Vehicle damage processing method, device and equipment based on large model, medium and product

Abstract

The embodiment of the invention provides a vehicle damage processing method and device based on a large model, equipment, a medium and a product, and belongs to the technical field of artificial intelligence. The method comprises the following steps: inputting a damage picture of a vehicle into a first model, identifying identity information, a damaged part and a damage degree of the vehicle through the first model, and determining a first damage value of the vehicle based on the identity information, the damaged part and the damage degree; inputting the first information into a second model, and identifying a claim settlement insurance fraud risk corresponding to the vehicle based on the first information through the second model; under the condition of determining that the vehicle does not have the insurance fraud risk based on the claim settlement insurance fraud risk, inputting the first damage value into a third model, and determining a claim settlement scheme corresponding to the vehicle based on the first damage value through the third model; and executing the claim settlement scheme in response to confirmation information of the user on the claim settlement scheme. The method can be applied to scenes such as financial science and technology and artificial intelligence, and the efficiency of an existing vehicle damage processing method can be improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, equipment, medium and product for vehicle damage processing based on a large model. Background Technology

[0002] As the number of cars on the road continues to increase, the number of people purchasing car insurance is also rising year by year. This not only brings more business opportunities to insurance companies, but also places higher demands on vehicle damage handling services.

[0003] Currently, traditional vehicle damage handling methods require strict human control at every stage. This is not only labor-intensive but also prone to human error, easily overlooking potential insurance fraud risks. Consequently, traditional vehicle damage handling methods are inefficient. Summary of the Invention

[0004] The main objective of this application is to propose a vehicle damage processing method, apparatus, equipment, medium, and product based on a large model, aiming to improve the efficiency of existing vehicle damage processing methods.

[0005] To achieve the above objectives, a first aspect of this application proposes a vehicle damage processing method based on a large model. The method includes: inputting a damaged image of a vehicle into a first model; identifying the vehicle's identity information, damaged parts, and damage extent through the first model; and determining a first damage value of the vehicle based on the identity information, the damaged parts, and the damage extent.

[0006] The first information is input into the second model, and the second model identifies the insurance fraud risk corresponding to the vehicle based on the first information; wherein, the first information includes at least one of the damage images and the owner's claim history.

[0007] If it is determined that the vehicle does not have the risk of insurance fraud based on the aforementioned insurance fraud risk, the first damage value is input into the third model, and the third model determines the corresponding claim plan for the vehicle based on the first damage value.

[0008] In response to the first user's confirmation of the claim settlement plan, the claim settlement plan is executed.

[0009] In some embodiments, after inputting the first information into a second model and identifying the insurance fraud risk corresponding to the vehicle based on the first information, and before determining that the vehicle does not have an insurance fraud risk based on the insurance fraud risk, inputting the first damage value into a third model and determining the corresponding claim plan for the vehicle based on the first damage value through the third model, the method further includes:

[0010] When the insurance fraud risk indicator suggests that the vehicle is suspected of insurance fraud, a first prompt message is output; wherein, the first prompt message is used to prompt the second user to confirm the insurance fraud risk corresponding to the vehicle;

[0011] In response to a first indication message received indicating that the vehicle is not at risk of insurance fraud, it is determined that the vehicle is not at risk of insurance fraud.

[0012] In some embodiments, after inputting the first information into a second model and identifying the insurance fraud risk corresponding to the vehicle based on the first information using the second model, the method further includes:

[0013] If it is determined that the vehicle is at risk of insurance fraud based on the aforementioned insurance fraud risk, a second prompt message is output; wherein, the second prompt message is used to prompt the first user to upload unedited pictures of the vehicle damage;

[0014] In response to the first user re-uploading images of the vehicle's damage, the process returns to inputting the images of the vehicle's damage into the first model. The first model identifies the vehicle's identity information, damaged parts, and degree of damage. Based on the identity information, the damaged parts, and the degree of damage, the first damage value of the vehicle is determined.

[0015] In some embodiments, the step of inputting the first damage value into a third model and determining the corresponding claim plan for the vehicle based on the first damage value using the third model includes:

[0016] The first damage value is input into the third model, and the third model determines the second damage value of the vehicle based on the first damage value by querying historical vehicle damage handling cases and industry standards.

[0017] Based on the second damage value, the corresponding compensation plan for the vehicle is determined.

[0018] In some embodiments, determining the compensation plan corresponding to the vehicle based on the second damage value includes:

[0019] In cases where the second damage value indicates that the vehicle is a total loss and cannot be repaired, the corresponding compensation plan for the vehicle is determined to be cash compensation.

[0020] If the second damage value represents partial loss of the vehicle and the repair cost is lower than the market value, the corresponding claim plan for the vehicle is determined to be repair service;

[0021] In cases where the second damage value represents damage to a component of the vehicle, the corresponding claim solution for the vehicle is determined to be a component replacement service.

