Intelligent Detection Method, Device, Computer Equipment, Medium and Product for Business Information

Abstract

This application relates to the field of artificial intelligence technology, and provides an intelligent detection method, device, computer device, storage medium, and computer program product for business information, which can be specifically applied to the financial field or other related fields. This application can improve the efficiency and accuracy of business information detection. Using OCR recognition technology to replace manual review has the advantages of fast and accurate recognition, strong versatility, and easy operation, and can realize the full-process automatic processing of remote centralized authorization, reducing the waste of human resources. The method includes: obtaining an image containing the business information to be detected, using a preset area selection template to obtain each image area in the image that contains the business elements to be detected, identifying the business elements to be detected contained in each image area, detecting the business elements to be detected contained in each image area, obtaining the detection results corresponding to the business elements to be detected, and obtaining the detection result of the business information to be detected according to the detection results corresponding to each business element to be detected.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and particularly to a method, device, computer device, storage medium and computer program product for intelligent detection of business information. Background Art

[0002] With the development of information technology, business institutions such as banks have a large amount of business information to be detected every day. Relevant business personnel often need to review counter or non-counter business, and manually check and compare to detect whether the business information is correct.

[0003] This traditional technology relies on manual detection and processing, and the efficiency of business information detection through this technology is relatively low. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method, device, computer device, computer-readable storage medium and computer program product for intelligent detection of business information.

[0005] In a first aspect, the present application provides a method for intelligent detection of business information. The method includes:

[0006] Obtain an image containing business information to be detected;

[0007] Use a preset region selection template to obtain each image region in the image that contains business elements to be detected;

[0008] Identify the business elements to be detected included in each image region;

[0009] Detect the business elements to be detected included in each image region to obtain a detection result corresponding to the business element to be detected;

[0010] According to the detection results corresponding to each business element to be detected, obtain the detection result of the business information to be detected.

[0011] In one embodiment, the method further includes:

[0012] Obtain the type of business to be processed;

[0013] According to the corresponding relationship between the preset business type and the detection rule, determine the target detection rule corresponding to the type of business to be processed;

[0014] According to the relationship between the preset detection rule and the business element, determine the business element to be detected corresponding to the target detection rule;

[0015] Select the region selection template corresponding to the business element to be detected from the pre-stored multiple region selection templates as the preset region selection template.

[0016] In one embodiment, the method further includes:

[0017] Obtain a plurality of image samples corresponding to each type of business information;

[0018] Generate a plurality of image layouts corresponding to each type of business information according to the plurality of image samples;

[0019] Perform region selection annotation on the plurality of image layouts according to the positions of the image regions containing business elements in each image sample in each image sample, and obtain a plurality of region selection templates corresponding to each type of business information.

[0020] In one embodiment, before detecting the business elements to be detected included in each image region and obtaining the detection results corresponding to the business elements to be detected, the method further includes:

[0021] Obtain the type of business information obtained by recognizing the recognition image;

[0022] Determine whether the business information to be detected meets the information correctness detection rule corresponding to the type of business information;

[0023] If so, perform the step of detecting the business elements to be detected included in each image region and obtaining the detection results corresponding to the business elements to be detected.

[0024] In one embodiment, detecting the business elements to be detected included in each image region includes:

[0025] Determine the region where the signature element is located in each image region, and obtain the signature to be detected in the region where the signature element is located;

[0026] Detect the similarity between the signature to be detected and the historical signature;

[0027] If the similarity meets the preset similarity threshold condition, detect other business elements to be detected included in the image region.

[0028] In one embodiment,

[0029] Obtaining the signature to be detected in the region where the signature element is located includes:

[0030] Recognize the signature to be detected in the region where the signature element is located through the trained signature element detection model;

[0031] The method further includes:

[0032] Obtain an image sample containing a handwritten signature and the true signature text corresponding to the image sample;

[0033] Train the signature element detection model to be trained by using the image sample and the true signature text, and obtain the trained signature element detection model.

[0034] In a second aspect, the present application also provides a business information intelligent detection device. The device includes:

[0035] An image acquisition module, configured to acquire an image containing the business information to be detected;

[0036] A region acquisition module, configured to use a preset region selection template to acquire each image region in the image that contains the business elements to be detected;

[0037] An element recognition module, configured to recognize the business elements to be detected included in each of the image regions;

[0038] An element detection module, configured to detect the business elements to be detected included in each of the image regions, and obtain a detection result corresponding to the business element to be detected;

[0039] A result obtaining module, configured to obtain a detection result of the business information to be detected according to the detection results corresponding to each business element to be detected.

