A bank customer information change and verification processing method based on RPA and OCR, a computer device and a storage medium

Abstract

The present invention discloses a bank customer information change and verification processing method based on RPA and OCR, a computer device and a storage medium. The method includes the steps of obtaining an updated ID image, a retained ID image and retained customer information, merging the updated ID image with the retained ID image to obtain a merged ID image, identifying the merged ID image to obtain updated customer information, checking and verifying the updated customer information based on the retained customer information, and updating the retained customer information based on the updated customer information when the check is qualified. The present invention can simplify the customer operation process; by checking the updated customer information based on the retained customer information, the negative impact caused by the instability of the image recognition technology used when performing OCR recognition on the merged ID image to obtain the updated customer information can be reduced, thereby improving the robustness of the bank customer information change processing method in the embodiment. The present invention is widely used in the field of bank digital technology.

Description

Technical Field

[0001] The present invention relates to the field of banking digital technology, and in particular to a method, computer device, and storage medium for changing and verifying bank customer information based on RPA and OCR. Background Art

[0002] Banks frequently handle changes in customer information, such as when individual customers upgrade their ID cards, or when corporate customers upgrade their business licenses or change their legal representatives. These changes require a visit to the bank. Previously, these changes required the customer or a representative to visit the bank, completing an application form, a cumbersome and complex process. Currently, some banks are promoting electronic transactions, eliminating the need for customers to complete paper documents, streamlining the process and saving customers time. However, currently, whether conducting electronic transactions in person at a bank or via mobile phone, customers still need to fill out electronic forms and other documents. This effectively digitizes the document filling process, but does not fundamentally simplify it. Furthermore, bank back-office staff currently rely heavily on manual visual verification to verify newly created and modified documents, which is inefficient, prone to errors, and lacks timeliness. Summary of the Invention

[0003] In response to at least one technical problem that the current process for changing bank customer information requires customers to fill out a large amount of materials and the process is cumbersome and complicated, the purpose of the present invention is to provide a bank customer information change and verification processing method based on RPA and OCR, a computer device and a storage medium.

[0004] In one aspect, an embodiment of the present invention includes a method for changing and verifying bank customer information based on RPA and OCR, comprising:

[0005] The RPA robot obtains the updated document image from the client;

[0006] The RPA robot retrieves the retained ID image and retained customer information; the retained ID image and the retained customer information are retained on the bank system side, and the retained ID image, the retained customer information, and the updated ID image correspond to the same customer;

[0007] The RPA robot merges the updated document image with the retained document image to obtain a merged document image;

[0008] The RPA robot uses the OCR algorithm to identify the combined document image and obtain updated customer information;

[0009] The RPA robot checks the updated customer information based on the retained customer information;

[0010] When the inspection of the updated customer information is qualified, the RPA robot updates the retained customer information according to the updated customer information.

[0011] Furthermore, merging the updated ID image with the retained ID image to obtain a merged ID image includes:

[0012] The RPA robot calls the OCR algorithm to identify each document image area in the updated document image and the retained document image;

[0013] The RPA robot calls the OCR algorithm to merge the image areas of each certificate into the same image according to the set relative position relationship to obtain the merged certificate image.

[0014] Furthermore, the bank customer information change processing method further includes:

[0015] When the check of the updated customer information fails, the RPA robot adjusts the relative position relationship between the areas of the document image;

[0016] The RPA robot merges the document image regions into a single image according to the new relative position relationship to obtain a new merged document image;

[0017] The RPA robot uses the OCR algorithm to recognize the newly merged document image and obtain the new updated customer information;

[0018] Based on the retained customer information, the new updated customer information is checked.

[0019] Furthermore, the bank customer information change processing method further includes:

[0020] When the cumulative number of unqualified checks on the updated customer information reaches a threshold, the RPA robot refuses to update the retained customer information.

[0021] Furthermore, the calling of an OCR algorithm to perform OCR recognition on the combined document image to obtain updated customer information includes:

[0022] The RPA robot calls the OCR algorithm to perform OCR recognition on the combined document image to obtain the first text information;

[0023] The RPA robot groups the first text information according to the document image area to obtain second text information and third text information; wherein the second text information is from the document image area in the retained document image, and the third text information is from the document image area in the updated document image;

[0024] performing cleaning processing on the second text information and the third text information;

[0025] The cleaned second text information and the third text information constitute the updated customer information.