[0022] In some embodiments, after executing the claim settlement plan in response to the first user's confirmation of the claim settlement plan, the method further includes:

[0023] Obtain the claims settlement plan and synchronize it to the big data system;

[0024] Reports are generated based on the big data system and sent to a second user via email or subscription account.

[0025] The feedback information from the first user and the report are input into the fourth model. Based on the report and the feedback information, the fourth model is optimized to obtain an optimized fourth model. The fourth model is used to optimize the subsequent claims settlement plan.

[0026] To achieve the above objectives, a second aspect of this application proposes a vehicle damage processing device based on a large model, the device comprising:

[0027] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0028] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0029] To achieve the above objectives, a fifth aspect of this application provides a computer program product stored in a storage medium, which is executed by at least one processor to implement the method described in the first aspect.

[0030] This application proposes a vehicle damage processing method, apparatus, equipment, medium, and product based on a large model. It inputs images of vehicle damage into a first model, which identifies the vehicle's identity information, damaged parts, and extent of damage. Based on these factors, a first damage value is determined. This first value is then input into a second model, which identifies the corresponding insurance fraud risk based on the first value. The first information includes at least one of the damaged images and the vehicle owner's claim history. If the insurance fraud risk is determined to be absent, the first damage value is input into a third model, which determines the corresponding claim plan. Upon user confirmation of the claim plan, the claim plan is executed. This application can determine the first damage value of a vehicle and identify insurance fraud risk through a model. When no insurance fraud risk is determined, a claim plan is determined based on the first damage value for compensation. This achieves automated vehicle damage processing, saves labor costs, reduces human error, enables a standardized process for vehicle damage processing, and provides fast, accurate results, thus improving the efficiency of vehicle damage processing methods. Attached Figure Description

[0031] Figure 1 This is a flowchart of the vehicle damage processing method based on a large model provided in an embodiment of this application;

[0032] Figure 2 yes Figure 1 The flowchart after step S102 and before step S103;

[0033] Figure 3 yes Figure 1 The flowchart following step S102;

[0034] Figure 4 yes Figure 1 The flowchart of step S103 in the process;

[0035] Figure 5 yes Figure 4 The flowchart of step S402 in the document;

[0036] Figure 6 yes Figure 1 The flowchart following step S104;

[0037] Figure 7 This is a schematic diagram illustrating the principle of the vehicle damage processing method based on a large model provided in the embodiments of this application;

[0038] Figure 8 This is a schematic diagram of the vehicle damage processing device based on a large model provided in an embodiment of this application;

[0039] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0041] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0043] First, let's analyze some of the terms used in this application:

[0044] Artificial intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.

[0045] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). NLP is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, intent recognition, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0046] Information extraction is a text processing technique that extracts factual information such as entities, relationships, and events from natural language text and outputs it as structured data. Information extraction is a technique for extracting specific information from text data. Text data is composed of specific units, such as sentences, paragraphs, and chapters. Text information is composed of smaller, specific units, such as characters, words, phrases, sentences, paragraphs, or combinations of these units. Extracting noun phrases, names of people, and place names from text data is an example of text information extraction. Of course, text information extraction techniques can extract information of various types.

[0047] Image captioning generates natural language descriptions for images, helping applications understand the semantics expressed in the visual scene. For example, image captioning can convert image retrieval into text retrieval, classify images, and improve retrieval results. While people can often describe the details of a visual scene with a quick glance, automatically adding descriptions to images is a comprehensive and challenging computer vision task, requiring the conversion of complex information contained in the image into natural language descriptions. Compared to ordinary computer vision tasks, image captioning not only requires identifying objects in an image but also associating the identified objects with natural semantics and describing them in natural language. Therefore, image captioning requires extracting deep features from the image, associating them with semantic features, and converting them to generate descriptions.

[0048] Insurance is a risk transfer mechanism whereby the insured (individual or business) transfers specific risks to an insurance company by paying premiums. The insurance company, through collecting premiums, establishes an insurance fund, which is used to provide financial compensation or pay insurance benefits to the insured for losses caused by insured events, under the conditions stipulated in the insurance contract. Property insurance refers to insurance that uses property and related interests as the insured object. It aims to protect the insured's property from losses caused by natural disasters, accidents, or other risks, and to provide financial compensation.

[0049] Auto insurance is a type of property insurance specifically designed to protect motor vehicles. It aims to compensate for vehicle damage, third-party personal injury or property loss, and related legal liabilities caused by traffic accidents or other unforeseen events. Auto insurance typically includes the following main types: Compulsory Traffic Accident Liability Insurance (CTALI), a legally mandated insurance for all motor vehicles, primarily used to compensate for losses to third parties (people and property); Commercial Auto Insurance, a voluntary insurance offering more comprehensive coverage, including vehicle damage insurance, third-party liability insurance, vehicle theft insurance, and passenger liability insurance; and Supplementary insurance, such as glass breakage insurance, vehicle scratch insurance, engine water damage insurance, and deductible waiver insurance.

