Risk identification method and device, computer equipment and storage medium

Abstract

The embodiment of the invention belongs to the technical field of image processing, and relates to a risk identification method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a to-be-identified original image set; performing image segmentation processing on the original image set to obtain an image segmentation result; performing feature extraction operation on the background segmentation image to obtain background image features; performing text recognition operation on the text segmentation image to obtain a text recognition result; performing integration processing on the background image features and the text recognition result to obtain comprehensive feature data; calling a historical database, and performing retrieval operation on the comprehensive feature data in the historical database to obtain a retrieval result sequence; and inputting the retrieval result sequence into a cluster case aggregation model for aggregation processing to obtain a risk aggregation result. According to the method, the early warning capability of a single risk is not influenced, and more accurate comprehensive risk judgment can be carried out by utilizing information fusion obtained by multiple models.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a risk identification method, device, computer equipment and storage medium. Background Art

[0002] In the field of auto credit, due to the development of digital technology, institutions and the public have gradually integrated into the digital wave, making online video remote review technology flourish. At the same time, along with the relevant loopholes in online remote approval, the fraudulent behavior of gangs has gradually surfaced.

[0003] In the auto dealer credit approval scenario, there are often cases of gang fraud and individual dealer fraud, such as existing auto finance dealers uploading home visit photos they have applied for in the past, new auto finance dealers uploading photos of other dealers' venues, different dealers using the same home visit photos, and uploaded photos being processed with Photoshop.

[0004] These often bring huge financial losses to credit companies. At the same time, it will also consume a lot of manpower and material resources to collect evidence from multiple parties, and it is very difficult to manually find gang fraud cases in massive data sets. In addition, the scene data and judgment logic are relatively complex, and there is no overall solution that fits the business needs well. For example, by using image retrieval technology in deep learning, image matching and similarity effects can be achieved visually, but in the actual application of specific business, the technology that relies solely on image retrieval often cannot meet the usage indicators that meet the business accuracy. It can be seen that the traditional risk identification method has the problem of low accuracy. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a risk identification method, apparatus, computer device and storage medium to solve the problem of low accuracy of traditional risk identification methods.

[0006] In order to solve the above technical problems, the embodiment of the present application provides a risk identification method, which adopts the following technical solution:

[0007] Obtaining a set of original images to be identified;

[0008] Performing image segmentation processing on the original image set to obtain image segmentation results, wherein the image segmentation results include a background segmentation image and a text segmentation image;

[0009] Performing a feature extraction operation on the background segmentation image to obtain background image features;

[0010] Performing a text recognition operation on the text segmentation image to obtain a text recognition result;

[0011] Integrate the background image features and the text recognition results to obtain comprehensive feature data;

[0012] Calling a historical database, performing a search operation on the comprehensive feature data in the historical database, and obtaining a search result sequence;

[0013] The search result sequence is input into the group case aggregation model for aggregation processing to obtain a risk aggregation result.

[0014] Furthermore, after the step of acquiring the original image set to be identified and before the step of performing image segmentation processing on the original image set to obtain the image segmentation result, the following steps are also included:

[0015] Performing an initial risk detection operation on the original image set according to the initial risk detection model to obtain an initial risk detection result, wherein the initial risk detection result includes the presence of an initial risk or the absence of an initial risk, the initial risk includes that the original image set is a PS image, and the initial risk also includes that a living body exists in the original image set;

[0016] If the initial risk detection result shows an initial risk, a risk warning signal is output;

[0017] If the initial risk detection result shows that there is no initial risk, image segmentation processing is performed on the original image set to obtain the image segmentation result.

[0018] Furthermore, the image segmentation result also includes a face segmentation image, and after the step of performing image segmentation processing on the original image set to obtain the image segmentation result, and before the step of integrating the background image features and the text recognition results to obtain comprehensive feature data, the following steps are also included:

[0019] Performing an identity recognition operation on the face segmentation image to obtain an identity recognition result;

[0020] The step of integrating the background image features and the text recognition results to obtain comprehensive feature data specifically includes the following steps:

[0021] Integrate the background image features, the text recognition results, and the identity recognition results to obtain the comprehensive feature data;

[0022] The step of calling the historical database, performing a search operation on the comprehensive feature data in the historical database, and obtaining a search result sequence specifically includes the following steps:

[0023] The historical customer manager investigation database is called, and a search operation is performed on the comprehensive feature data in the historical customer manager investigation database to obtain a search result sequence.