[0026] Furthermore, the cleaning process of the second text information and the third text information includes:

[0027] The RPA robot filters out illegal characters in the second text information and the third text information;

[0028] The RPA robot calls the OCR algorithm to identify characteristic characters in the second text information and the third text information;

[0029] The RPA robot segments the second text information and the third text information based on the characteristic characters.

[0030] Furthermore, updating the retained customer information according to the updated customer information includes:

[0031] The RPA robot invokes an OCR algorithm to identify the orientation of the document image region from which the second text information and the third text information originate;

[0032] The RPA robot identifies the third text information as replacement information or new information based on the orientation of the corresponding document image area;

[0033] For the third text information that is replacement information, part of the content of the retained customer information is replaced with the third text information; wherein the orientation of the ID image area corresponding to the replaced part of the retained customer information is the same as the orientation of the ID image area from which the third text information comes;

[0034] For the third text information that is newly added information, the same content as the third text information is added to the retained customer information.

[0035] Furthermore, the checking of the updated customer information based on the retained customer information includes:

[0036] The RPA robot compares the second text information with the retained customer information to obtain consistency;

[0037] When the consistency reaches a preset threshold, the RPA robot determines that the inspection of the updated customer information is qualified; otherwise, it determines that the inspection of the updated customer information is unqualified.

[0038] On the other hand, an embodiment of the present invention also includes a computer device, including a memory and a processor, wherein the memory is used to store at least one program, and the processor is used to load the at least one program to execute the bank customer information change processing method in the embodiment.

[0039] On the other hand, an embodiment of the present invention further includes a storage medium storing a program executable by a processor, wherein the program executable by the processor executes the bank customer information change processing method in the embodiment.

[0040] The beneficial effects of the present invention are as follows: the bank customer information change processing method in the embodiment only needs to obtain the updated certificate image required for handling the customer information change, and obtain the updated customer information through the RPA robot calling the OCR recognition technology to automatically update the retained customer information, so as to obtain the information required for handling the customer information change and complete the handling of the customer information change business, which can enable customers to avoid filling out forms and other materials multiple times, simplify the operation process, and simplify the back-end personnel's management process of customer information, improve customer information management efficiency, and reduce the error rate of customer information verification; at the same time, by checking the updated customer information based on the retained customer information, it can reduce the negative impact of the instability of the image recognition technology used when identifying the merged certificate image to obtain the updated customer information, thereby improving the robustness of the bank customer information change processing method in the embodiment. The RPA robot combined with the automatic processing technology of OCR effectively improves the back-end management efficiency of the bank customer information change processing method, reduces the error rate, and effectively improves customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 Flowchart of the bank customer information change and verification processing method in the embodiment;

[0042] Figure 2 Schematic diagram of updating a certificate image in an embodiment;

[0043] Figure 3 Schematic diagram of retaining a document image in an embodiment;

[0044] Figure 4 is a schematic diagram of merging document images in an embodiment;

[0045] Figure 5 This is a schematic diagram of the step of checking whether the content of the updated customer information is consistent with the content recorded in the merged document image based on the retained customer information in the embodiment. DETAILED DESCRIPTION

[0046] In this embodiment, refer to Figure 1 , the bank customer information change processing method includes the following steps:

[0047] S1. The RPA robot obtains the updated document image from the client backend server;

[0048] S2. The RPA robot retrieves the retained document image and retained customer information;

[0049] S3. The RPA robot merges the updated ID image with the retained ID image to obtain a merged ID image.

[0050] S4. The RPA robot uses the OCR algorithm to perform OCR recognition on the combined document image to obtain updated customer information.

[0051] S5. The RPA robot checks the updated customer information based on the retained customer information;

[0052] S6. When the RPA robot passes the inspection of the updated customer information, it updates the retained customer information based on the updated customer information.

[0053] In this embodiment, steps S1-S6 may be executed by the bank's server, and the customer establishes a connection with the server through a client such as a mobile phone or tablet computer to perform data communication.