[0040] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0041] Acquire an image containing the business information to be detected; use a preset region selection template to acquire each image region in the image that contains the business elements to be detected; recognize the business elements to be detected included in each image region; detect the business elements to be detected included in each image region, and obtain a detection result corresponding to the business element to be detected; obtain a detection result of the business information to be detected according to the detection results corresponding to each business element to be detected.

[0042] In a fourth aspect, the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0043] Acquire an image containing the business information to be detected; use a preset region selection template to acquire each image region in the image that contains the business elements to be detected; recognize the business elements to be detected included in each image region; detect the business elements to be detected included in each image region, and obtain a detection result corresponding to the business element to be detected; obtain a detection result of the business information to be detected according to the detection results corresponding to each business element to be detected.

[0044] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0045] Obtain an image containing the business information to be detected; use a preset area selection template to obtain each image area in the image that contains the business elements to be detected; identify the business elements to be detected contained in each image area; detect the business elements to be detected contained in each image area to obtain the detection results corresponding to the business elements to be detected; and obtain the detection result of the business information to be detected according to the detection results corresponding to each business element to be detected.

[0046] The above-mentioned intelligent detection method, device, computer device, storage medium and computer program product for business information obtain an image containing the business information to be detected, use a preset area selection template to obtain each image area in the image that contains the business elements to be detected, identify the business elements to be detected contained in each image area, detect the business elements to be detected contained in each image area to obtain the detection results corresponding to the business elements to be detected, and obtain the detection result of the business information to be detected according to the detection results corresponding to each business element to be detected. This solution can be applied to remote centralized authorization. A remote authorization image data automatic detection device is constructed and a set of rule engines are set in advance to automatically process the image data detection according to the rules. By obtaining the image containing the business information to be detected after scanning the business information to be detected, applying the preset area selection template to this image, and obtaining the image selection area corresponding to this image, where each image area contains some business elements in the business information to be detected, identify the business elements to be detected contained in each image area through OCR recognition technology. Using OCR recognition technology instead of manual review has the advantages of fast and accurate recognition, strong versatility and easy operation. Then, detect whether the business elements to be detected contained in each image area are correct to obtain the detection results corresponding to the business elements to be detected, and obtain the detection result of the business information to be detected according to the detection results corresponding to each business element to be detected, thereby improving the efficiency and accuracy of business information detection and realizing the full-process automatic processing of remote centralized authorization, reducing the waste of human resources. Brief Description of the Drawings

[0047] Figure 1 It is a schematic flowchart of the intelligent detection method for business information in an embodiment;

[0048] Figure 2 It is an application scenario diagram of the intelligent detection method for business information in an embodiment;

[0049] Figure 3 It is a schematic flowchart of the intelligent detection method for business information in another embodiment;

[0050] Figure 4 It is a schematic flowchart of the intelligent detection method for business information in yet another embodiment;

[0051] Figure 5Schematic flowchart of steps for determining a region selection template in an embodiment;

[0052] Figure 6 Block diagram of a business information intelligent detection device in an embodiment;

[0053] Figure 7 Internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0054] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] In one embodiment, as Figure 1 shown, a business information intelligent detection method is provided. In this embodiment, it is exemplified that the method is applied to a server or a terminal, and includes the following steps:

[0056] Step S101, obtain an image containing the business information to be detected.

[0057] In this step, the business information to be detected may be information for handling a certain type of business to be handled. For example, the business information to be detected is a certificate such as Certificate A, Certificate B, etc.; as Figure 2 shown, the image containing the business information to be detected may be an image (such as video data) obtained by scanning the business information to be detected with a scanner, and the business information to be detected is displayed on the image.

[0058] Specifically, as Figure 2 shown, an image containing the business information to be detected is obtained by scanning the business information to be detected with a scanner. The scanner sends the image to the terminal of the front desk teller, and the terminal of the front desk teller sends the image to the server (which can be called an automatic review device for remote authorization video data). The server receives the image containing the business information to be detected (as Figure 3 shown, that is, obtain the video data input by the terminal).

[0059] Step S102, use a preset region selection template to obtain each image region in the image that contains the business elements to be detected.