[0024] Furthermore, after the step of performing an identity recognition operation on the face segmentation image to obtain an identity recognition result, the following steps are also included:

[0025] Performing an identity authentication operation on the identity recognition result according to the text recognition result to obtain an identity authentication result;

[0026] If the identity verification result is abnormal, a risk warning signal is output;

[0027] If the identity verification result is normal, a normal identity signal is output.

[0028] Furthermore, the image segmentation result also includes a vehicle segmentation image, and after the step of performing image segmentation processing on the original image set to obtain the image segmentation result, and before the step of integrating the background image features and the text recognition results to obtain comprehensive feature data, the following steps are also included:

[0029] Performing vehicle segmentation processing on the background segmentation image according to the vehicle segmentation image to obtain a vehicle segmentation image;

[0030] The step of performing a feature extraction operation on the background segmentation image to obtain background image features specifically includes the following steps:

[0031] Performing a feature extraction operation on the vehicle segmentation image to obtain background image features;

[0032] The step of calling the historical database, performing a search operation on the comprehensive feature data in the historical database, and obtaining a search result sequence specifically includes the following steps:

[0033] A historical dealer site database is called, and a search operation is performed on the comprehensive feature data in the historical dealer site database to obtain a search result sequence.

[0034] Furthermore, the step of inputting the search result sequence into the group case aggregation model for aggregation processing to obtain the risk aggregation result specifically includes the following steps:

[0035] Performing an image matching operation on the retrieval result sequence according to a preset matching degree threshold to obtain a matching sequence image;

[0036] Constructing a relationship graph according to the original image set and the matching sequence image;

[0037] A risk aggregation operation is performed on the relationship graph according to a community discovery algorithm to obtain the risk aggregation result.

[0038] In order to solve the above technical problems, the embodiment of the present application further provides a risk identification device, which adopts the following technical solution:

[0039] The original image acquisition module is used to acquire the original image set to be identified;

[0040] An image segmentation module is used to perform image segmentation processing on the original image set to obtain image segmentation results, wherein the image segmentation results include a background segmentation image and a text segmentation image;

[0041] A feature extraction module is used to perform a feature extraction operation on the background segmentation image to obtain background image features;

[0042] A text recognition module is used to perform a text recognition operation on the text segmentation image to obtain a text recognition result;

[0043] An integration module, used for integrating the background image features and the text recognition results to obtain comprehensive feature data;

[0044] A retrieval module, used for calling a historical database, performing a retrieval operation on the comprehensive feature data in the historical database, and obtaining a retrieval result sequence;

[0045] The aggregation module is used to input the search result sequence into the group case aggregation model for aggregation processing to obtain risk aggregation results.

[0046] Furthermore, the device also includes:

[0047] an initial risk detection module, configured to perform an initial risk detection operation on the original image set according to an initial risk detection model to obtain an initial risk detection result, wherein the initial risk detection result includes the presence of an initial risk or the absence of an initial risk, the initial risk includes that the original image set is a PS image, and the initial risk also includes that a living body exists in the original image set;

[0048] A first initial risk detection result module, configured to output a risk warning signal if the initial risk detection result contains an initial risk;

[0049] The second initial risk detection result module is used to perform image segmentation processing on the original image set to obtain the image segmentation result if the initial risk detection result shows that there is no initial risk.

[0050] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0051] It includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the risk identification method as described above when executing the computer-readable instructions.

[0052] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0053] The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps of the risk identification method described above are implemented.