[0054] Before executing steps S1-S6, the server may open a login port to the client, and the client accesses the login port and logs in using a pre-registered account and password. The server may identify the client's corresponding identity through the account and password or the client's network address. In this embodiment, the client may refer to a personal account, a basic corporate account, a general corporate account, etc. The client has provided real-name authentication materials during registration, such as a personal ID card, a business license, an ID card of the legal representative of the enterprise, an ID card of the authorized agent of the enterprise, an ID card of the financial director of the enterprise, an ID card of the contact person of the enterprise, etc. These documents can be photographed or scanned by the bank to retain the document image, or the customer can photograph or scan the document image to retain the document image and send it to the bank, which will store it on the server and can read, modify, delete, and replace it.

[0055] After the client logs in, the server can display service options to the client through the client in the form of buttons. The client can click the button to choose to enter the customer information change service, thereby triggering the execution of steps S1-S6.

[0056] In this embodiment, the customer information change is due to the customer changing the certificate or changing the representative, which requires changing the information retained in the bank system. In step S1, the server can instruct the client to display prompt content, thereby prompting the customer to prepare new certificates, such as a new business license, a new legal representative ID card, a new contact ID card, etc., and then operate the client to take pictures of these certificates. In this embodiment, if the customer needs to change multiple certificates, he can refer to Figure 2 , put all the documents together, and thus take a picture of all the documents as one updated document image.

[0057] Reference Figure 3 In this embodiment, the updated document image includes a business license with the address of "A Province B City C2 District D2 Road" and an ID card with the name "C" and the number "778899".

[0058] In step S2, the bank's server reads the customer's stored ID image and stored customer information. These images can be stored in the bank's system when the customer registers and opens an account. If the customer changes their information after account opening, the stored ID image and information will be the most recent stored ID image and information. The stored ID image can be a photograph or scan of the ID, and the stored customer information can be extracted from the stored ID image using techniques such as optical character recognition (OCR), or manually compiled from the stored ID image.

[0059] Reference Figure 3 In this embodiment, the retained document images include a business license with the address "Province A, City B, District C1, Road D1," an ID card with the name "A" and the ID number "112233," and an ID card with the name "B" and the ID number "445566." A is the legal representative of the customer's company, and B is the financial officer of the customer's company. Accordingly, the retained customer information in text format might read "Business License: Unified Social Security Code - AA123456; Address - Province A, City B, District C1, Road D1; Legal Representative: Name - A, ID Number - 112233; Financial Officer: Name - B, ID Number - 445566."

[0060] The actual business license also records information such as the legal representative and business scope, and the actual ID card also records information such as the address. This embodiment only mentions the fictitious residence information in the business license and the fictitious name and number information in the ID card, so as to briefly explain the bank customer information change processing method in this embodiment.

[0061] In step S3, Figure 2 The updated document image shown is the same as Figure 3 The retained document images shown are merged to obtain Figure 4 The combined document image shown. Specifically, Figure 2 Updated document image shown and Figure 3 In the updated document image shown, the business license and identity card are located in their respective document image areas. When merging the updated document image with the retained document image, it is best that these document image areas do not overlap with each other. The positional relationship between the document image areas can be determined randomly, or a fixed rule such as "the document image area from the updated document image is on the top, and the document image area from the retained document image is on the bottom" can be adopted.

[0062] In step S3, after the updated ID image and the retained ID image are merged into a merged ID image, marking information may be added to the merged ID image to indicate whether each ID image region in the merged ID image is from the updated ID image or the retained ID image. The marking information may include the pixel coordinates of the center point of each ID image region, as well as a numerical value indicating whether the region is from the updated ID image or the retained ID image.

[0063] In this embodiment, when executing step S4, that is, calling the OCR algorithm to perform OCR recognition on the combined document image to obtain updated customer information, the following steps can be specifically performed:

[0064] S401. The RPA robot performs OCR on the combined document image to obtain first text information.

[0065] In step S401, Figure 4 The combined document image shown is subjected to optical character recognition (OCR), and the first text information obtained may be "Uniform Social Security Number AA123456, Residence A, Province B, City C1, District D1, Road Name C, ID Number 778899, Name B, ID Number 445566, Name A, ID Number 112233, Unified Social Security Number AA123456, Residence A, Province B, City C2, District D2, Road" and the like;

[0066] S402. The RPA robot groups the first text information according to the region of the ID image from which it originated, to obtain second and third text information.