[0050] As the number of cars on the road continues to increase, the number of people purchasing car insurance is also rising year by year. This not only brings more business opportunities to insurance companies, but also places higher demands on vehicle damage handling services.

[0051] Currently, traditional vehicle damage handling methods require strict human control at every stage. This is not only labor-intensive but also prone to human error, easily overlooking potential insurance fraud risks. Consequently, traditional vehicle damage handling methods are inefficient.

[0052] Based on this, embodiments of this application provide a vehicle damage processing method, apparatus, equipment, medium, and product based on a large model, which can be applied to scenarios such as fintech and artificial intelligence, aiming to improve the efficiency of existing vehicle damage processing methods.

[0053] The vehicle damage processing method, apparatus, equipment, medium, and product based on a large model provided in this application are specifically illustrated through the following embodiments. First, the vehicle damage processing method based on a large model in this application is described.

[0054] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0055] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0056] The vehicle damage processing method based on a large model provided in this application relates to the field of artificial intelligence technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the vehicle damage processing method based on a large model, but is not limited to the above forms.

[0057] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0058] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0059] Figure 1 This is an optional flowchart of the vehicle damage processing method based on a large model provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S104.

[0060] Step S101: Input the damaged image of the vehicle into the first model, identify the vehicle's identity information, damaged parts and the extent of damage through the first model, and determine the first damage value of the vehicle based on the identity information, the damaged parts and the extent of damage;

[0061] In step S101 of some embodiments, a picture of vehicle damage is input into a first model. The first model uses image recognition technology to analyze the vehicle features in the picture, identify the vehicle's identity information, damaged parts and the extent of damage, and then calculates the economic value of the initial damage to the vehicle, i.e., the first damage value, based on preset evaluation rules and algorithms and in combination with the vehicle's identity information, damaged parts and the extent of damage.

[0062] It should be noted that the identity information can include the vehicle's brand, model, and year, etc.

[0063] In some implementations, images of vehicle damage are acquired, and the images are then identified using a first model to obtain the vehicle's identity information, the damaged parts, and the extent of the damage. Based on the vehicle's identity information, the market value of the vehicle is obtained; based on the damaged parts and the extent of the damage, the repair cost of the vehicle is obtained; and finally, based on the vehicle's market value and repair cost, the first damage value of the vehicle is obtained.

[0064] In other implementations, the vehicle damage images are preprocessed, and then the preprocessed vehicle damage images are input into a first model. The first model uses image recognition technology to analyze the vehicle features in the images, identify the vehicle's identity information, damaged parts, and degree of damage. Then, the first model calculates the first damage value of the vehicle based on preset evaluation rules and algorithms, combined with the vehicle's identity information, damaged parts, and degree of damage.

[0065] Exemplarily, assume that a car has a collision accident, and the car owner takes pictures of the damaged front bumper of the vehicle and uploads them. The first model identifies that the license plate number of the vehicle is "Xiang A12345", the damaged part is the "front bumper", and the degree of damage is "slight deformation and paint scratches". According to the preset evaluation rules, the first model calculates the first damage value to be 500 yuan.

[0066] Step S102: Input the first information into the second model, and based on the first information, the second model identifies the claim fraud risk corresponding to the vehicle; wherein, the first information includes at least one of the damaged pictures and the vehicle owner's claim history record;

[0067] In step S102 of some embodiments, at least one of the vehicle damaged pictures and the vehicle owner's claim history record is input into the second model as the first information. The second model evaluates the claim fraud risk by analyzing the first information.

[0068] It should be noted that the claim fraud risk may be the possibility of fraud by the vehicle owner during the claim process, such as deliberately creating an accident, exaggerating losses, etc.

[0069] In some embodiments, by analyzing the metadata of the vehicle damaged pictures through the second model, it is identified whether there are signs of forgery or tampering, so as to evaluate the claim fraud risk. Among them, the metadata may be the shooting time, shooting location, etc. of the vehicle damaged pictures.

[0070] In some other embodiments, the second model cross - verifies the vehicle owner's claim history record to determine whether there is a fraud risk.

[0071] In some other embodiments, the second model cross - verifies the vehicle identity information and the vehicle owner's claim history record to determine whether there is a fraud risk.

[0072] Exemplarily, in the vehicle damaged pictures of the vehicle owner, it shows that the front bumper of the vehicle is slightly deformed, but the claim history record shows that the vehicle owner has claimed for similar accidents many times in the past year. The second model evaluates that there is a claim fraud risk based on this information.

[0073] Step S103: When it is determined that the vehicle has no fraud risk based on the claim fraud risk, input the first damage value into the third model, and based on the first damage value, the third model determines the claim settlement plan corresponding to the vehicle;

[0074] In step S103 of some embodiments, when the evaluation result of the second model shows that the vehicle has no fraud risk, input the first damage value into the third model, and the third model generates the claim settlement plan corresponding to the vehicle according to the first damage value.

[0075] It should be noted that the claims settlement plan may include information such as the amount of compensation, repair recommendations (such as original factory repair or non-original factory repair), repair location (such as designated repair shop), and parts replacement.