[0060] In this step, the preset region selection template may be a template made according to the business information for determining the image regions to be selected in the image; the business information to be detected may include multiple business elements to be detected, and the business elements to be detected may be part of the business information to be detected.

[0061] Specifically, call the pre-set area selection module to determine the image areas to be selected in the image, where each image area to be selected contains business elements to be detected.

[0062] Step S103: Identify the business elements to be detected included in each image area.

[0063] Specifically, as Figure 3 shown, identify the business elements to be detected included in each image area through OCR recognition technology (OCR recognition module M002) to obtain the business elements to be detected.

[0064] Step S104: Detect the business elements to be detected included in each image area to obtain the detection results corresponding to the business elements to be detected.

[0065] In this step, the detection may include detecting whether the business elements to be detected included in each image area are within the validity period, are complete, have a certain seal, have information for online verification, are correct, are consistent with the business element information entered historically, etc.

[0066] Specifically, as Figure 3 and 4 shown, there can be multiple detection rules. The verification can be performed according to all the rules in the detection rules (all the rules in the verification rule engine). Use the machine learning model M003 to detect whether the materials are complete and detect the business elements to be detected included in each image area (verification of each element area) to obtain one or more detection results corresponding to the business elements to be detected.

[0067] Step S105: Obtain the detection result of the business information to be detected according to the detection results corresponding to each business element to be detected.

[0068] Specifically, as Figure 3 and 4 shown, obtain the detection result of the business information to be detected according to the detection results corresponding to each business element to be detected. For example, when the detection results corresponding to each business element to be detected are all correct (i.e., all verifications pass), obtain the detection result that the business information to be detected is correct, and relevant authorization passing processing can be performed for the business information to be detected, output the authorization result information and end the detection process; when the detection results corresponding to each business element to be detected are not all correct (i.e., not all verifications pass), obtain the detection result that the business information to be detected is incorrect, and relevant authorization rejection processing can be performed for the business information to be detected, output the authorization result information and end the detection process.

[0069] In the above intelligent detection method for business information, an image containing the business information to be detected is obtained, a preset region selection template is used to obtain each image region in the image that contains the business elements to be detected, the business elements to be detected contained in each image region are identified, the business elements to be detected contained in each image region are detected to obtain the detection results corresponding to the business elements to be detected, and based on the detection results corresponding to each business element to be detected, the detection result of the business information to be detected is obtained. This solution can be applied to remote centralized authorization. An automatic detection device for remote authorization image data is constructed and a set of rule engines are set in advance. According to the rules, automatic processing of image data detection is performed. By obtaining the image containing the business information to be detected after scanning the business information to be detected, the preset region selection template is applied to this image to obtain the image selection region corresponding to this image, where each image region contains some business elements in the business information to be detected. The business elements to be detected contained in each image region are identified through OCR recognition technology. Using OCR recognition technology to replace manual review has the advantages of fast and accurate recognition, strong versatility, and easy operation. Then, it is detected whether the business elements to be detected contained in each image region are correct to obtain the detection results corresponding to the business elements to be detected. Based on the detection results corresponding to each business element to be detected, the detection result of the business information to be detected is obtained, thereby improving the efficiency and accuracy of business information detection and enabling automatic processing of the entire process of remote centralized authorization, reducing the waste of human resources.

[0070] In one embodiment, the above method can also determine the preset region selection template through the following steps, which specifically include: obtaining the type of business to be processed; determining the target detection rule corresponding to the type of business to be processed according to the corresponding relationship between the preset business type and the detection rule; determining the business elements to be detected corresponding to the target detection rule according to the relationship between the preset detection rule and the business elements; and selecting the region selection template corresponding to the business elements to be detected from the multiple pre-stored region selection templates as the preset region selection template.

[0071] In this embodiment, the type of business to be processed may be a certain type of transaction business (such as the transaction of online modifying the name of a savings account). For example, in the counter business of a bank, there are various transaction businesses, and the type of business to be processed may be one of them. Each type of transaction business can be configured with a specific transaction code. According to the transaction code, it can be distinguished what transaction (business type) it is and what materials are required (such as what business information is required). Through the transaction code, the serial number of the materials that must be uploaded can be queried. The corresponding relationship between the preset business type and the detection rule means that each business type has a corresponding detection rule. For example, the detection rule corresponding to the business type of the transaction of online modifying the name of a savings account may be the detection rules such as the mandatory input items of the vouchers corresponding to business information such as certificates, cards, remote authorization applications, and signature service confirmation letters, the consistency check of voucher linkage, the check of voucher validity period, the check of card number, and the signature matching check. The relationship between the preset detection rule and the business element means that each detection rule has a corresponding business element to be detected. For example, the business element to be detected corresponding to the detection rule of the transaction of online modifying the name of a savings account may be the card number of the card.