[0067] In step S402, the first text information "Uniform Social Security Number AA123456 Residence A Province B City C1 District D1 Road Name C ID Number 778899 Name B ID Number 445566 Name A ID Number 112233 Unified Social Security Number AA123456 Residence A Province B City C2 District D2 Road" includes "Uniform Social Security Number AA123456 Residence A Province B City C1 District D1 Road Name B ID Number 445566 Name A ID Number 112233" which is obtained by recognizing the document image region in the retained document image and is grouped into the second text information; the remaining content "Name C ID Number 778899 Unified Social Security Number AA123456 Residence A Province B City C2 District D2 Road" is obtained by recognizing the document image region in the updated document image and is grouped into the third text information. That is, after processing in step S402, the following can be obtained:

[0068] Second text information: Unified social security number AA123456, address A, province B, city C1, road D1, name B, ID number 445566, name A, ID number 112233;

[0069] Third text information: Name C, ID number 778899, Unified Social Security Number AA123456, Address A, Province B, City C2, Road D2;

[0070] S403. The RPA robot cleans the second text message and the third text message.

[0071] The second and third text messages are cleaned to delete illegal characters such as spaces in the second and third text messages. On the other hand, characteristic characters in documents related to banking services can be set in the system, including "name", "unified social code", "identity card number", "residence", "address", etc. The second and third text messages are segmented based on these characteristic characters. Semicolons and other symbols can be added between segments, and colons and other symbols can be inserted between characteristic characters and other characters in the same segment. After processing in step S403, the following is obtained:

[0072] Second text information: Unified Social Security Number: AA123456; Address: D1 Road, C1 District, B City, Province A; Name: Yi, ID Number: 445566; Name: Jia, ID Number: 112233;

[0073] Third text information: Name: C; ID number: 778899; Unified Social Security Number: AA123456; Address: D2 Road, C2 District, B City, Province A;

[0074] S404. The RPA robot combines the cleaned second text message and the third text message to update the customer information.

[0075] In step S404, the second text information "unified social code: AA123456; residence: D1 Road, C1 District, B City, Province A; name: B, ID number: 445566; name: A, ID number 112233" that has been cleaned and the third text information "name: C; ID number: 778899; unified social code: AA123456; residence: D2 Road, C2 District, B City, Province A" constitute the updated customer information.

[0076] In this embodiment, when explaining step S404, it is assumed that the optical character recognition performed when executing step S401 can obtain completely accurate recognition results. Therefore, for the example of updated customer information "Uniform Social Security Number: AA123456; Residence: D1 Road, C1 District, B City, A Province; Name: B, ID Number: 445566; Name: A, ID Number: 112233; Name: C; ID Number: 778899; Uniform Social Security Number: AA123456; Residence: D2 Road, C2 District, B City, A Province" and Figure 4 The content displayed in the merged ID image is consistent. However, in actual applications, the optical character recognition algorithm used may produce erroneous results due to noise interference and other factors. That is, the updated customer information obtained by executing steps S401-S404 may not be completely consistent with the content recorded in the merged ID image obtained by executing step S3. Step S5 can be executed to check whether the content of the updated customer information is consistent with the content recorded in the merged ID image.

[0077] In step S5, the RPA robot can check whether the updated customer information is consistent with the content recorded in the merged ID image based on the retained customer information. In this embodiment, when executing step S5, the following steps can be specifically performed:

[0078] S501. The RPA robot compares the second text information with the retained customer information to obtain consistency information;

[0079] S502. When the consistency information reaches a preset threshold, the RPA robot determines that the inspection of the updated customer information is qualified; otherwise, it determines that the inspection of the updated customer information is unqualified.

[0080] In step S501, the second text message and the retained customer information are both in text format. A character-by-character correspondence check can be performed to determine consistency. The number of matching characters is calculated as a percentage of the total number of characters counted to determine the degree of consistency. During this character-by-character check, punctuation marks and specific characters indicating document type, such as "business license" and "identity card," can be omitted, with only characters such as letters and numbers being checked. Before this character-by-character check, the characters in the second text message and the retained customer information can be reordered so that they represent the contents of each document in the same order.