[0054] The present application provides a risk identification method, including: obtaining a set of original images to be identified; performing image segmentation processing on the original image set to obtain image segmentation results, wherein the image segmentation results include background segmentation images and text segmentation images; performing feature extraction operations on the background segmentation images to obtain background image features; performing text recognition operations on the text segmentation images to obtain text recognition results; integrating the background image features and the text recognition results to obtain comprehensive feature data; calling a historical database, performing a search operation on the comprehensive feature data in the historical database to obtain a search result sequence; inputting the search result sequence into a group case aggregation model for aggregation processing to obtain a risk aggregation result. Compared with the prior art, the present application inputs the search result sequence into the group case aggregation model for aggregation processing. Through the complex calculation and reasoning capabilities of the model, it is possible to comprehensively consider the risk characteristics of multiple similar cases, thereby obtaining a more accurate and reliable risk aggregation result. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0056] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0057] Figure 2 This is a flow chart for implementing the risk identification method provided in Example 1 of the present application;

[0058] Figure 3 is a schematic diagram of the structure of the risk identification device provided in Example 2 of the present application;

[0059] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0061] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0062] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0063] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0064] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0065] The terminal device 101 can be various electronic devices with a display screen and supporting web browsing. In addition to the laptop computer 1011, the tablet computer 1012 or the mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop computer, a desktop computer, etc.

[0066] The server 103 may be a server that provides various services, such as a background server that provides support for a web page displayed on the terminal device 101 .

[0067] It should be noted that the risk identification method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the risk identification device is generally set in the server / terminal device.

[0068] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0069] Embodiment 1

[0070] Continue to refer Figure 2 , shows a flow chart of an embodiment of a risk identification method according to the present application. The risk identification method comprises: step S201, step S202, step S203, step S204, step S205, step S206 and step S207.

[0071] In step S201, a set of original images to be identified is obtained.

[0072] In the embodiment of the present application, the original image set refers to the image basis required for evaluation when applying for automobile credit, wherein the original image set may include photos of dealer sites, customer manager inspection photos, etc. It should be understood that the examples of the original image set here are only for convenience of understanding and are not used to limit the present application.

[0073] In step S202, image segmentation processing is performed on the original image set to obtain image segmentation results, wherein the image segmentation results include a background segmentation image and a text segmentation image.

[0074] In the embodiment of the present application, the background segmentation image refers to the image after the interference elements are removed, which is mainly used to identify the background information of the image scene.

[0075] In the embodiments of the present application, a text segmentation image refers to an image element carrying text content. The text segmentation image may be a watermark image, a work badge image, etc. It should be understood that the examples of text segmentation images given here are only for ease of understanding and are not used to limit the present application.

[0076] In step S203, a feature extraction operation is performed on the background segmentation image to obtain background image features.

[0077] In the embodiment of the present application, the feature extraction operation is mainly used to identify important information in the image, and the feature extraction operation can be:

[0078] (1) Shape features: such as bounding box, perimeter, area, circularity, aspect ratio, etc., which are used to describe the basic shape attributes of an object;

[0079] (2) Texture features: Use gray-level co-occurrence matrix (GLCM), local binary pattern (LBP) and other methods to extract texture information of image regions;

[0080] (3) Color features: describing the color distribution and color attributes of an image through methods such as color histogram and color moment;

[0081] (4) Spatial relationship features: describing the position, direction, and spatial relationship of objects in the image;

[0082] (5) Advanced features: Use deep learning techniques (such as convolutional neural networks (CNNs)) to automatically learn and extract deep features of images. These features often have higher abstraction and better generalization capabilities.

[0083] In step S204, a text recognition operation is performed on the text segmentation image to obtain a text recognition result.

[0084] In the embodiment of the present application, a watermark detection model is used to determine the watermark position for the two types of input images, and then the OCR recognition technology is used to identify the corresponding time and address information of the area.

[0085] In step S205, the background image features and the text recognition results are integrated to obtain comprehensive feature data.

[0086] In the embodiment of the present application, the integration process is mainly used to combine the visual features obtained by image analysis with the text information extracted by text recognition technology to form a richer and more comprehensive data representation, so as to perform subsequent retrieval operations and improve the accuracy of retrieval results. The integration process can be:

[0087] (1) Feature concatenation: directly concatenate text features (such as word embedding vectors, TF-IDF vectors, etc.) and image features (such as CNN feature vectors) into a longer feature vector. This method is simple and intuitive, but may face the problem of too high feature dimension;

[0088] (2) Feature fusion: Use deep learning models (such as multimodal fusion networks) to fuse text features and image features at a deeper level. This method can learn the complex relationship between text and images and generate more representative comprehensive features;

[0089] (3) Attention mechanism: Introducing the attention mechanism in the feature fusion process enables the model to dynamically focus on the areas in the image that are most relevant to the text, thereby extracting more targeted comprehensive features.

