Object screening method, device and equipment and storage medium thereof

Abstract

The embodiment of the invention belongs to the technical field of artificial intelligence, is applied to a vehicle maintenance recommendation scene, and relates to an object screening method, device and equipment and a storage medium thereof. Inputting the to-be-recognized image into the first prediction model to obtain a first list; extracting information of vehicle accessories needing to be replaced or / and maintained in the first list; inputting the vehicle accessory information and the priority screening condition into a second prediction model to obtain a second list; and extracting the comprehensive information of the maintenance manufacturer from the second list, and pushing the comprehensive information to the user side. According to the object screening method provided by the invention, the information of the vehicle accessories needing to be replaced or / and maintained during vehicle maintenance and the maintenance manufacturers meeting the priority screening conditions are sequentially predicted in combination with previous and later predictions, so that a user side can conveniently select the maintenance manufacturers, and more excellent and desirable vehicle maintenance recommendation services can be provided for a vehicle owner.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and is applied to vehicle maintenance recommendation scenarios, and in particular to an object screening method, device, equipment and storage medium thereof. Background Art

[0002] Cars are one of the most widely used means of transportation in daily life, so vehicle maintenance has become one of the problems to be solved in daily life. When vehicle maintenance services are needed, how to quickly find a satisfactory vehicle maintenance service provider becomes the current problem to be solved. In the field of automobile maintenance, the screening of car repair parts and repair manufacturers is a key link in improving the efficiency and accuracy of damage assessment. At present, although there are some related systems and methods, there are still many defects. Existing screening schemes often rely too much on simple keyword matching and fixed sorting rules, and cannot fully consider the complexity of vehicle damage, the diversity of accessories, and the variability of maintenance conditions.

[0003] At present, there is a widespread problem of low efficiency in the connection between vehicle maintenance service providers and vehicle owners, and the effective allocation of vehicle maintenance service provider resources has not been achieved. When car owners look for vehicle maintenance service providers, they often rely on their original experience, introductions from others, etc., and it is difficult to find a satisfactory vehicle maintenance service provider. The efficiency of resource connection between car owners and vehicle maintenance service providers is low. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose an object screening method, device, equipment and storage medium thereof to solve the problem that the resource docking efficiency between existing car owners and vehicle maintenance service providers is low and it is difficult to find a satisfactory vehicle maintenance service provider.

[0005] In order to solve the above technical problems, the present application embodiment provides an object screening method, which adopts the following technical solution:

[0006] An object screening method comprises the following steps:

[0007] Obtain the image to be identified and the priority screening conditions uploaded by the user;

[0008] Inputting the image to be recognized into a pre-trained first prediction model to obtain a first list output by the first prediction model;

[0009] Extracting text content from the first list to extract information about vehicle parts that need to be replaced or / and repaired in the first list;

[0010] Inputting the vehicle accessory information and the priority screening condition into a pre-trained second prediction model to obtain a second list output by the second prediction model;

[0011] The comprehensive information of the target repair manufacturer is extracted from the second list, and the comprehensive information of the target repair manufacturer is pushed to the user end, wherein the comprehensive information includes repair technology level information, repair service attitude information, total repair price information, repair time information, repair shop address information, and repair shop name information.

[0012] Furthermore, the step of obtaining the image to be identified and the priority screening conditions uploaded by the user terminal specifically includes:

[0013] Acquire the image to be identified and the priority screening conditions uploaded by the user according to the preset prompt content, wherein the preset prompt content includes an image clarity compliance field, an image import position indication field, and screening condition information for prompting the user to check, the image to be identified includes an image of a damaged or destroyed part of a vehicle, and the screening condition information includes maintenance technology level information, maintenance service attitude information, total maintenance price information, maintenance time information, and maintenance shop address information.

[0014] Furthermore, the first prediction model includes a vehicle parts prediction model, the first list includes information about vehicle parts that need to be replaced or / and repaired, and the vehicle parts information includes parts names and parts models. Before executing the step of inputting the image to be recognized into the pre-trained first prediction model to obtain the first list output by the first prediction model, the method further includes:

[0015] Acquire a batch of vehicle accident images collected in advance and a list of vehicle repair parts corresponding to the batch of vehicle accident images, and construct a training data set with a one-to-one mapping between the images and the list;

[0016] Inputting the training data set into a vehicle parts prediction model to be trained, performing learning training on the vehicle parts prediction model, and obtaining a pre-trained vehicle parts prediction model, wherein the vehicle parts prediction model includes a multimodal graph-text relationship prediction model based on LLM;

[0017] The step of inputting the image to be recognized into a pre-trained first prediction model to obtain a first list output by the first prediction model specifically includes:

[0018] Inputting the image of the damaged or destroyed part of the vehicle into a pre-trained vehicle parts prediction model to predict the name and model of the parts that need to be replaced or / and repaired by the vehicle parts prediction model;

[0019] The names and models of the parts that need to be replaced and / or repaired are output in the form of a list.