[0072] Specifically, as Figure 3 shown, the rule engine module M001 is used to provide processing rules for the remote authorization image data automatic review device, and can abstract corresponding rules according to the actual business scenario and provide them for the device to use. One of the rules of the rule engine may be to establish the numbering rule of business information (vouchers). First, the business information can be classified and numbered in the AABB format, where AA can be the major category (such as the major category of vouchers), and BB is the minor category (such as the minor category of vouchers). An example of the numbering of business information is shown in Table 1 below:

[0073]

[0074]

[0075] Table 1

[0076] If the service information is a certificate, the certificate can be represented by a service information number of 01. Service information such as Certificate A and Certificate B belonging to the certificate can be represented by service information numbers of 0100 and 0101 respectively. If the service information is a card, the card can be represented by a service information number of 02. Card A and Card B belonging to the card can be represented by service information numbers of 0200 and 0201 respectively. If the service information is a remote authorization application form, the remote authorization application form can be represented by a service information number of 0300. If the service information is a signing service confirmation letter, the signing service confirmation letter can be represented by a service information number of 0400. The service information and the corresponding service information numbers will not be elaborated here. Detection rules corresponding to each service type can be preset, and business elements corresponding to each detection rule can be set. Obtain the service type to be processed. According to the corresponding relationship between the preset service type and the detection rule, determine the target detection rule corresponding to the service type to be processed. According to the relationship between the preset detection rule and the business elements to be detected, determine the business elements to be detected corresponding to the target detection rule. Select the region selection template corresponding to the business elements to be detected from multiple pre-stored region selection templates, and use the selected region selection template to obtain each image region containing the business elements to be detected in the image containing the service information to be detected.

[0077] The technical solution of this embodiment is beneficial to more accurately obtain the preset region selection template by selecting the region selection template corresponding to the service type to be processed from multiple pre-stored region selection templates as the preset region selection template, thereby facilitating more accurate acquisition of each image region of the business elements to be detected subsequently.

[0078] In one embodiment, the above method can also obtain multiple region selection templates corresponding to each service information type through the following steps, specifically including: obtaining multiple image samples corresponding to each service information type; generating multiple image layouts corresponding to each service information type according to the multiple image samples; performing region annotation on the multiple image layouts according to the positions of the image regions containing the business elements in each image sample in each image sample, and obtaining multiple region selection templates corresponding to each service information type.

[0079] In this embodiment, the types of business information may refer to the types of business information. For example, the types of business information may be the types of vouchers, which may include Document A, Document B, Card A, or Card B, etc.; the image samples may refer to the image samples corresponding to each type of business information. For example, the sample images obtained by scanning the business information corresponding to each type of business information with a scanner, or the images containing the business information to be detected obtained; the image layout may be the image layout obtained after making layout configurations for each type of business information (voucher type) based on multiple image samples; the selection area may refer to one or more image selection areas on each image layout, which may be rectangular image areas or circular image areas.

[0080] Specifically, before obtaining the type of business to be handled, multiple image samples corresponding to various voucher types (types of business information) can be obtained. Multiple image layouts corresponding to each voucher type are generated based on the multiple image samples. According to the positions of the image areas containing business elements in each image sample among the various image samples, corresponding area markings are made on the layout configuration, and the positions of the areas are outlined for the business elements to be detected or the business elements that may need to be detected. A voucher element (business information element) number can be set for each area. On the basis of the business information number, the element number can be added and extended into the AABBCC format, where the first 4 digits represent the business information (voucher), and the last 2 digits represent the element number (for example, the document text element can be set as 010001, the document digital element can be set as 010002, etc.), obtaining multiple area selection templates corresponding to each voucher type. Each of the area selection templates is used to select different business elements. For each type of business information, there is one or more corresponding area selection templates. The area selection templates corresponding to the same type of business information are used to select different business elements of this type of business information, that is, the multiple area selection templates corresponding to each type of business information obtained also correspond to the business elements.

[0081] The technical solution of this embodiment is beneficial to selecting a more accurate and appropriate area selection template when selecting a preset area selection template by obtaining multiple area selection templates, thereby facilitating more accurately obtaining each image area of the business elements to be detected subsequently.