[0081] In step S502, a preset threshold can be set to 100%. That is, if each character in the second text message has a corresponding character in the retained customer information, the check for the updated customer information is determined to have passed. For example, in this embodiment, the second text message is "Unified Social Security Code: AA123456; Address: Province A, City B, District C1, Road D1; Name: Yi, ID Number: 445566; Name: Jia, ID Number 112233," and the retained customer information is "Business License: Unified Social Security Code - AA123456; Address - Province A, City B, District C1, Road D1; Legal Representative: Name - Jia, ID Number - 112233; Financial Manager: Name - Yi, ID Number - 445566." Without checking punctuation marks and specific characters indicating the type of document, such as "business license" or "ID card," each character in the second text message has a corresponding character in the retained customer information. Therefore, the consistency reaches the preset threshold of 100%, and the check for the updated customer information is determined to have passed.

[0082] The principle of steps S501-S502 is: Figure 5Since the retained customer information is retained by the customer when registering and opening an account or when the last information change was made, it can be assumed that the retained customer information is consistent with the content recorded in the merged ID image. The updated customer information includes two parts: the second text information and the third text information. The second text information is obtained by recognizing the portion of the merged ID image that comes from the retained ID image. The degree of consistency between the second text information and the retained ID image is basically determined by the running result of the OCR recognition algorithm used in step S401. When the second text information has a high degree of consistency with the content of the retained ID image, it can be inferred that the OCR recognition algorithm used in step S401 is running correctly, that is, the recognition result of the OCR recognition algorithm is reliable. Then, there is also a high degree of confidence in inferring that the third text information recognized by the OCR recognition algorithm is consistent with the content in the updated ID image. Therefore, the inspection result of the updated customer information is qualified. When the second text information has a low degree of consistency with the content of the retained ID image, it can be inferred that the OCR recognition algorithm used in step S401 has an error or is affected by adverse factors such as noise interference. Then there is also a high degree of confidence in inferring that the third text information recognized by the OCR recognition algorithm is inconsistent with the content in the updated ID image, and therefore the inspection result of the updated customer information is unqualified.

[0083] From the principles of steps S501-S502, it can be seen that by executing step S5, the RPA robot checks the updated customer information based on the retained customer information, which can reduce the impact of the uncertainty of the result of identifying the combined document image in step S4, thereby making it easier to obtain accurate updated customer information.

[0084] If the inspection result of the updated customer information is qualified in step S5, it means that the updated customer information can accurately reflect the content recorded in the updated ID image. The RPA robot automatically executes step S6 to update the retained customer information based on the updated customer information. It can also save the updated ID image and record the saving time.

[0085] If the check result of the updated customer information in step S5 is unqualified, it means that the updated customer information cannot accurately reflect the content recorded in the updated certificate image, and the following steps can be performed:

[0086] S7. If the updated customer information fails to pass inspection, the RPA robot adjusts the relative positions of the image regions of the ID card.

[0087] S8. The RPA robot merges the document image regions into a single image based on the new relative positional relationships, obtaining a new merged document image.

[0088] S9. The RPA robot recognizes the newly merged ID image and obtains the new updated customer information;

[0089] S10.RPA robots check new and updated customer information based on retained customer information.

[0090] The principles of steps S7 and S8 are the same as those of step S3, the principles of step S9 are the same as those of step S4, and the principles of step S10 are the same as those of step S5. Figure 4 When the document images are merged as shown, the relative position relationship between the document image regions is randomly arranged, and after executing steps S7 and S8, the contents of the document image regions contained in the new merged document image are the same as those in the previous step. Figure 4 The updated customer information and the updated document image are identical, but the relative positional relationships between the areas of the document image have changed. Step S9 is then executed to rerun the OCR recognition algorithm for recognition. Since the relative positional relationships between the areas of the document image have changed in the newly merged document image, the effects of accidental factors such as the relative positional relationships between the areas of the document image on the OCR recognition algorithm are corrected. Step S10 is then executed to check whether the updated customer information and the content in the updated document image are identical. If they are, the process proceeds to step S6. If they are not, the cumulative number of failed checks on the updated customer information is incremented by 1, and the process proceeds to step S7.