[0020] Furthermore, the step of extracting text content from the first list to extract information about vehicle parts that need to be replaced or / and repaired in the first list specifically includes:

[0021] The natural language content recognition component in the LLM-based multimodal image-text relationship prediction model is used to extract text content from the first list, and the names and models of the parts that need to be replaced or / and repaired in the first list are extracted as the vehicle parts information.

[0022] or,

[0023] Using natural language recognition technology based on NLP, the names and models of the parts that need to be replaced or / and repaired are extracted from the first list as the vehicle parts information,

[0024] or,

[0025] The OCR optical character extraction technology is used to identify and extract the names and models of the parts that need to be replaced and / or repaired from the first list as the vehicle parts information.

[0026] Furthermore, the second prediction model includes a maintenance object screening prediction model, and the maintenance object refers to a maintenance manufacturer that repairs the accident vehicle. Before executing the step of inputting the vehicle parts information and the priority screening condition into the pre-trained second prediction model to obtain the second list output by the second prediction model, the method further includes:

[0027] Extracting identification information of the repair manufacturers one by one from the vehicle repair parts lists corresponding to the batch of vehicle accident images, wherein the identification information includes name information of the repair manufacturers;

[0028] Using the identification information as an index field, multi-dimensional feature information corresponding to all maintenance vendors is obtained from a plurality of preset data sources, wherein the multi-dimensional feature information includes maintenance technology level information, maintenance service attitude information, maintenance total price information, maintenance time consumption information, and maintenance factory address information;

[0029] The multi-dimensional feature information is input as multi-dimensional training data into the maintenance object screening prediction model to be trained, and the maintenance object screening prediction model to be trained is trained to learn the maintenance technology level information, maintenance service attitude information of all maintenance manufacturers, and maintenance price information and maintenance time information of all maintenance manufacturers when repairing different vehicle parts, so as to obtain a pre-trained maintenance object screening prediction model.

[0030] Furthermore, the step of training the maintenance object screening prediction model to be trained to learn the maintenance technology level information, maintenance service attitude information of all maintenance manufacturers, and maintenance price information and maintenance time consumption information of all maintenance manufacturers when repairing different vehicle parts specifically includes:

[0031] Analyzing the multi-dimensional feature information corresponding to all maintenance manufacturers respectively obtained from the preset multiple data sources;

[0032] Obtaining preset confidence weights of the plurality of preset data source terminals;

[0033] Based on the confidence weights of the multiple data sources and the multi-dimensional feature information, a weighted average method is adopted for each dimensional feature to obtain the maintenance technology level score, maintenance service attitude score, maintenance price score, maintenance time consumption score and maintenance factory distance score corresponding to all maintenance manufacturers;

[0034] The maintenance technology level scores, maintenance service attitude scores, maintenance price scores, maintenance time scores and maintenance factory distance scores corresponding to all maintenance manufacturers are used as reference knowledge and deployed into the maintenance object screening prediction model; and,

[0035] According to the maintenance technology level score, maintenance service attitude score, maintenance price score, maintenance time score and maintenance plant distance score, score levels of different dimensions are divided respectively;

[0036] The score levels of the different dimensions are set as checkable options in the priority screening conditions and sent to the user end interface, so that the user can select the maintenance screening conditions when uploading the image to be identified.

[0037] Furthermore, the step of inputting the vehicle accessory information and the priority screening condition into the pre-trained second prediction model to obtain the second list output by the second prediction model specifically includes:

[0038] Inputting the vehicle parts information and the priority screening conditions into the pre-trained maintenance object screening prediction model;

[0039] By parsing and identifying the vehicle accessory information, all vehicle accessory names and accessory models contained in the vehicle accessory information are determined;

[0040] By analyzing and identifying the priority screening conditions, determining the priority screening conditions of the maintenance vendors selected by the user through the user terminal;

[0041] According to the maintenance manufacturer screening priority condition, the maintenance manufacturers in the pre-trained maintenance object screening prediction model are initially screened to obtain an initial screening result;

[0042] Generate a repair quotation result and comprehensive information of the target repair manufacturer based on the repair price information and repair time information of all repair manufacturers when repairing different vehicle parts in the initial screening results, and all vehicle parts names and parts models contained in the vehicle parts information;

[0043] The maintenance quotation result and the comprehensive information of the target maintenance manufacturer are output in a list form from the output position of the pre-trained maintenance object screening prediction model to obtain the second list.