[0082] In one embodiment, the above method may also perform the step of detecting the business elements to be detected included in each image area respectively to obtain the detection results corresponding to the business elements to be detected through the following steps, which specifically include: obtaining the type of business information obtained by recognizing the image; determining whether the business information to be detected meets the information correctness detection rule corresponding to the type of business information; if so, performing the step of detecting the business elements to be detected included in each image area respectively to obtain the detection results corresponding to the business elements to be detected.

[0083] In this embodiment, the information correctness detection rule may be a detection rule for information correctness and information validity. For example, it detects whether there are required fields that are not filled, whether the validity period corresponding to the service information is valid, checks whether the card number information is correct, and detects whether the service information is consistent with the input information, etc.

[0084] Specifically, before detecting the service elements to be detected included in each image region and obtaining the detection results corresponding to the service elements to be detected, as Figure 5 shown, through the recognition of the format of the service information (voucher) (including the recognition of the header, lines, confusing combinations, etc. First, recognize according to the fixed header features, then according to the line features. If there are line confusing combinations, then recognize the confusing header), after recognizing which type of service information the service information (voucher) belongs to, match the corresponding service information (voucher) number through the rule engine (subsequently, the text or image information of the region can be extracted according to the format configuration region of the rule engine). According to the recognized service information, call the information correctness detection rule corresponding to this type of service information, and judge whether the service information to be detected meets this information correctness detection rule. If so, execute the step of detecting the service elements to be detected included in each image region and obtaining the detection results corresponding to the service elements to be detected. If not, the detection result that does not meet the information correctness detection rule can be returned.

[0085] The technical solution of this embodiment is beneficial to the service information to be detected being within the validity period and conforming to the basic logical correctness by judging whether the service information to be detected meets the information correctness detection rule corresponding to the type of service information, thereby facilitating the improvement of the accuracy of the detection results.

[0086] In one embodiment, the detection of the service elements to be detected included in each image region in the above step S104 specifically includes: determining the region where the signature element is located in each image region, obtaining the signature to be detected in the region where the signature element is located; detecting the similarity between the signature to be detected and the historical signature; if the similarity meets the preset similarity threshold condition, then detect the other service elements to be detected included in the image region.

[0087] In this embodiment, the signature element may be the signature of the user or the handwritten font of the user; the signature to be detected may be the signature of the user or the handwritten font of the user; the historical signature may be the historical signature of the user or the historical handwritten font of the user; the similarity threshold condition may be the condition that the similarity is greater than Z%.

[0088] Specifically, after recognizing the service elements to be detected included in each image region, as Figure 4 and Figure 5As shown, the corresponding detection rules (rule engine information) can be retrieved according to the business information or the type of business information (voucher or voucher type). The template is selected for the image area of the business elements to be archived according to the corresponding area of the detection rules (layout configuration area according to the rule engine). The image area of the business elements to be archived is cut, the area where the signature element is located in each image area is determined, and the image of the signature to be detected in the area where the signature element is located is obtained (which can be called the current signature fragment, and the element fragment can be stored in binary form). The image of the signature to be detected and the image of the historical signature (which can be called the historical signature fragment) are subjected to similarity detection to obtain the corresponding similarity. If the similarity meets the preset similarity threshold condition (i.e., the similarity is relatively high), it can be determined that the person corresponding to the business information to be detected has handled it. Then, other business elements to be detected included in the image area are detected. If the similarity does not meet the preset similarity threshold condition (i.e., the similarity is relatively low), it is determined that it is not handled by the person corresponding to the business information to be detected, and it may be a proxy or impersonation, etc. In this case, additional information detection is required.

[0089] In the technical solution of this embodiment, by detecting whether the similarity meets the preset similarity threshold condition, it is beneficial to improve the accuracy of signature detection of the business information to be detected, and verify the identity of the applicant for the business information to be detected. Therefore, it is beneficial to prevent criminals from impersonating the person corresponding to the business information to be detected to handle business information and improve the security of business handling.

[0090] In one embodiment, the above method can also obtain a trained signature element detection model and identify the signature to be detected in the area where the signature element is located through the trained signature element detection model, specifically including: obtaining an image sample containing a handwritten signature and the corresponding real signature text of the image sample; training the signature element detection model to be trained by using the image sample and the real signature text to obtain a trained signature element detection model; identifying the signature to be detected in the area where the signature element is located through the trained signature element detection model.