[0076] The first model, the second model, and the third model can be integrated into a single model or can be independently operating models.

[0077] In some implementations, if the second model's assessment results show that the vehicle does not pose a risk of insurance fraud, the first damage value is input into the third model. The third model determines the corresponding claim plan for the vehicle based on the first damage value, combined with industry standards and historical cases with similar damage conditions obtained from the claims database.

[0078] In other implementations, if the second model's evaluation results show that the vehicle does not pose a risk of insurance fraud, the first damage value is input into the third model. The third model determines the corresponding claim plan for the vehicle based on the first damage value and in conjunction with factors such as insurance terms.

[0079] For example, if the first damage value is 500 yuan and the second model assesses that there is no risk of insurance fraud in the vehicle, the third model generates a claim plan based on industry standards and historical cases: the compensation amount is 500 yuan, and it is recommended to have the vehicle repaired at the designated repair shop.

[0080] Step S104: In response to the first user's confirmation of the claim settlement plan, execute the claim settlement plan.

[0081] In step S104 of some embodiments, a claims settlement plan is presented to the vehicle owner, and a confirmation option is provided. After receiving confirmation from the vehicle owner regarding the claims settlement plan, the claims settlement plan is executed.

[0082] In some implementations, if a car owner disagrees with the claim settlement plan, they can apply for a manual review process, whereby the insurance company will manually review and adjust the claim settlement plan.

[0083] In other implementations, during the claims process, the repair progress can be tracked in real time, and repair status updates can be pushed to the vehicle owner.

[0084] For example, after the car owner receives the claim settlement plan, they confirm their agreement to its implementation. The insurance company will then automatically pay 500 yuan to the designated repair shop for repairs and notify the repair shop to begin vehicle repairs.

[0085] In the steps S101 to S104 illustrated in the embodiments of the present application, by inputting the damaged pictures of the vehicle into the first model, the identity information, damaged parts and damage degree of the vehicle are identified through the first model, and based on the identity information, damaged parts and damage degree, the first damage value of the vehicle is determined; the first information is input into the second model, and the second model identifies the claim fraud risk corresponding to the vehicle based on the first information; wherein, the first information includes at least one of the damaged pictures and the owner's claim history; in the case that it is determined based on the claim fraud risk that the vehicle does not have a fraud risk, the first damage value is input into the third model, and the third model determines the claim settlement plan corresponding to the vehicle based on the first damage value; in response to the confirmation information of the user for the claim settlement plan, the claim settlement plan is executed. The present application can determine the first damage value of the vehicle and identify the fraud risk through the model, and then when it is determined that the vehicle does not have a fraud risk, determine the claim settlement plan based on the first damage value for compensation, realizing automated vehicle damage processing, saving labor costs, reducing human errors, enabling a standardized process for vehicle damage processing, with a fast process speed, accurate vehicle damage processing results, and improving the efficiency of the vehicle damage processing method.

[0086] Please refer to Figure 2 , in some embodiments, after step S102 and before step S103, it may include but is not limited to steps S201 to S202:

[0087] Step S201, in the case that the claim fraud risk indicates that the vehicle has a fraud suspicion, output a first prompt message; wherein, the first prompt message is used to prompt the second user to determine the claim fraud risk corresponding to the vehicle;

[0088] In step S201 of some embodiments, when the second model identifies that the vehicle has a fraud suspicion, the mechanism for outputting the first prompt message is triggered, and the first prompt message is sent to the claim settlement reviewers or relevant staff of the insurance company.

[0089] Exemplarily, assume that the second model identifies that a vehicle with a license plate number of "Xiang A12345" has a fraud suspicion, then the first prompt message is sent to the claim settlement reviewers, with the content: "The vehicle with the license plate number of Xiang A12345 has a fraud suspicion, the claim frequency is abnormal and there are signs of forgery in the damaged pictures. Please further confirm the claim fraud risk of this vehicle."

[0090] Step S202, in response to the first instruction message indicating that the vehicle does not have a fraud risk received, determine that the vehicle does not have a fraud risk.

[0091] In step S202 of some embodiments, after receiving the first prompt message, the claims examiner or relevant staff member reviews the vehicle's risk of insurance fraud in claims settlement. After receiving the first indication message indicating that there is no risk of insurance fraud in the vehicle, the review result is recorded, and the status of the vehicle is updated to "no risk of insurance fraud".

[0092] Exemplarily, after receiving the first prompt message, the claims examiner conducts a detailed review of the vehicle with the license plate number "Xiang A12345". The claims examiner confirms that there is no risk of insurance fraud in this vehicle. Subsequently, the claims examiner sends the first indication message through the system interface. After the system receives it, the status of the vehicle is updated to "no risk of insurance fraud", and the subsequent claims settlement process continues.