[0044] In order to solve the above technical problems, the embodiment of the present application also provides an object screening device, which adopts the following technical solution:

[0045] An object screening device, comprising:

[0046] The user-end uploaded information acquisition module is used to obtain the image to be identified and the priority screening conditions uploaded by the user end;

[0047] A first prediction module, used for inputting the image to be recognized into a pre-trained first prediction model to obtain a first list output by the first prediction model;

[0048] A vehicle parts information extraction module, used to extract text content from the first list to extract information about vehicle parts that need to be replaced or / and repaired in the first list;

[0049] A second prediction module, used for inputting the vehicle parts information and the priority screening condition into a pre-trained second prediction model to obtain a second list output by the second prediction model;

[0050] The push module to the user terminal is used to extract the comprehensive information of the target repair manufacturer from the second list and push the comprehensive information of the target repair manufacturer to the user terminal, wherein the comprehensive information includes maintenance technology level information, maintenance service attitude information, total maintenance price information, maintenance time information, maintenance plant address information, and maintenance plant name information.

[0051] 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:

[0052] A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the object screening method described above when executing the computer-readable instructions.

[0053] 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:

[0054] A computer-readable storage medium stores computer-readable instructions, which implement the steps of the object screening method described above when executed by a processor.

[0055] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0056] The object screening method described in the embodiment of the present application obtains the image to be identified and the priority screening conditions uploaded by the user end; inputs the image to be identified into the first prediction model to obtain the first list; extracts the vehicle parts information that needs to be replaced or / and repaired in the first list; inputs the vehicle parts information and the priority screening conditions into the second prediction model to obtain the second list; extracts the comprehensive information of the repair manufacturer from the second list, and pushes the comprehensive information of the repair manufacturer to the user end, wherein the comprehensive information includes the repair technology level information, the repair service attitude information, the total repair price information, the repair time information, the repair shop address information, and the repair shop name information. The object screening method described in the present application combines the two previous and subsequent predictions to sequentially predict the vehicle parts information that needs to be replaced or / and repaired during vehicle maintenance, and predicts the repair shops that meet the priority screening conditions, so as to facilitate the user end to select the repair shop, be more intelligent and automated, and provide the car owner with a better quality and more satisfactory vehicle maintenance recommendation service. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] 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.

[0058] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0059] Figure 2 is a flow chart of an embodiment of the object screening method according to the present application;

[0060] Figure 3 is a flowchart of a specific embodiment of pre-training the first prediction model in the object screening method described in the present application;

[0061] Figure 4 yes Figure 2 A flowchart of a specific embodiment of step 202 is shown;

[0062] Figure 5 is a flowchart of a specific embodiment of pre-training the second prediction model in the object screening method described in the present application;

[0063] Figure 6 yes Figure 5 A flowchart of a specific embodiment of step 503 is shown;

[0064] Figure 7 yes Figure 2 A flowchart of a specific embodiment of step 204 is shown;

[0065] Figure 8 is a schematic structural diagram of an embodiment of an object screening device according to the present application;

[0066] Fig. 9 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0067] 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.

[0068] 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.

[0069] 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.

[0070] like Figure 1As 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.

[0071] 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.

[0072] 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 and a desktop computer, etc.

[0073] The server 103 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal device 101 .

[0074] It should be noted that the object screening method provided in the embodiment of the present application is generally executed by a terminal device, and accordingly, the object screening device is generally arranged in the terminal device.

[0075] 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.

[0076] Continue to refer Figure 2 , shows a flow chart of an embodiment of an object screening method according to the present application. The object screening method comprises the following steps:

[0077] Step 201, obtaining the image to be identified and the priority screening conditions uploaded by the user;

[0078] Specifically, the method obtains an image to be identified and a priority screening condition uploaded by a user terminal, wherein the image to be identified is an image of a damaged or destroyed part of a vehicle, and the priority screening condition refers to a repair priority condition that can be checked by a user. The object screening method described in the present application is applied to a vehicle repair recommendation scenario, and can combine the image of the damaged or destroyed part of the vehicle and the repair priority condition checked by the user to recommend a suitable repair vendor to the user terminal, thereby providing the user with a higher quality and more satisfactory vehicle repair recommendation service in a more intelligent manner.