[0091] In this embodiment, after executing steps S7-S10 once, step S11 is executed, where the RPA robot checks whether the cumulative number of failed checks on the updated customer information has reached a threshold. If the cumulative number of failed checks on the updated customer information reaches the threshold, it can be determined that the identified updated customer information is different from the content in the updated ID image. This is not due to accidental factors such as the relative positional relationship between the areas of the ID image, but rather to a defect in the optical character recognition algorithm itself. The identified updated customer information may not necessarily be consistent with the content of the updated ID image. Therefore, the update of the retained customer information can be rejected, and any of steps S1-S11 will no longer be executed. Manual updating of the retained customer information can be performed.

[0092] Since the change to customer information includes adding new customer information based on the original customer information and replacing part of the original customer information with the new customer information, in this embodiment, when executing step S6, that is, the step of updating the retained customer information according to the updated customer information, the following steps can be specifically performed:

[0093] S601. The RPA robot uses the OCR algorithm to identify the orientation of the document image region from which the second and third text information originate.

[0094] S602. The RPA robot identifies the third text information as replacement information or new information based on the orientation of the corresponding document image area;

[0095] S603. The RPA robot replaces a portion of the retained customer information with the third text information, which is the replacement information. The orientation of the ID image area corresponding to the replaced portion of the retained customer information is the same as the orientation of the ID image area from which the third text information originated.

[0096] S604. The RPA robot adds the same content as the third text information to the retained customer information for the third text information which is the newly added information.

[0097] When executing step S1, the RPA robot can read Figure 3 The retained document image shown above uses the OCR algorithm to identify the orientation of each document image area, for example, the business license is facing forward, the ID card of the legal representative of the enterprise is facing forward, and the ID card of the financial director of the enterprise is facing right. The bank system can generate prompt information based on the orientation of each document image area in the retained document image. The content of the prompt information can be expressed in the form of text or schematic diagrams. Specifically, it can be: (1) When shooting and updating the document image, shooting the ID card facing forward indicates that the ID card of the legal representative of the enterprise needs to be replaced, (2) shooting the ID card facing right indicates that the ID card of the financial director of the enterprise needs to be replaced, and (3) shooting the ID card facing other directions indicates that the ID cards of other personnel need to be added. The bank system sends the prompt information to the client and displays it, thereby guiding the customer to select the appropriate document orientation.

[0098] When executing step S4, the RPA robot can call the OCR algorithm to identify the orientation of each ID image area in the merged ID image, and mark the second text information or the third text information obtained by the orientation of the ID image area. In steps S601 and S602, referring to the rules indicated by the prompt information, the third text information is identified as replacement information or new information according to the orientation of the corresponding ID image area, for example Figure 4 In the merged document image shown, the business license with the address of "Road D2, District C2, City B, Province A" has the same orientation as the business license with the address of "Road D1, District C1, City B, Province A", which indicates that the customer wants to replace the business license with the address of "Road D2, District C2, City B, Province A" with the business license with the address of "Road D2, District C2, City B, Province A". Therefore, the third text information identified from the business license with the address of "Road D2, District C2, City B, Province A" is identified as replacement information; similarly, the third text information identified from the ID card with the name "C" is identified as replacement information.

[0099] In step S603, if the third text information is replacement information, the RPA robot replaces part of the retained customer information with the third text information. For example, the third text information "A Province B City C2 District D2 Road" identified from the business license with the address "A Province B City C2 District D2 Road" is used to replace the corresponding part of the retained customer information "A Province B City C1 District D1 Road". The retained customer information after the replacement becomes "Business License: Unified Social Security Code - AA123456; Address - A Province B City C2 District D2 Road; Legal Representative: Name - A, ID Number - 112233; Financial Person in Charge: Name - B, ID Number - 445566"; The third text information "Name: C, ID Number 778899" identified from the ID card of the name "C" replaces the corresponding portion of the retained customer information "Name - A, ID Number - 112233". The retained customer information after the replacement becomes "Business License: Unified Social Security Code - AA123456; Address - D2 Road, C2 District, B City, Province A; Legal Representative: Name - C, ID Number - 778899; Financial Person in Charge: Name - B, ID Number - 445566". At the same time, the updated ID image from which the third text information, representing the replacement information, originated can be stored, replacing the corresponding ID image region in the retained ID image.

[0100] In step S604, if the third text information is newly added, the RPA robot adds the same content as the third text information to the retained customer information. For example, when using a linked list to store the retained customer information, the third text information can be written to the linked list starting at the end of the original retained customer information. Simultaneously, the ID image region from which the newly added third text information originated can be stored.