[0093] In the embodiments of the present application, when a suspicion of insurance fraud is identified, the second user is promptly notified for manual intervention to ensure the accuracy and fairness of claims settlement review and prevent the occurrence of insurance fraud. Secondly, through the review and confirmation of the second user, it is finally determined whether there is a risk of insurance fraud in the vehicle, providing an accurate basis for the subsequent claims settlement process and ensuring the rationality and legality of claims settlement.

[0094] Please refer to Figure 3 , in some embodiments, after step S102, it may include but is not limited to steps S301 to S302:

[0095] Step S301, when it is determined based on the risk of insurance fraud in claims settlement that the vehicle has a risk of insurance fraud, output a second prompt message; wherein, the second prompt message is used to prompt the first user to upload unedited pictures of the vehicle damage;

[0096] In step S301 of some embodiments, the second model identifies the risk of insurance fraud in claims settlement of the vehicle based on the first information (such as pictures of vehicle damage and the owner's claims settlement history, etc.). When the second model determines that the vehicle has a risk of insurance fraud, a mechanism for outputting the second prompt message will be triggered. The second prompt message clearly requires the first user to upload unedited pictures of the vehicle damage.

[0097] Exemplarily, assume that the second model identifies a vehicle with the license plate number "Xiang A12345" as having a suspicion of insurance fraud, then a second prompt message is sent to the first user (the vehicle owner), with the content: "Dear vehicle owner, we have identified that there may be abnormalities in the pictures of vehicle damage you submitted. For further confirmation, please re-upload unedited pictures of the vehicle damage. Please ensure that the pictures are clear and complete, and are taken from multiple angles. The upload link is: [link address]".

[0098] Step S302: In response to the damaged vehicle pictures re - uploaded by the first user, return to execute the operation of inputting the damaged vehicle pictures into the first model. Identify the vehicle's identity information, damaged parts, and degree of damage through the first model, and based on the identity information, damaged parts, and degree of damage, determine the first damage value of the vehicle.

[0099] In step S302 of some embodiments, the first user re - uploads unedited damaged vehicle pictures according to the requirements of the second prompt information. After receiving the damaged vehicle pictures re - uploaded by the first user, input the re - uploaded damaged vehicle pictures into the first model. The first model analyzes the vehicle features in the pictures through image recognition technology to identify the vehicle's identity information, damaged parts, and degree of damage. The first model calculates the first damage value of the vehicle according to the preset evaluation rules and algorithms, in combination with the vehicle's identity information, damaged parts, and degree of damage.

[0100] In some implementation manners, before the first user re - uploads the pictures, a picture editing detection tool can be provided to help the user ensure that the uploaded pictures have not been edited.

[0101] In other implementation manners, if there are still problems with the pictures uploaded by the first user, the user can be prompted again to re - upload.

[0102] Exemplarily, after the vehicle owner receives the second prompt information, re - takes and uploads pictures of the damaged parts of the vehicle. After receiving the pictures, input the pictures into the first model. The first model identifies that the license plate number of the vehicle is "Xiang A12345", the damaged part is the "front bumper", and the degree of damage is "slightly deformed, paint scratched". According to the preset evaluation rules, the first model calculates the first damage value of the vehicle to be 500 yuan.

[0103] In the embodiments of this application, when insurance fraud suspicion is identified, the first user is required to provide real and unedited damaged vehicle pictures to re - evaluate the damage situation of the vehicle, ensuring the fairness and accuracy of the claim settlement process. By re - uploading unedited damaged vehicle pictures, re - evaluate the damage situation of the vehicle to ensure the accuracy and reliability of the evaluation result of the vehicle damage value, thus providing accurate data support for subsequent claim settlement plans.

[0104] Please refer to Figure 4 , in some embodiments, step S103 may include but is not limited to steps S401 to S402:

[0105] Step S401: Input the first damage value into the third model, and through the third model, query historical vehicle damage handling cases and industry standards based on the first damage value to determine the second damage value of the vehicle;

[0106] In step S401 of some embodiments, the first damage value is input into the third model. The third model uses an internal algorithm to query historical vehicle damage handling cases similar to the current vehicle damage situation, and combines industry standards to correct and adjust the first damage value. The third model outputs the adjusted second damage value.

[0107] It should be noted that historical vehicle damage cases can be similar vehicle damage cases that have been handled in the past, including information such as the damage details, repair costs, and claim amounts.

[0108] Industry standards can be the repair cost standards and claims standards that are generally accepted in the auto repair industry or the insurance industry.

[0109] The second damage value can be the final economic value of the vehicle damage, or it can be the economic value of the vehicle damage reassessed using a third model based on the first damage value, combined with historical vehicle damage handling cases and industry standards.

[0110] For example, suppose the value of the first damage is 500 yuan, the damaged part is the front bumper, and the damage is minor deformation. The third model queries historical vehicle damage cases and finds that the average repair cost for similar damage is 450 yuan. Meanwhile, combining industry standards, the repair cost for minor deformation of the front bumper ranges from 400 to 500 yuan. After considering the above information, the third model determines the value of the second damage to be 480 yuan.