[0079] Step 202, inputting the image to be recognized into a pre-trained first prediction model to obtain a first list output by the first prediction model;

[0080] Specifically, the first prediction model is a vehicle parts prediction model, which can predict the names and models of parts that need to be replaced and / or repaired based on the input images of damaged or destroyed parts of the vehicle, and output the names and models of parts that need to be replaced and / or repaired in the form of a first list.

[0081] In this embodiment, a pre-trained vehicle parts prediction model is used to predict the image of the damaged or destroyed part of the vehicle input by the user, and the name and model of the parts that need to be replaced and / or repaired are predicted, and the name and model of the parts that need to be replaced and / or repaired are output in the form of a first list to facilitate the subsequent screening of repair manufacturers based on the name and model of the parts that need to be replaced and / or repaired.

[0082] Step 203, extracting text content from the first list to extract information about vehicle parts that need to be replaced or / and repaired in the first list;

[0083] Specifically, the names and models of the parts that need to be replaced or / and repaired in the first list are extracted, so that the subsequent second prediction model can screen the repair manufacturers in combination with the names and models of the parts that need to be replaced or / and repaired;

[0084] In addition, after extracting the text content of the first list and extracting the information of the vehicle parts that need to be replaced or / and repaired in the first list, the method further includes: structuring the names and models of the parts that need to be replaced or / and repaired to generate structured data. The purpose of structuring and generating structured data is to facilitate the rapid recognition and content extraction of the computer processor.

[0085] Step 204, inputting the vehicle accessory information and the priority screening condition into the pre-trained second prediction model to obtain a second list output by the second prediction model;

[0086] Specifically, the second prediction model includes a maintenance object screening prediction model, where the maintenance object refers to the maintenance manufacturer that repairs the accident vehicle. By inputting the vehicle parts information and the priority screening conditions into the pre-trained maintenance object screening prediction model, it is possible to provide car owners with better quality and more satisfactory vehicle maintenance recommendation services and screen out more reasonable maintenance manufacturers.

[0087] Step 205, extracting comprehensive information of the target repair manufacturer from the second list, and pushing the comprehensive information of the target repair manufacturer to the user terminal, wherein the comprehensive information includes repair technology level information, repair service attitude information, total repair price information, repair time information, repair shop address information, and repair shop name information.

[0088] Specifically, by extracting the comprehensive information of the target repair manufacturer from the second list, and pushing the comprehensive information of the target repair manufacturer to the user terminal, the user can fully understand the repair technology level, repair service attitude, total repair price, repair time, repair shop address and name information of the target repair manufacturer.

[0089] In this embodiment, the image to be identified and the priority screening conditions uploaded by the user end are obtained; the image to be identified is input into the first prediction model to obtain the first list; the vehicle parts information that needs to be replaced or / and repaired in the first list is extracted; the vehicle parts information and the priority screening conditions are input into the second prediction model to obtain the second list; the comprehensive information of the repair manufacturer is extracted from the second list, and the comprehensive information of the repair manufacturer is pushed to the user end, wherein the comprehensive information includes the repair technology level information, the repair service attitude information, the total repair price information, the repair time information, the repair shop address information, and the repair shop name information. The object screening method described in the present application combines the two previous and subsequent predictions to sequentially predict the vehicle parts information that needs to be replaced or / and repaired during vehicle maintenance, and predicts the repair shops that meet the priority screening conditions, so as to facilitate the user end to select the repair shop, be more intelligent and automated, and provide the car owner with a better quality and more satisfactory vehicle maintenance recommendation service.

[0090] In some optional implementations, step 201 is specifically implemented as follows: obtaining the image to be identified and the priority filtering conditions uploaded by the user terminal according to the preset prompt content, wherein the preset prompt content includes an image clarity compliance field, an image import position indication field, and filtering condition information for prompting the user to check the box, the image to be identified includes an image of a damaged or destroyed part of the vehicle, and the filtering condition information includes maintenance technology level information, maintenance service attitude information, total maintenance price information, maintenance time information, and maintenance shop address information.