[0101] By executing steps S601-S604, the customer can determine whether to replace part of the retained customer information with the photographed document or to add it to the retained customer information without changing the original part of the retained customer information by placing the photographed document in the direction of the photographed document, thereby reducing the need to fill out electronic forms and other materials and simplifying the customer's operation.

[0102] An RPA robot that executes the bank customer information change processing method in this embodiment can be combined with an OCR program, and the computer program can be written into a computer device or storage medium. When the computer program is read out and run, the bank customer information change processing method in this embodiment is executed, thereby achieving the same technical effect as the bank customer information change processing method in the embodiment.

[0103] It should be noted that, unless otherwise specified, when a feature is referred to as being "fixed" or "connected" to another feature, it may be directly fixed or connected to the other feature, or it may be indirectly fixed or connected to the other feature. In addition, the descriptions of up, down, left, right, etc. used in this disclosure are only relative to the relative positional relationship of the components of the present disclosure in the accompanying drawings. The singular forms of "a", "said" and "the" used in this disclosure are also intended to include the plural forms, unless the context clearly indicates otherwise. In addition, unless otherwise defined, all technical and scientific terms used in this embodiment have the same meaning as those generally understood by those skilled in the art. The terms used in the description of this embodiment are only for describing specific embodiments and are not intended to limit the present invention. The term "and / or" used in this embodiment includes any combination of one or more related listed items.

[0104] It should be understood that, although the present disclosure may adopt the term first, second, third etc. to describe various elements, these elements should not be limited to these terms.These terms are only used to distinguish the elements of the same type from each other.For example, without departing from the scope of the present disclosure, the first element may also be referred to as the second element, and similarly, the second element may also be referred to as the first element.The use of any and all examples or exemplary language ("for example", "such as" etc.) provided by the present embodiment is only intended to better illustrate embodiments of the present invention, and unless otherwise required, the scope of the present invention will not be limited.

[0105] It should be appreciated that embodiments of the present invention can be implemented or practiced by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable memory. The methods can be implemented in a computer program using standard programming techniques, including a non-transitory computer-readable storage medium configured with a computer program, wherein the storage medium so configured causes the computer to operate in a specific and predefined manner, according to the methods and figures described in the specific embodiments. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system. However, if desired, the program can be implemented in assembly or machine language. In any case, the language can be a compiled or interpreted language. In addition, the program can be run on a programmed application-specific integrated circuit for this purpose.

[0106] In addition, the operations of the processes described in this embodiment may be performed in any suitable order, unless otherwise indicated in this embodiment or otherwise clearly contradicted by the context. The processes described in this embodiment (or variations and / or combinations thereof) may be performed under the control of one or more computer systems configured with executable instructions, and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) that is executed collectively on one or more processors, by hardware, or a combination thereof. The computer program includes a plurality of instructions that can be executed by one or more processors.

[0107] Furthermore, the methods can be implemented in any type of computing platform that is operably connected to a suitable computer, including but not limited to a personal computer, a minicomputer, a mainframe, a workstation, a network or distributed computing environment, a separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, etc. Various aspects of the present invention can be implemented as machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, an optical read and / or write storage medium, RAM, ROM, etc., so that it can be read by a programmable computer, and when the storage medium or device is read by the computer, it can be used to configure and operate the computer to perform the processes described herein. In addition, the machine-readable code, or portions thereof, can be transmitted over a wired or wireless network. When such media includes instructions or programs that implement the steps described above in conjunction with a microprocessor or other data processor, the invention described in this embodiment includes these and other different types of non-transitory computer-readable storage media. When programmed according to the methods and techniques of the present invention, the present invention also includes the computer itself.

[0108] The computer program can be applied to input data to perform the functions described in the present embodiment, thereby converting the input data to generate output data that is stored in a non-volatile memory. The output information can also be applied to one or more output devices such as a display. In a preferred embodiment of the present invention, the converted data represents a physical and tangible object, including a specific visual depiction of the physical and tangible object produced on the display.

[0109] The above description is merely a preferred embodiment of the present invention. The present invention is not limited to the aforementioned embodiments. As long as the technical effects of the present invention are achieved by the same means, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention. Within the scope of protection of the present invention, various modifications and variations of the technical solutions and / or implementation methods are possible.