[0111] Step S402: Based on the second damage value, determine the corresponding compensation plan for the vehicle.

[0112] In step S402 of some embodiments, the third model generates a claim plan corresponding to the vehicle based on the second damage value.

[0113] In some implementations, multiple claims options can be generated for users to choose from, such as cash compensation options, repair service options, and options combining cash compensation and repair services.

[0114] In other implementations, the first user is allowed to provide feedback or raise objections to the claims settlement plan, adjustments are made based on user feedback, and corresponding explanations are provided.

[0115] For example, based on the second damage value of 480 yuan, the third model generates the following claim plan: compensation amount of 480 yuan, and a recommendation to have the vehicle repaired at a designated repair shop with the address "XX Province XX District XX No." This claim plan is output and provided to the vehicle owner. After the vehicle owner confirms it, the system automatically pays the repair shop 480 yuan for repairs and notifies the repair shop to begin vehicle repairs.

[0116] In this embodiment, a third model is used to query historical vehicle damage handling cases and industry standards to correct and adjust the first damage value, thereby determining a second damage value. This helps improve the accuracy of damage value assessment. Based on the second damage value, a reasonable claims settlement plan is then generated, ensuring transparency in the claims process while improving claims efficiency and customer satisfaction.

[0117] Please see Figure 5 In some embodiments, step S402 may also include, but is not limited to, steps S501 to S503:

[0118] Step S501: In the case where the second damage value indicates that the vehicle is a total loss and cannot be repaired, the compensation plan for the vehicle is determined to be cash compensation.

[0119] It should be noted that total loss can be defined as damage to a vehicle that reaches or exceeds its market value, making it impossible to restore it to its pre-accident condition through repair.

[0120] Unrepairable could be due to severe damage to the vehicle that is technically impossible to repair, or excessively high repair costs.

[0121] Step S502: If the second damage value represents partial loss of the vehicle and the repair cost is lower than the market value, determine that the corresponding claim plan for the vehicle is repair service;

[0122] It should be noted that partial loss may refer to situations where the vehicle damage is not total and can be repaired to restore it to its pre-accident condition.

[0123] Step S503: If the second damage value indicates that a component of the vehicle is damaged, determine that the corresponding claim plan for the vehicle is a component replacement service.

[0124] It should be noted that parts replacement can be a service provided by the insurance company to replace damaged parts. Car owners can choose to have the parts replaced at a designated repair shop, and the insurance company will pay for the replacement.

[0125] This application embodiment provides reasonable compensation plans based on different vehicle damage situations, ensuring the fairness and transparency of the compensation process, while improving compensation efficiency and customer satisfaction.

[0126] Please see Figure 6 In some embodiments, after step S104, steps S601 to S603 are included, but are not limited to:

[0127] Step S601: Obtain the claim settlement plan and synchronize the claim settlement plan to the big data system;

[0128] In step S601 of some embodiments, after the first user confirms the claim settlement plan, the relevant data of the claim settlement plan is obtained. The relevant data of the claim settlement plan (such as vehicle information, damage condition, compensation amount, repair method, etc.) is synchronized to the big data system.

[0129] Step S602: Generate a report based on the big data system and send the report to the second user via email or subscription account;

[0130] In step S602 of some embodiments, a statistical report is generated based on the claims settlement data stored in the big data system. The report is then sent to a second user via email or a subscription account (such as a WeChat official account or WeChat Work).

[0131] Step S603: Input the feedback information of the first user and the report into the fourth model, and obtain the optimized fourth model based on the report and the feedback information; wherein the fourth model is used to optimize the subsequent claims settlement plan.

[0132] In step S603 of some embodiments, feedback information from the first user regarding the claims settlement plan is collected, and then this feedback information, along with the report generated by the big data system, is input into the fourth model. The fourth model, through data analysis and machine learning algorithms, combines the report data and feedback information to adjust model parameters and optimize the claims settlement plan generation logic. The optimized fourth model is then output for subsequent claims settlement plan generation.

[0133] It should be noted that the fourth model can be the third model or other models. The fourth model can be integrated into a single model with the first model, the second model, etc., or it can be a model that runs independently.

[0134] For example, suppose the car owner's feedback on the current claim settlement is that the compensation amount is low and the repair service quality is average. This feedback, along with report data, is then input into the fourth model. The fourth model's analysis reveals that similar cases generally have low compensation amounts and poor repair service quality ratings. After adjusting the model's parameters, the claim settlement generation logic is optimized, improving the reasonableness of the compensation amount and recommending higher-quality repair shops. The optimized fourth model generates more reasonable claim settlement plans in subsequent claims, increasing car owner satisfaction.

[0135] In this embodiment, claims settlement data is synchronized to a big data system, and reports are generated based on the big data system and sent to a second user. This helps managers monitor the claims settlement process and identify potential problems. The fourth model is optimized through user feedback and data analysis to improve the accuracy of the claims settlement plan and user satisfaction, thereby enhancing the efficiency and quality of the claims settlement process.