[0090] In step S206, the historical database is called, and a search operation is performed on the comprehensive feature data in the historical database to obtain a search result sequence.

[0091] In step S207, the search result sequence is input into the group case aggregation model for aggregation processing to obtain a risk aggregation result.

[0092] In the embodiment of the present application, the group case aggregation model contains a risk setting module, which can customize the risk rules according to the information obtained automatically, such as face, time, address and other information, and filter the search result sequence to obtain the sequence under the corresponding rules. For example, when different dealers use the same home visit photo, the time address and other information will be consistent, and the same dealer uploads photos of the venue that have been submitted in previous years. After accumulating a certain amount of search results (parameters can be set), the initiating search image and the search result sequence image are connected in the form of a node relationship graph, and then the community discovery algorithm (such as infomap) is used to divide the relationship node graph into communities and aggregate group cases, so as to obtain a potential fraud group case sequence, and finally obtain a potential fraud group case sequence under the corresponding risk rules.

[0093] In an embodiment of the present application, a risk identification method is provided, including: obtaining an original image set to be identified; performing image segmentation processing on the original image set to obtain image segmentation results, wherein the image segmentation results include background segmentation images and text segmentation images; performing feature extraction operations on the background segmentation images to obtain background image features; performing text recognition operations on the text segmentation images to obtain text recognition results; integrating the background image features and the text recognition results to obtain comprehensive feature data; calling a historical database, performing a search operation on the comprehensive feature data in the historical database to obtain a search result sequence; inputting the search result sequence into a group case aggregation model for aggregation processing to obtain a risk aggregation result. Compared with the prior art, the present application inputs the search result sequence into a group case aggregation model for aggregation processing, and through the complex calculation and reasoning capabilities of the model, it is able to comprehensively consider the risk characteristics of multiple similar cases, thereby obtaining a more accurate and reliable risk aggregation result.

[0094] In some optional implementations of the embodiments of the present application, after step S201 and before step 202, the following steps are further included:

[0095] Performing an initial risk detection operation on the original image set according to the initial risk detection model to obtain an initial risk detection result, wherein the initial risk detection result includes the presence of an initial risk or the absence of an initial risk, the initial risk includes that the original image set is a PS image, and the initial risk also includes that a living body exists in the original image set;

[0096] If the initial risk detection result shows an initial risk, a risk warning signal is output;

[0097] If the initial risk detection result shows that there is no initial risk, the original image set is subjected to image segmentation processing to obtain an image segmentation result.

[0098] In the embodiment of the present application, the fraud types include PS detection, liveness detection, etc. At the same time, the solution reserves the ability to add new fraud types. If it is determined that there are risks such as PS and non-liveness, the corresponding risk warning is directly output without performing subsequent risk identification operations.

[0099] In the embodiments of the present application, efficient, accurate and safe processing of the original image set is achieved through rapid preliminary screening, immediate risk warning, precise segmentation processing, and improved overall processing efficiency, safety and reliability. This is of great significance for application scenarios that require processing large amounts of image data and conducting risk assessments.

[0100] In some optional implementations of the embodiments of the present application, the image segmentation result also includes a face segmentation image, and after step S202 and before step 205, the following steps are also included:

[0101] Performing identity recognition operation on the face segmentation image to obtain an identity recognition result;

[0102] Step S205 specifically includes the following steps:

[0103] Integrate the background image features, the text recognition results and the identity recognition results to obtain the comprehensive feature data;

[0104] Step S206 specifically includes the following steps:

[0105] The historical customer manager investigation database is called, and a search operation is performed on the comprehensive feature data in the historical customer manager investigation database to obtain a search result sequence.