[0091] By combining the preset prompt content, the images to be identified uploaded by the user meet the upload identification requirements, and can be combined with the filtering condition information checked by the user to screen out more suitable repair manufacturers and provide better quality and more satisfactory vehicle repair recommendation services.

[0092] Continue to refer Figure 3 In some optional implementations, before executing step 202, a step of pre-training the first prediction model is also included. Figure 3 is a flowchart of a specific embodiment of pre-training the first prediction model in the object screening method described in the present application, comprising the following steps:

[0093] Step 301, obtaining a batch of vehicle accident images collected in advance and a list of vehicle repair parts corresponding to the batch of vehicle accident images, and constructing a training data set with a one-to-one mapping between the images and the list;

[0094] Specifically, the main purpose of constructing a training data set that maps one-to-one between images and lists is to construct the vehicle parts information that needs to be replaced or / and repaired contained in each vehicle accident image according to the vehicle repair parts list, such as: part name and part model, to facilitate subsequent vehicle parts prediction model learning and training.

[0095] Step 302: input the training data set into the vehicle parts prediction model to be trained, perform learning training on the vehicle parts prediction model, and obtain a pre-trained vehicle parts prediction model, wherein the vehicle parts prediction model includes a multimodal graph-text relationship prediction model based on LLM.

[0096] Specifically, in the selection of the vehicle parts prediction model, a multimodal image-text relationship prediction model based on LLM (Large Language Model, LLM for short) can be selected. The multimodal image-text relationship prediction model based on LLM can not only understand and identify the text content in the training data set, such as: vehicle parts information that need to be replaced or / and repaired, such as: parts name and part model, but also can fully learn the image knowledge of each vehicle accident image, so as to combine text knowledge and image knowledge to train the vehicle parts prediction model;

[0097] In the learning and training stage of the first prediction model, the input to the model is a training data set, that is, a batch of vehicle accident images and the vehicle parts information that needs to be replaced or / and repaired contained in each vehicle accident image, and the output is actually the mapping relationship between different image features and the vehicle parts information that needs to be replaced or / and repaired, for example: the mapping relationship between the collision depth, collision damage location, collision damage area size and the vehicle parts information that needs to be replaced or / and repaired;

[0098] When the pre-trained vehicle parts prediction model is actually used, it is possible to predict the vehicle parts information that need to be replaced or / and repaired based on the image to be identified uploaded by the user. That is, it can be understood that in actual use, the input to the vehicle parts prediction model is the image of the damaged or destroyed part of the vehicle. Different image features are obtained based on the image of the damaged or destroyed part of the vehicle, such as the collision depth, collision damage location, and collision damage area size. The output is the vehicle parts information that need to be replaced or / and repaired in combination with the image features.

[0099] Continue to refer Figure 4 , in some optional implementations, Figure 4 yes Figure 2 The flowchart of a specific embodiment of step 202 shown includes the following steps:

[0100] Step 401, inputting the image of the damaged or destroyed part of the vehicle into a pre-trained vehicle parts prediction model, and predicting the name and model of the parts that need to be replaced or / and repaired by the vehicle parts prediction model;

[0101] Specifically, it can be understood that the image of the damaged or destroyed part of the vehicle is input into a pre-trained vehicle parts prediction model, and the vehicle parts prediction model is used to extract image features of the image of the damaged or destroyed part of the vehicle, and identify the collision depth, collision damage location, collision damage area size and other information contained in the image, and determine the name and model of the parts that need to be replaced or / and repaired by the vehicle parts prediction model based on the above information.

[0102] Step 402: Output the names and models of the accessories that need to be replaced and / or repaired in the form of a list.

[0103] By adopting the first prediction model, the name and model of the accessory that needs to be replaced or / and repaired are output according to the image to be identified input by the user, which is more intelligent.

[0104] In some optional implementations, step 203 may optionally include the following steps: using the natural language content recognition component in the LLM-based multimodal image-text relationship prediction model to extract text content from the first list, and extracting the names and models of accessories that need to be replaced or / and repaired in the first list as the vehicle accessories information; or, using natural language recognition technology based on NLP to extract the names and models of accessories that need to be replaced or / and repaired from the first list as the vehicle accessories information; or, using OCR optical text extraction technology to identify and extract the names and models of accessories that need to be replaced or / and repaired from the first list as the vehicle accessories information.