[0136] Please see Figure 7This application also provides a schematic diagram of a vehicle damage processing method based on a large model. Its implementation includes: using AI image recognition technology to perform AI recognition on images uploaded by the user (first user), including but not limited to: AI recognition of vehicle brand, vehicle model, vehicle year, damaged parts, and damage extent. Based on the AI ​​image recognition results, combined with existing claims processes and compared with claims plans from previous years, AI damage assessment is performed, and an online quote is provided. If the user agrees to the quote, online claims can be processed directly, reducing many complex and cumbersome processes and greatly improving claims efficiency.

[0137] The specific implementation steps are as follows:

[0138] 1. Write specific prompts and embed an image recognition plugin to enable the large model to recognize images. Provide specific examples to help the large model identify risk claims scenarios and continuously optimize the prompts.

[0139] 2. Apply for an API interface. Users upload pictures of vehicle damage. The backend service calls the API interface to transmit the image information to the API in real time (historical data can be transmitted to the API asynchronously offline). A large-scale AI model is used for image recognition to analyze the user-uploaded photos. In other words, the system calls the AI ​​image recognition interface to transmit images in real time.

[0140] 3. AI identifies vehicle information, including brand, model, and year.

[0141] 4. AI identifies the damaged parts and, combined with the vehicle information from the previous step, makes a rough estimate of the damaged parts to obtain the first damage value.

[0142] 5. AI-powered risk control identifies insurance fraud to prevent losses to the company. The large-scale model returns a fraud assessment result, and the backend service processes the results based on the AI ​​identification to provide a final quote and claims settlement plan.

[0143] If the insurance fraud assessment confirms that insurance fraud has occurred, determine whether it is necessary to re-upload the images. If yes, require the first user to re-upload images of the vehicle damage; otherwise, apply for the manual processing channel.

[0144] If the insurance fraud assessment results indicate suspicion of insurance fraud, a second user will manually identify whether it is insurance fraud. If it is, no compensation will be paid; if not, AI will assess the damage, i.e., the second damage value, based on the claims settlement plan of previous years, i.e., historical vehicle damage handling cases. Then, the system will quote a claim based on the AI ​​damage assessment and the insured amount.

[0145] If the insurance fraud assessment results indicate that there is no insurance fraud, the AI ​​will determine the loss based on the claims settlement plan of previous years. Then, the system will quote a claim based on the AI's loss assessment and the insured amount.

[0146] 6. The user (first user) confirms the quotation information (claims plan). If the user does not agree to confirm the quotation information, the user applies for a manual channel, where the salesperson, i.e., the second user, renegotiates the quotation based on the vehicle information and the damaged parts, and then confirms the quotation information with the first user again.

[0147] Once the quote information is confirmed and agreed upon, the system automatically processes the claim payment. The backend service asynchronously records the claim plan and generates reports in the big data system for easy analysis. The system administrator (the second user) is notified via email and the Happy Peace subscription account. The claim plan will be continuously optimized in the future.

[0148] Please see Figure 8 This application also provides a vehicle damage processing device based on a large model, which can implement the above-mentioned vehicle damage processing method based on a large model. The device includes:

[0149] The first determining module 801 is used to input the damaged image of the vehicle into the first model, identify the vehicle's identity information, damaged parts and the extent of damage through the first model, and determine the first damage value of the vehicle based on the identity information, the damaged parts and the extent of damage;

[0150] The identification module 802 is used to input the first information into the second model, and the second model identifies the insurance fraud risk corresponding to the vehicle based on the first information; wherein, the first information includes at least one of the damage pictures and the owner's claim history;

[0151] The second determining module 803 is used to input the first damage value into the third model when it is determined that the vehicle does not have the risk of insurance fraud based on the risk of insurance fraud, and to determine the corresponding claim plan for the vehicle based on the first damage value through the third model.

[0152] The execution module 804 is used to execute the claim settlement plan in response to the user's confirmation information.

[0153] The specific implementation of the large-model-based vehicle damage processing device is basically the same as the specific implementation of the large-model-based vehicle damage processing method described above, and will not be repeated here.

[0154] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned vehicle damage processing method based on a large model. This electronic device can be any smart terminal, including a tablet computer or an in-vehicle computer.

[0155] Please see Figure 9 , Figure 9The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0156] The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0157] The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and called and executed by the processor 901 to implement the vehicle damage processing method based on a large model according to the embodiments of this application.

[0158] The input / output interface 903 is used to implement information input and output;

[0159] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0160] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0161] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0162] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described vehicle damage processing method based on a large model.