[0106] In an embodiment of the present application, when a face segmentation image exists in the image segmentation result, the identity information of the face segmentation image is extracted and integrated into the comprehensive feature data, so as to construct more comprehensive feature data and improve the breadth and depth of data analysis; in addition, user identity information, as an important reference information, can reduce the errors and uncertainties of other data sources to a certain extent and improve the accuracy and reliability of data analysis.

[0107] In some optional implementations of the embodiments of the present application, after the step of performing an identity recognition operation on the face segmentation image to obtain an identity recognition result, the following steps are also included:

[0108] Performing an identity authentication operation on the identity recognition result according to the text recognition result to obtain an identity authentication result;

[0109] If the identity verification result is abnormal, a risk warning signal is output;

[0110] If the identity verification result is normal, a normal identity signal is output.

[0111] In the embodiment of the present application, identity is recognized through face recognition, and identity matching is performed through work badge detection and OCR recognition. If the match is consistent, subsequent risk identification operations are performed, otherwise a risk of identity information mismatch is warned.

[0112] In some optional implementations of the embodiments of the present application, the image segmentation result also includes a vehicle segmentation image, and after step S202 and before step 205, the following steps are also included:

[0113] Performing vehicle segmentation processing on the background segmentation image according to the vehicle segmentation image to obtain a vehicle segmentation image;

[0114] Step S203 specifically includes the following steps:

[0115] Performing a feature extraction operation on the vehicle segmentation image to obtain background image features;

[0116] Step S206 specifically includes the following steps:

[0117] A historical dealer site database is called, and a search operation is performed on the comprehensive feature data in the historical dealer site database to obtain a search result sequence.

[0118] In some optional implementations of the embodiments of the present application, the above step S207 specifically includes the following steps:

[0119] Performing an image matching operation on the retrieval result sequence according to a preset matching degree threshold to obtain a matching sequence image;

[0120] Constructing a relationship graph according to the original image set and the matching sequence image;

[0121] A risk aggregation operation is performed on the relationship graph according to a community discovery algorithm to obtain the risk aggregation result.

[0122] In the embodiment of the present application, the risk assessment and aggregation of the original image set and its related images are realized through steps such as image matching, relationship graph construction and community discovery algorithm. This method not only improves the accuracy and efficiency of risk assessment, but also provides strong support for subsequent risk response and decision-making.

[0123] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0124] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0125] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0126] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0127] Embodiment 2

[0128] Further references Figure 3 , as a response to the above Figure 2 In order to realize the method shown in the figure, the present application provides an embodiment of a risk identification device, and the device embodiment is Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0129] like Figure 3 As shown, the risk identification device 200 described in the embodiment of the present application includes:

[0130] The original image acquisition module 210 is used to acquire the original image set to be identified;

[0131] An image segmentation module 220 is used to perform image segmentation processing on the original image set to obtain an image segmentation result, wherein the image segmentation result includes a background segmentation image and a text segmentation image;

[0132] A feature extraction module 230 is used to perform a feature extraction operation on the background segmentation image to obtain background image features;

[0133] A text recognition module 240 is used to perform a text recognition operation on the text segmentation image to obtain a text recognition result;

[0134] An integration module 250 is used to integrate the background image features and the text recognition results to obtain comprehensive feature data;

[0135] A retrieval module 260 is used to call a historical database, perform a retrieval operation on the comprehensive feature data in the historical database, and obtain a retrieval result sequence;

[0136] The aggregation module 270 is used to input the search result sequence into the group case aggregation model for aggregation processing to obtain a risk aggregation result.

[0137] In an embodiment of the present application, a risk identification device 200 is provided, including: an original image acquisition module 210, used to acquire a set of original images to be identified; an image segmentation module 220, used to perform image segmentation processing on the original image set to obtain image segmentation results, wherein the image segmentation results include background segmentation images and text segmentation images; a feature extraction module 230, used to perform feature extraction operations on the background segmentation images to obtain background image features; a text recognition module 240, used to perform text recognition operations on the text segmentation images to obtain text recognition results; an integration module 250, used to integrate the background image features and the text recognition results to obtain comprehensive feature data; a retrieval module 260, used to call a historical database, perform a retrieval operation on the comprehensive feature data in the historical database, and obtain a retrieval result sequence; an aggregation module 270, used to input the retrieval result sequence into a group case aggregation model for aggregation processing to obtain a risk aggregation result. Compared with the prior art, this application inputs the search result sequence into the group case aggregation model for aggregation processing. Through the complex calculation and reasoning capabilities of the model, it can comprehensively consider the risk characteristics of multiple similar cases, thereby obtaining more accurate and reliable risk aggregation results.