[0105] The names and models of the parts that need to be replaced or / and repaired are identified and extracted from the first list as the vehicle parts information, so that the names and models of the parts that need to be replaced or / and repaired can be subsequently input into the second prediction model.

[0106] Continue to refer Figure 5 In some optional implementations, before executing step 204, a step of pre-training the second prediction model is also included. Figure 5 is a flowchart of a specific embodiment of pre-training the second prediction model in the object screening method described in the present application, comprising the following steps:

[0107] Step 501, extracting identification information of repair manufacturers one by one from the vehicle repair parts lists corresponding to the batch of vehicle accident images, wherein the identification information includes name information of the repair manufacturers;

[0108] Specifically, the identification information of the repair manufacturer is directly extracted from the vehicle repair parts list collected when the first prediction model is trained, and the historical data collected in the past is reasonably used.

[0109] Step 502, using the identification information as an index field, obtaining multi-dimensional feature information corresponding to all maintenance manufacturers from multiple preset data sources, wherein the multi-dimensional feature information includes maintenance technology level information, maintenance service attitude information, maintenance total price information, maintenance time consumption information, and maintenance factory address information;

[0110] Specifically, it should be understood that the preset multiple data sources may include authoritative maintenance management data sources, maintenance data provided by auto insurance service providers, or maintenance data self-built by maintenance manufacturers. Due to different sources, there are certain differences in reference value.

[0111] Step 503, input the multi-dimensional feature information as multi-dimensional training data into the maintenance object screening prediction model to be trained, and perform learning and training on the maintenance object screening prediction model to be trained, learn the maintenance technology level information, maintenance service attitude information of all maintenance manufacturers, and the maintenance price information and maintenance time information of all maintenance manufacturers when repairing different vehicle parts, and obtain a pre-trained maintenance object screening prediction model.

[0112] Specifically, it should be understood that during the learning and training stage of the maintenance object screening prediction model, the training data input into the model includes the maintenance technology level information, maintenance service attitude information, total maintenance price information, maintenance time information, and maintenance plant address information corresponding to all maintenance manufacturers, and the information obtained by learning is the scores corresponding to different maintenance manufacturers in different dimensions.

[0113] By training the maintenance object screening prediction model, it is possible to directly select suitable maintenance vendors based on the model input content during subsequent actual use, and provide users with maintenance recommendation services.

[0114] Continue to refer Figure 6 , in some optional implementations, Figure 6 yes Figure 5 The flowchart of a specific embodiment of step 503 shown includes the following steps:

[0115] Step 601, parsing the multi-dimensional feature information corresponding to all maintenance manufacturers respectively obtained from the preset multiple data sources;

[0116] Specifically, the maintenance technology level information, maintenance service attitude information, total maintenance price information, maintenance time information, and maintenance shop address information corresponding to all maintenance manufacturers on different data source ends are parsed. For example, there are three different data source ends mentioned above, namely, data source A operated by the regulatory agency, data source B for automobile maintenance related forums, and data source C for the data source provided by auto insurance service providers. On different data source ends, car owners' evaluations of different maintenance manufacturers may vary greatly.

[0117] Step 602, obtaining the preset confidence weights of the plurality of preset data source terminals;

[0118] Specifically, due to different sources, data from different data sources have certain differences in reference value, that is, different data sources correspond to different confidence weights. For example, the confidence weight of data source A is greater than that of data source B, and the confidence weight of data source B is greater than that of data source C.

[0119] Step 603: Based on the confidence weights of the multiple data sources and the multi-dimensional feature information, a weighted average method is adopted for each dimensional feature to obtain the maintenance technology level score, maintenance service attitude score, maintenance price score, maintenance time consumption score and maintenance factory distance score corresponding to all maintenance manufacturers;

[0120] For example, taking the maintenance technical level information as an example, assuming that the scores provided by data source A, data source B, and data source C are 8 points, 3 points, and 7 points respectively, and the confidence weights are 0.90, 0.89, and 0.40 respectively, then the total score S is That is, S is about 4.22, which means the actual maintenance technology level is 4.22;

[0121] In addition, the maintenance technical level score can also be calculated based on the collected qualification certificates of maintenance personnel in different maintenance manufacturers. For example, there are 10 maintenance personnel in maintenance manufacturer A, including 2 with advanced maintenance certificates, 1 with intermediate maintenance certificate, 1 with low-level maintenance certificate, and 4 apprentices. Assuming that each certificate is given different scores according to different levels, the average score is calculated to obtain the actual maintenance technical level score.