[0091] In this embodiment, the image sample containing a handwritten signature can be the signature of the user himself / herself and / or a non-user, or the handwritten font of the user himself / herself and / or a non-user; the real signature text corresponding to the image sample can be the real signature of the user himself / herself, or the real handwritten font of the user himself / herself; the signature element detection model to be trained can be a machine learning model.

[0092] Specifically, obtain an image sample containing the user's and non-user's handwritten signature (or handwritten font) and an image sample (or text) of the user's real signature (or real handwritten font) corresponding to the image sample, and input them into the machine learning model to be trained for training to obtain a trained machine learning model. Identify the signature to be detected (or handwritten font to be detected) in the area where the signature elements are located through the trained machine learning model.

[0093] The technical solution of this embodiment uses the trained signature element detection model to identify the signature to be detected in the area where the signature elements are located, which is beneficial to improving the accuracy of obtaining the signature to be detected in the area where the signature elements are located, thereby facilitating the subsequent improvement of the accuracy of signature detection of the business information to be detected, and thus improving the security of business handling.

[0094] It should be understood that although the steps in the flowcharts involved in the above-mentioned embodiments are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps or stages in other steps.

[0095] Based on the same inventive concept, the embodiments of the present application also provide a business information intelligent detection device for implementing the above-mentioned business information intelligent detection method. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the business information intelligent detection device provided below can refer to the limitations on the business information intelligent detection method in the above text, and will not be repeated here.

[0096] In one embodiment, as Figure 6 shown, a business information intelligent detection device is provided. The device 600 may include:

[0097] An image acquisition module 601, configured to acquire an image containing the business information to be detected;

[0098] A region acquisition module 602, configured to use a preset region selection template to acquire each image region in the image that contains the business elements to be detected;

[0099] An element recognition module 603, configured to recognize the business elements to be detected included in each of the image regions;

[0100] An element detection module 604, configured to detect service elements to be detected included in each of the image regions, and obtain a detection result corresponding to the service element to be detected;

[0101] A result obtaining module 605, configured to obtain a detection result of the service information to be detected according to the detection results corresponding to the service elements to be detected.

[0102] In one embodiment, the apparatus 600 further includes: a region selection template selection module, configured to obtain a service type to be processed; determine a target detection rule corresponding to the service type to be processed according to a corresponding relationship between a preset service type and a detection rule; determine service elements to be detected corresponding to the target detection rule according to a relationship between a preset detection rule and service elements; and select, as the preset region selection template, a region selection template corresponding to the service element to be detected from multiple pre-stored region selection templates.

[0103] In one embodiment, the apparatus 600 further includes: a region selection template generation module, configured to obtain multiple image samples corresponding to each service information type; generate multiple image layouts corresponding to each service information type according to the multiple image samples; and perform region annotation on the multiple image layouts according to positions of image regions including service elements in each of the image samples in each of the image samples, to obtain multiple region selection templates corresponding to each service information type.

[0104] In one embodiment, the apparatus 600 further includes: an information correctness detection rule judgment module, configured to obtain a service information type obtained by recognizing the image; judge whether the service information to be detected meets an information correctness detection rule corresponding to the service information type; and if so, execute the step of detecting service elements to be detected included in each of the image regions to obtain a detection result corresponding to the service element to be detected.

[0105] In one embodiment, the element detection module 604 is further configured to determine a region where a signature element is located in each of the image regions, obtain a signature to be detected in the region where the signature element is located; detect a similarity between the signature to be detected and a historical signature; and if the similarity meets a preset similarity threshold condition, detect other service elements to be detected included in the image region.

[0106] In one embodiment, the element detection module 604 is further configured to identify a signature to be detected in the area where the signature element is located through a trained signature element detection model; the apparatus 600 further includes: a model training module, configured to obtain an image sample including a handwritten signature and a true signature text corresponding to the image sample; and use the image sample and the true signature text to train the signature element detection model to be trained, so as to obtain the trained signature element detection model.

[0107] Each module in the above business information intelligent detection apparatus can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or stored in a memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0108] It should be noted that the method and apparatus for intelligent detection of business information provided in this application can be used in application fields involving intelligent detection of business information in the financial field, and can also be used in the processing of intelligent detection of business information in any field other than the financial field. The application fields of the method and apparatus for intelligent detection of business information provided in this application are not limited.