[0138] In some optional implementations of the embodiments of the present application, the risk identification device 200 further includes:

[0139] an initial risk detection module, configured to perform an initial risk detection operation on the original image set according to an initial risk detection model to obtain an initial risk detection result, wherein the initial risk detection result includes the presence of an initial risk or the absence of an initial risk, the initial risk includes that the original image set is a PS image, and the initial risk also includes that a living body exists in the original image set;

[0140] A first initial risk detection result module, configured to output a risk warning signal if the initial risk detection result contains an initial risk;

[0141] The second initial risk detection result module is used to perform image segmentation processing on the original image set to obtain the image segmentation result if the initial risk detection result shows that there is no initial risk.

[0142] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of a computer device according to an embodiment of the present application.

[0143] The computer device 300 includes a memory 310, a processor 320, and a network interface 330 that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device 300 having components 310-330, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital processors (DSPs), embedded devices, etc.

[0144] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0145] The memory 310 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 310 can be an internal storage unit of the computer device 300, such as a hard disk or memory of the computer device 300. In other embodiments, the memory 310 can also be an external storage device of the computer device 300, such as a plug-in hard disk equipped on the computer device 300, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card, etc. Of course, the memory 310 can also include both the internal storage unit of the computer device 300 and its external storage device. In the embodiment of the present application, the memory 310 is generally used to store the operating system and various application software installed on the computer device 300, such as computer-readable instructions of the risk identification method, etc. In addition, the memory 310 can also be used to temporarily store various data that have been output or are to be output.

[0146] The processor 320 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 320 is generally used to control the overall operation of the computer device 300. In the embodiment of the present application, the processor 320 is used to run the computer-readable instructions stored in the memory 310 or process data, such as computer-readable instructions for running the risk identification method.

[0147] The network interface 330 may include a wireless network interface or a wired network interface. The network interface 330 is generally used to establish a communication connection between the computer device 300 and other electronic devices.

[0148] The computer device provided in this application inputs the search result sequence into the group case aggregation model for aggregation processing. Through the complex calculation and reasoning capabilities of the model, it can comprehensively consider the risk characteristics of multiple similar cases, thereby obtaining more accurate and reliable risk aggregation results.

[0149] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the risk identification method as described above.

[0150] The computer-readable storage medium provided in the present application inputs the search result sequence into the group case aggregation model for aggregation processing. Through the complex calculation and reasoning capabilities of the model, it can comprehensively consider the risk characteristics of multiple similar cases, thereby obtaining a more accurate and reliable risk aggregation result.

[0151] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0152] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A risk identification method, characterized in that: The steps include: Obtaining a set of original images to be identified; Performing image segmentation processing on the original image set to obtain image segmentation results, wherein the image segmentation results include a background segmentation image and a text segmentation image; Performing a feature extraction operation on the background segmentation image to obtain background image features; Performing a text recognition operation on the text segmentation image to obtain a text recognition result; Integrate the background image features and the text recognition results to obtain comprehensive feature data; Calling a historical database, performing a search operation on the comprehensive feature data in the historical database, and obtaining a search result sequence; The search result sequence is input into the group case aggregation model for aggregation processing to obtain a risk aggregation result.

2. The risk identification method according to claim 1, characterized in that: After the step of acquiring the original image set to be identified and before the step of performing image segmentation processing on the original image set to obtain an image segmentation result, the following steps are also included: Performing an initial risk detection operation on the original image set according to the initial risk detection model to obtain an initial risk detection result, wherein the initial risk detection result includes the presence of an initial risk or the absence of an initial risk, the initial risk includes that the original image set is a PS image, and the initial risk also includes that a living body exists in the original image set; If the initial risk detection result shows an initial risk, a risk warning signal is output; If the initial risk detection result shows that there is no initial risk, image segmentation processing is performed on the original image set to obtain the image segmentation result.