[0122] Accordingly, the maintenance service attitude information is obtained by comprehensively calculating the service scores of the maintenance shops disclosed by multiple data sources;

[0123] Accordingly, the repair price score is obtained by comprehensively calculating the repair price rationality scores of the repair shops disclosed by multiple data sources;

[0124] Correspondingly, the maintenance time consumption score is obtained by comprehensively calculating the maintenance time consumption scores of the maintenance shops disclosed by multiple data sources. Of course, it can also be obtained by combining the time required to replace and / or repair the parts, whether there is a strict maintenance sequence between the parts, whether they can be repaired in parallel, etc.

[0125] The maintenance technology level scores, maintenance service attitude scores, maintenance price scores, maintenance time scores and maintenance plant distance scores corresponding to all the above-mentioned maintenance manufacturers can be obtained according to different comprehensive analysis strategies, and the specific strategies are freely set by the second prediction model training personnel.

[0126] Step 604, using the maintenance technology level scores, maintenance service attitude scores, maintenance price scores, maintenance time scores and maintenance factory distance scores corresponding to all maintenance manufacturers as reference knowledge, and deploying them into the maintenance object screening prediction model; and

[0127] Step 605, dividing the score levels of different dimensions according to the maintenance technology level score, maintenance service attitude score, maintenance price score, maintenance time score and maintenance plant distance score;

[0128] The maintenance technology level scores, maintenance service attitude scores, maintenance price scores, maintenance time scores and maintenance plant distance scores corresponding to all maintenance manufacturers are used as reference knowledge and deployed into the maintenance object screening prediction model, so that when the second prediction model is used for actual prediction, maintenance manufacturers that meet the priority screening conditions can be screened.

[0129] Step 606: Setting the score levels of the different dimensions as selectable options in the priority screening conditions is sent to the user interface, so that the user can select the repair screening conditions when uploading the image to be identified.

[0130] By setting the score levels of the different dimensions as checkable options in the priority screening conditions and sending them to the user-end interface, the user can select the maintenance screening conditions when uploading the image to be identified, so that the maintenance screening conditions can be set to the user operation interface for the user to perform priority screening, so as to unify the user screening conditions and the model prediction conditions, and improve the prediction efficiency and interpretability of the second prediction model.

[0131] Continue to refer Figure 7 , in some optional implementations, Figure 7 yes Figure 2 The flowchart of a specific embodiment of step 204 shown includes the following steps:

[0132] Step 701, inputting the vehicle parts information and the priority screening condition into the pre-trained maintenance object screening prediction model;

[0133] Step 702, by parsing and identifying the vehicle accessory information, determining all vehicle accessory names and accessory models contained in the vehicle accessory information;

[0134] Step 703, by analyzing and identifying the priority screening conditions, determining the priority screening conditions of the maintenance vendors selected by the user through the user terminal;

[0135] Step 704, performing a primary screening on the maintenance manufacturers in the pre-trained maintenance object screening prediction model according to the maintenance manufacturer screening priority condition to obtain a primary screening result;

[0136] Specifically, according to the maintenance manufacturer screening priority condition, the maintenance manufacturers in the pre-trained maintenance object screening prediction model are initially screened to obtain the initial screening result. It can be set to only screen out one optimal result, or a certain number of optimal solutions can be screened out for the user to click and select. No mandatory setting is performed here.

[0137] Step 705, generating a repair quotation result and comprehensive information of a target repair manufacturer based on the repair price information and repair time information of all repair manufacturers when repairing different vehicle parts in the initial screening result, and all vehicle parts names and parts models included in the vehicle parts information;

[0138] Step 706: Output the maintenance quotation result and the comprehensive information of the target maintenance manufacturer in a list form from the output position of the pre-trained maintenance object screening prediction model to obtain the second list.

[0139] By combining the vehicle parts information predicted by the first prediction model and the maintenance vendor screening priority conditions selected by the user, a certain number of maintenance vendors are initially screened out, and then combined with the maintenance price information and maintenance time information of the certain number of maintenance vendors when repairing different vehicle parts, maintenance quotation results and comprehensive information of target maintenance vendors are generated for the user to choose, which is more intelligent and automated, and provides car owners with better quality and more satisfactory vehicle maintenance recommendation services.