[0109] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 7 shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device can be used to store preset area selection template data. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer device further includes an input / output interface. The input / output interface is a connection circuit for exchanging information between the processor and external devices, and they are connected to the processor through a bus, simply referred to as an I / O interface. When the computer program is executed by the processor, it implements a method for intelligent detection of business information.

[0110] Those skilled in the art can understand that Figure 7 the structure shown in

[0111] In one embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0112] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0113] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0114] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties.

[0115] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include Read-Only Memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0116] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0117] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. An intelligent detection method for service information, characterized in that, The method includes: Obtaining the type of business to be processed; Determining the target detection rule corresponding to the type of business to be processed according to the correspondence between the preset business type and the detection rule; Determining the business elements to be detected corresponding to the target detection rule according to the relationship between the preset detection rule and the business elements; Selecting the region selection template corresponding to the business elements to be detected from multiple pre-stored region selection templates as the preset region selection template; Scanning the business information to be detected through a scanner to obtain an image containing the business information to be detected; wherein, the business information to be detected is the information for processing the type of business to be processed; Using the preset region selection template to obtain each image region in the image that contains the business elements to be detected; Identifying the business elements to be detected included in each image region; Detecting the business elements to be detected included in each image region to obtain the detection result corresponding to the business elements to be detected; Obtaining the detection result of the business information to be detected according to the detection results corresponding to each business element to be detected; The detecting the business elements to be detected included in each image region includes: Determining the region where the signature element is located in each image region, and obtaining the signature to be detected in the region where the signature element is located; Detecting the similarity between the signature to be detected and the historical signature; If the similarity meets the preset similarity threshold condition, then detecting other business elements to be detected included in the image region; The obtaining the signature to be detected in the region where the signature element is located includes: Identifying the signature to be detected in the region where the signature element is located through a trained signature element detection model; The method further includes: Obtaining an image sample containing a handwritten signature and the true signature text corresponding to the image sample; Training the signature element detection model to be trained using the image sample and the true signature text to obtain the trained signature element detection model.

2. The method according to claim 1, wherein The method further includes: Obtaining multiple image samples corresponding to each type of business information; Generating multiple image layouts corresponding to each type of business information according to the multiple image samples; Performing region annotation on the multiple image layouts according to the positions of the image regions containing business elements in each image sample in each image sample to obtain multiple region selection templates corresponding to each type of business information.

3. The method according to claim 1, wherein Before detecting the business elements to be detected included in each image region to obtain the detection result corresponding to the business elements to be detected, the method further includes: Obtaining the type of business information obtained by recognizing the image; Judging whether the business information to be detected meets the information correctness detection rule corresponding to the type of business information; If so, performing the step of detecting the business elements to be detected included in each image region to obtain the detection result corresponding to the business elements to be detected.

4. An intelligent detection device for service information, characterized in that, The device includes: An image acquisition module, configured to scan the business information to be detected through a scanner to obtain an image containing the business information to be detected; wherein, the business information to be detected is the information for processing the type of business to be processed; A region acquisition module, configured to use a preset region selection template to acquire each image region in the image that contains business elements to be detected; An element recognition module, configured to recognize the business elements to be detected included in each of the image regions; An element detection module, configured to detect the business elements to be detected included in each of the image regions, and obtain a detection result corresponding to the business element to be detected; A result obtaining module, configured to obtain a detection result of the business information to be detected according to the detection results corresponding to the business elements to be detected; A region selection template selection module, configured to obtain the type of business to be handled; determine a target detection rule corresponding to the type of business to be handled according to the corresponding relationship between the preset business type and the detection rule; determine the business element to be detected corresponding to the target detection rule according to the relationship between the preset detection rule and the business element; select the region selection template corresponding to the business element to be detected from a plurality of pre-stored region selection templates as the preset region selection template; The element detection module is further configured to determine the region where the signature element is located in each of the image regions, obtain the signature to be detected in the region where the signature element is located; detect the similarity between the signature to be detected and the historical signature; if the similarity meets the preset similarity threshold condition, then detect other business elements to be detected included in the image region; The element detection module is further configured to recognize the signature to be detected in the region where the signature element is located through a trained signature element detection model; A model training module, configured to obtain an image sample containing a handwritten signature and the true signature text corresponding to the image sample; use the image sample and the true signature text to train a signature element detection model to be trained, and obtain the trained signature element detection model.

5. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 3 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 3 are implemented.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images