[0163] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0164] The vehicle damage processing method, apparatus, equipment, medium, and product based on a large model provided in this application embodiment involves inputting images of vehicle damage into a first model, which identifies the vehicle's identity information, damaged parts, and extent of damage, and determines a first damage value based on these factors. This first information is then input into a second model, which identifies the corresponding insurance fraud risk based on the first information. The first information includes at least one of the damaged images and the vehicle owner's claim history. If the insurance fraud risk is determined to be absent, the first damage value is input into a third model, which determines the corresponding claim plan based on this value. The claim plan is then executed in response to user confirmation. This application can determine the first damage value of a vehicle and identify insurance fraud risk through a model. Furthermore, when the insurance fraud risk is determined to be absent, a claim plan is determined based on the first damage value for compensation. This achieves automated vehicle damage processing, saves labor costs, reduces human error, enables a standardized process for vehicle damage processing, and provides fast, accurate, and efficient results, thus improving the overall efficiency of vehicle damage processing methods.

[0165] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0166] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0167] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0168] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0169] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0170] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0171] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0172] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0173] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0174] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0175] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A vehicle damage processing method based on a large model, characterized in that, The method includes: The damaged vehicle image is input into the first model, which identifies the vehicle's identity information, damaged parts, and extent of damage. Based on the identity information, the damaged parts, and the extent of damage, the first damage value of the vehicle is determined. The first information is input into the second model, and the second model identifies the insurance fraud risk corresponding to the vehicle based on the first information; wherein, the first information includes at least one of the damage images and the owner's claim history. If it is determined that the vehicle does not have the risk of insurance fraud based on the aforementioned insurance fraud risk, the first damage value is input into the third model, and the third model determines the corresponding claim plan for the vehicle based on the first damage value. In response to the first user's confirmation of the claim settlement plan, the claim settlement plan is executed.

2. The method according to claim 1, characterized in that, After inputting the first information into the second model and identifying the insurance fraud risk corresponding to the vehicle based on the first information, and before determining that the vehicle does not have an insurance fraud risk based on the insurance fraud risk, the method further includes: When the insurance fraud risk indicator suggests that the vehicle is suspected of insurance fraud, a first prompt message is output; wherein, the first prompt message is used to prompt the second user to confirm the insurance fraud risk corresponding to the vehicle; In response to a first indication message received indicating that the vehicle is not at risk of insurance fraud, it is determined that the vehicle is not at risk of insurance fraud.

3. The method according to claim 1, characterized in that, After inputting the first information into the second model, and using the second model to identify the insurance fraud risk corresponding to the vehicle based on the first information, the method further includes: If it is determined that the vehicle is at risk of insurance fraud based on the aforementioned insurance fraud risk, a second prompt message is output; wherein, the second prompt message is used to prompt the first user to upload unedited pictures of the vehicle damage; In response to the first user re-uploading images of the vehicle's damage, the process returns to inputting the images of the vehicle's damage into the first model. The first model identifies the vehicle's identity information, damaged parts, and degree of damage. Based on the identity information, the damaged parts, and the degree of damage, the first damage value of the vehicle is determined.

4. The method according to claim 1, characterized in that, The step of inputting the first damage value into the third model and determining the corresponding compensation plan for the vehicle based on the first damage value through the third model includes: The first damage value is input into the third model, and the third model determines the second damage value of the vehicle based on the first damage value by querying historical vehicle damage handling cases and industry standards. Based on the second damage value, the corresponding compensation plan for the vehicle is determined.

5. The method according to claim 4, characterized in that, The determination of the corresponding compensation plan for the vehicle based on the second damage value includes: In cases where the second damage value indicates that the vehicle is a total loss and cannot be repaired, the corresponding compensation plan for the vehicle is determined to be cash compensation. If the second damage value represents partial loss of the vehicle and the repair cost is lower than the market value, the corresponding claim plan for the vehicle is determined to be repair service; In cases where the second damage value represents damage to a component of the vehicle, the corresponding claim solution for the vehicle is determined to be a component replacement service.

6. The method according to claim 1, characterized in that, After responding to the first user's confirmation of the claim settlement plan and executing the claim settlement plan, the method further includes: Obtain the claims settlement plan and synchronize it to the big data system; Reports are generated based on the big data system and sent to a second user via email or subscription account. The feedback information from the first user and the report are input into the fourth model. Based on the report and the feedback information, the fourth model is optimized to obtain an optimized fourth model. The fourth model is used to optimize the subsequent claims settlement plan.

7. A vehicle damage processing device based on a large model, characterized in that, The device includes: The first determining module is used to input images of vehicle damage into a first model, identify the vehicle's identity information, damaged parts, and degree of damage through the first model, and determine the first damage value of the vehicle based on the identity information, the damaged parts, and the degree of damage. The identification module is used to input first information into a second model, and the second model identifies the insurance fraud risk corresponding to the vehicle based on the first information; wherein, the first information includes at least one of the damage images and the owner's claim history. The second determining module is used to input the first damage value into the third model when it is determined that the vehicle does not have the risk of insurance fraud based on the risk of insurance fraud, and to determine the corresponding claim plan for the vehicle based on the first damage value through the third model. The execution module is used to execute the claim settlement plan in response to the user's confirmation information.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the vehicle damage processing method based on a large model as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the vehicle damage processing method based on a large model as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device performs the vehicle damage processing method based on a large model as described in any one of claims 1 to 6.

Images