and the processing unit 304 are stored as program units in the memory, and the processor executes the above-mentioned program units stored in the memory to implement corresponding functions.

The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be set with one or more than one, and the problem that the efficiency of opening the financial account for the client is lower due to the fact that the image material of the client is audited manually when the financial account is opened for the client in the related technology is solved by adjusting the kernel parameters.

The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.

The embodiment of the invention provides a computer readable storage medium, wherein a program is stored on the computer readable storage medium, and the program realizes a financial account opening method when being executed by a processor.

The embodiment of the invention provides a processor, which is used for running a program, wherein the program runs to execute a financial account opening method.

As shown in fig. 4, an embodiment of the present invention provides an electronic device, where the device includes a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, the following steps are implemented: receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object; inputting a plurality of target images into a target network model, and identifying a first target area of the plurality of target images through an area identification module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, and the target text information is proving information required for processing an account opening request, wherein the target network model is obtained by training based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model; performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image; and processing the opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object.

Optionally, identifying, by the area identifying module in the target network model, a first target area of the plurality of target images, where obtaining an image corresponding to the first target area includes: identifying first target areas of the plurality of target images through an area identification module to obtain target position information corresponding to the first target areas; and dividing the plurality of target images according to the target position information to obtain images corresponding to the first target area.

Optionally, processing the opening request according to the target text information to obtain a processing result, including: matching the target text information according to a preset first keyword to obtain a first matching result, wherein the first matching result is used for representing whether the first text information exists in the target text information, and the first text information is used for representing identification information of a target object; if the first matching result represents that the first text information exists in the target text information, matching the target text information according to a preset second keyword to obtain a second matching result, wherein the second matching result is used for representing whether the second text information exists in the target text information or not, and the second text information is used for representing insurance payment proving information of the target object; and processing the opening request according to the second matching result to obtain a processing result.

Optionally, before acquiring the account opening request of the target object and the plurality of target images corresponding to the target object, the method further includes: acquiring a training sample set, wherein the training sample set at least comprises a plurality of sample images, sample text information corresponding to each sample image and position information of the sample text information in each sample image; and training the initial network model according to the training sample set to obtain a target network model.

Optionally, training the initial network model according to the training sample set, and obtaining the target network model includes: dividing a training sample set to obtain a target training set and a target test set; training the initial network model according to the target training set to obtain a trained initial network model; testing the initial network model after training according to the target test set to obtain the test score of the initial network model after training; if the test score is greater than the preset value, determining the obtained trained initial network model as a target network model.

Optionally, training the initial network model according to the target training set, and obtaining the trained initial network model includes: identifying a second target area in the sample image through the initial network model to obtain an image corresponding to the second target area; performing character recognition on the image corresponding to the second target area through the initial network model to obtain predicted character information corresponding to the sample image; and training the initial network model according to the predicted text information and the image corresponding to the second target area to obtain a trained initial network model.

Optionally, training the initial network model according to the predicted text information and the image corresponding to the second target area, and obtaining the trained initial network model includes: calculating according to the predicted text information and the sample text information to obtain a first loss function; calculating according to the position information corresponding to the image corresponding to the second target area and the position information of the sample text information to obtain a second loss function; determining a target loss function according to the first loss function and the second loss function; and training the initial network model according to the target loss function to obtain a trained initial network model.

The present application also provides a computer program product adapted to perform, when executed on a data processing device, a program initialized with the method steps of: receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object; inputting a plurality of target images into a target network model, and identifying a first target area of the plurality of target images through an area identification module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, and the target text information is proving information required for processing an account opening request, wherein the target network model is obtained by training based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model; performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image; and processing the opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims

1. A method for opening a financial account, comprising:

receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object;

inputting the plurality of target images into a target network model, and identifying a first target area of the plurality of target images through an area identification module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, and the target text information is proving information required for processing the account opening request, wherein the target network model is obtained by training based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model;

performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image;

and processing the account opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object.

2. The method of claim 1, wherein identifying, by the region identification module in the target network model, a first target region of the plurality of target images, the image corresponding to the first target region comprises:

identifying a first target area of the plurality of target images through the area identification module to obtain target position information corresponding to the first target area;

and dividing the plurality of target images according to the target position information to obtain images corresponding to the first target area.

3. The method of claim 1, wherein processing the account opening request according to the target text information includes:

matching the target text information according to a preset first keyword to obtain a first matching result, wherein the first matching result is used for representing whether first text information exists in the target text information, and the first text information is used for representing identification information of the target object;

if the first matching result represents that the first text information exists in the target text information, matching the target text information according to a preset second keyword to obtain a second matching result, wherein the second matching result is used for representing whether the second text information exists in the target text information or not, and the second text information is used for representing insurance payment proving information of the target object;

And processing the account opening request according to the second matching result to obtain the processing result.

4. The method of claim 1, wherein prior to acquiring the request for the opening of the account of the target object and the plurality of target images corresponding to the target object, the method further comprises:

acquiring a training sample set, wherein the training sample set at least comprises a plurality of sample images, sample text information corresponding to each sample image and position information of the sample text information in each sample image;

and training the initial network model according to the training sample set to obtain the target network model.

5. The method of claim 4, wherein training an initial network model in accordance with the training sample set to obtain the target network model comprises:

dividing the training sample set to obtain a target training set and a target test set;

training the initial network model according to the target training set to obtain a trained initial network model;

testing the trained initial network model according to the target test set to obtain the test score of the trained initial network model;

And if the test score is larger than a preset value, determining the initial network model after training as the target network model.

6. The method of claim 5, wherein training the initial network model based on the target training set to obtain a trained initial network model comprises:

identifying a second target area in the sample image through the initial network model to obtain an image corresponding to the second target area;

performing text recognition on the image corresponding to the second target area through the initial network model to obtain predicted text information corresponding to the sample image;

and training the initial network model according to the predicted text information and the image corresponding to the second target area to obtain the trained initial network model.

7. The method of claim 6, wherein training the initial network model based on the predicted text information and the predicted location information to obtain the trained initial network model comprises:

calculating according to the predicted text information and the sample text information to obtain a first loss function;

Calculating according to the position information corresponding to the image corresponding to the second target area and the position information of the sample text information to obtain a second loss function;

determining the target loss function according to the first loss function and the second loss function;

and training the initial network model according to the target loss function to obtain the trained initial network model.

8. An opening device for a financial account, comprising:

the receiving unit is used for receiving an account opening request of a target object to a financial account and receiving a plurality of target images corresponding to the target object;

the first recognition unit is used for inputting the plurality of target images into a target network model, recognizing a first target area of the plurality of target images through an area recognition module in the target network model to obtain an image corresponding to the first target area, wherein the first target area is an area with target text information, the target text information is proving information required for processing the account opening request, the target network model is obtained through training based on a target loss function, and the target loss function is obtained based on predicted position information output by an initial network model and predicted text information output by the target network model;

The second recognition unit is used for performing character recognition according to the image corresponding to the first target area through a character recognition module in the target network model to obtain target character information corresponding to each target image;

the processing unit is used for processing the account opening request according to the target text information to obtain a processing result, wherein the processing result is used for representing whether the financial account is opened for the target object.

9. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the storage medium is controlled to perform the opening method of the financial account of any one of claims 1 to 7 at a device when the program is run.

10. An electronic device comprising one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of opening a financial account of any of claims 1-7.