3. The risk identification method according to claim 1, characterized in that: The image segmentation result also includes a face segmentation image. After the step of performing image segmentation processing on the original image set to obtain the image segmentation result and before the step of integrating the background image features and the text recognition results to obtain comprehensive feature data, the following steps are also included: Performing an identity recognition operation on the face segmentation image to obtain an identity recognition result; The step of integrating the background image features and the text recognition results to obtain comprehensive feature data specifically includes the following steps: Integrate the background image features, the text recognition results, and the identity recognition results to obtain the comprehensive feature data; The step of calling the historical database, performing a search operation on the comprehensive feature data in the historical database, and obtaining a search result sequence specifically includes the following steps: The historical customer manager investigation database is called, and a search operation is performed on the comprehensive feature data in the historical customer manager investigation database to obtain a search result sequence.

4. The risk identification method according to claim 3, characterized in that: After the step of performing an identity recognition operation on the face segmentation image to obtain an identity recognition result, the following steps are also included: Performing an identity authentication operation on the identity recognition result according to the text recognition result to obtain an identity authentication result; If the identity verification result is abnormal, a risk warning signal is output; If the identity verification result is normal, a normal identity signal is output.

5. The risk identification method according to claim 1, characterized in that: The image segmentation result also includes a vehicle segmentation image. After the step of performing image segmentation processing on the original image set to obtain the image segmentation result and before the step of integrating the background image features and the text recognition results to obtain comprehensive feature data, the following steps are also included: Performing vehicle segmentation processing on the background segmentation image according to the vehicle segmentation image to obtain a vehicle segmentation image; The step of performing a feature extraction operation on the background segmentation image to obtain background image features specifically includes the following steps: Performing a feature extraction operation on the vehicle segmentation image to obtain background image features; The step of calling the historical database, performing a search operation on the comprehensive feature data in the historical database, and obtaining a search result sequence specifically includes the following steps: A historical dealer site database is called, and a search operation is performed on the comprehensive feature data in the historical dealer site database to obtain a search result sequence.

6. The risk identification method according to claim 1, characterized in that: The step of inputting the search result sequence into the group case aggregation model for aggregation processing to obtain the risk aggregation result specifically includes the following steps: Performing an image matching operation on the retrieval result sequence according to a preset matching degree threshold to obtain a matching sequence image; Constructing a relationship graph according to the original image set and the matching sequence image; A risk aggregation operation is performed on the relationship graph according to a community discovery algorithm to obtain the risk aggregation result.

7. A risk identification device, characterized in that: include: The original image acquisition module is used to acquire the original image set to be identified; An image segmentation module is used to perform image segmentation processing on the original image set to obtain image segmentation results, wherein the image segmentation results include a background segmentation image and a text segmentation image; A feature extraction module is used to perform a feature extraction operation on the background segmentation image to obtain background image features; A text recognition module is used to perform a text recognition operation on the text segmentation image to obtain a text recognition result; An integration module, used for integrating the background image features and the text recognition results to obtain comprehensive feature data; A retrieval module, used for calling a historical database, performing a retrieval operation on the comprehensive feature data in the historical database, and obtaining a retrieval result sequence; The aggregation module is used to input the search result sequence into the group case aggregation model for aggregation processing to obtain risk aggregation results.

8. The risk identification device according to claim 7, characterized in that: The device also includes: an initial risk detection module, configured to perform an initial risk detection operation on the original image set according to an initial risk detection model to obtain an initial risk detection result, wherein the initial risk detection result includes the presence of an initial risk or the absence of an initial risk, the initial risk includes that the original image set is a PS image, and the initial risk also includes that a living body exists in the original image set; A first initial risk detection result module, configured to output a risk warning signal if the initial risk detection result contains an initial risk; The second initial risk detection result module is used to perform image segmentation processing on the original image set to obtain the image segmentation result if the initial risk detection result shows that there is no initial risk.

9. A computer device comprising a memory and a processor, characterized in that: The memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the risk identification method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the risk identification method according to any one of claims 1 to 6 are implemented.

Images