Claims

1. A method for changing and verifying bank customer information based on RPA and OCR, characterized by: The bank customer information change and verification processing method based on RPA and OCR includes: The RPA robot obtains the updated document image from the client; The RPA robot retrieves the retained ID image and retained customer information; the retained ID image and the retained customer information are retained on the bank system side, and the retained ID image, the retained customer information, and the updated ID image correspond to the same customer; The RPA robot merges the updated ID image with the retained ID image to obtain a merged ID image, and adds marking information to the merged ID image; the marking information marks each ID image region in the merged ID image as coming from the updated ID image or the retained ID image; The RPA robot calls the OCR algorithm to perform OCR recognition on the combined document image to obtain updated customer information; The RPA robot checks the updated customer information based on the retained customer information; When the updated customer information passes the inspection, the RPA robot updates the retained customer information according to the updated customer information; The step of merging the updated ID image with the retained ID image to obtain a merged ID image includes: The RPA robot identifies each document image region in the updated document image and the retained document image using an OCR algorithm; The RPA robot merges the document image regions into a single image according to a set relative position relationship to obtain the merged document image.

2. The bank customer information change and verification processing method based on RPA and OCR according to claim 1 is characterized in that: The RPA and OCR-based bank customer information change and verification processing method also includes: When the RPA robot fails to pass the inspection of the updated customer information, the relative position relationship between the areas of the certificate image is adjusted; The RPA robot merges the document image regions into a single image according to the new relative position relationship to obtain a new merged document image; The RPA robot recognizes the new merged document image and obtains the new updated customer information; The RPA robot checks the new updated customer information based on the retained customer information.

3. The bank customer information change and verification processing method based on RPA and OCR according to claim 2 is characterized in that: The RPA and OCR-based bank customer information change and verification processing method also includes: When the cumulative number of unqualified checks on the updated customer information reaches a threshold, the RPA robot refuses to update the retained customer information.

4. The method for changing and verifying bank customer information based on RPA and OCR according to any one of claims 1 to 3, characterized in that: The calling of an OCR algorithm to perform OCR recognition on the combined document image to obtain updated customer information includes: The RPA robot performs OCR recognition on the combined document image to obtain first text information; The RPA robot groups the first text information according to the ID image region from which it comes, to obtain second text information and third text information; wherein the second text information comes from the ID image region in the retained ID image, and the third text information comes from the ID image region in the updated ID image; performing cleaning processing on the second text information and the third text information; The cleaned second text information and the third text information constitute the updated customer information.

5. The method for changing and verifying bank customer information based on RPA and OCR according to claim 4 is characterized in that: The cleaning process of the second text information and the third text information includes: The RPA robot filters out illegal characters in the second text information and the third text information; The RPA robot calls the OCR algorithm to identify characteristic characters in the second text information and the third text information; The RPA robot segments the second text information and the third text information based on the characteristic characters.

6. The method for changing and verifying bank customer information based on RPA and OCR according to claim 5 is characterized in that: The updating of the retained customer information according to the updated customer information includes: An OCR algorithm identifies the orientation of the document image region from which the second text information and the third text information originate; The RPA robot identifies the third text information as replacement information or new information based on the orientation of the corresponding document image area; For the third text information that is replacement information, part of the content of the retained customer information is replaced with the third text information; wherein the orientation of the ID image area corresponding to the replaced part of the retained customer information is the same as the orientation of the ID image area from which the third text information comes; For the third text information that is newly added information, the RPA robot adds the same content as the third text information to the retained customer information.

7. The method for changing and verifying bank customer information based on RPA and OCR according to claim 4 is characterized in that: The checking of the updated customer information based on the retained customer information includes: The RPA robot compares the second text information with the retained customer information to obtain consistency; When the consistency reaches a preset threshold, the RPA robot determines that the inspection of the updated customer information is qualified; otherwise, it determines that the inspection of the updated customer information is unqualified.

8. A computer device, characterized in that: It includes a memory and a processor, the memory is used to store at least one program, and the processor is used to load the at least one program to execute the bank customer information change and verification processing method based on RPA and OCR as described in any one of claims 1 to 7.

9. A storage medium storing a program executable by a processor, characterized in that: The processor-executable program is used to execute the bank customer information change and verification processing method based on RPA and OCR when executed by the processor.

Images