[0140] The present application obtains the image to be identified and the priority screening conditions uploaded by the user end; inputs the image to be identified into the first prediction model to obtain the first list; extracts the vehicle parts information that needs to be replaced or / and repaired in the first list; inputs the vehicle parts information and the priority screening conditions into the second prediction model to obtain the second list; extracts the comprehensive information of the repair manufacturer from the second list, and pushes the comprehensive information of the repair manufacturer to the user end, wherein the comprehensive information includes the repair technology level information, the repair service attitude information, the total repair price information, the repair time information, the repair shop address information, and the repair shop name information. The object screening method described in the present application combines the two previous and subsequent predictions to sequentially predict the vehicle parts information that needs to be replaced or / and repaired during vehicle maintenance, and predicts the repair shops that meet the priority screening conditions, so as to facilitate the user end to select the repair shop, be more intelligent and automated, and provide the car owner with a better quality and more satisfactory vehicle maintenance recommendation service.

[0141] 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.

[0142] 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.

[0143] In the embodiment of the present application, by obtaining the image to be identified and the priority screening conditions uploaded by the user end; inputting the image to be identified into the first prediction model to obtain the first list; extracting the vehicle parts information that need to be replaced or / and repaired in the first list; inputting the vehicle parts information and the priority screening conditions into the second prediction model to obtain the second list; extracting the comprehensive information of the repair manufacturer from the second list, and pushing the comprehensive information of the repair manufacturer to the user end, wherein the comprehensive information includes the repair technology level information, the repair service attitude information, the total repair price information, the repair time information, the repair shop address information, and the repair shop name information. The object screening method described in the present application combines the two previous and subsequent predictions to sequentially predict the vehicle parts information that need to be replaced or / and repaired during vehicle maintenance, and predicts the repair shops that meet the priority screening conditions, so as to facilitate the user end to select the repair shop, be more intelligent and automated, and provide the car owner with a better quality and more satisfactory vehicle maintenance recommendation service.

[0144] Further references Figure 8 , 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 an object screening device, and the device embodiment is Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0145] like Figure 8 As shown, the object screening device 800 described in this embodiment includes: a user-end uploaded information acquisition module 801, a first prediction module 802, a vehicle parts information extraction module 803, a second prediction module 804 and a user-end push module 805. Among them:

[0146] The user-end uploaded information acquisition module 801 is used to acquire the image to be identified and the priority screening conditions uploaded by the user end;

[0147] A first prediction module 802, configured to input the image to be recognized into a pre-trained first prediction model to obtain a first list output by the first prediction model;

[0148] The vehicle parts information extraction module 803 is used to extract text content from the first list to extract the vehicle parts information that needs to be replaced or / and repaired in the first list;

[0149] A second prediction module 804, configured to input the vehicle parts information and the priority screening condition into a pre-trained second prediction model to obtain a second list output by the second prediction model;

[0150] The push module 805 to the user terminal is used to extract the comprehensive information of the target repair manufacturer from the second list and push the comprehensive information of the target repair manufacturer to the user terminal, wherein the comprehensive information includes maintenance technology level information, maintenance service attitude information, total maintenance price information, maintenance time information, maintenance plant address information, and maintenance plant name information.

[0151] The present application obtains the image to be identified and the priority screening conditions uploaded by the user end; inputs the image to be identified into the first prediction model to obtain the first list; extracts the vehicle parts information that needs to be replaced or / and repaired in the first list; inputs the vehicle parts information and the priority screening conditions into the second prediction model to obtain the second list; extracts the comprehensive information of the repair manufacturer from the second list, and pushes the comprehensive information of the repair manufacturer to the user end, wherein the comprehensive information includes the repair technology level information, the repair service attitude information, the total repair price information, the repair time information, the repair shop address information, and the repair shop name information. The object screening method described in the present application combines the two previous and subsequent predictions to sequentially predict the vehicle parts information that needs to be replaced or / and repaired during vehicle maintenance, and predicts the repair shops that meet the priority screening conditions, so as to facilitate the user end to select the repair shop, be more intelligent and automated, and provide the car owner with a better quality and more satisfactory vehicle maintenance recommendation service.

[0152] 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).

[0153] 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.

[0154] To solve the above technical problems, the present application also provides a computer device. Fig. 9 , Fig. 9 This is a basic structural block diagram of the computer device in this embodiment.

[0155] The computer device 9 includes a memory 9a, a processor 9b, and a network interface 9c which are interconnected through a system bus. It should be noted that: Fig. 9 Only a computer device 9 having a component memory 9a, a processor 9b, and a network interface 9c is shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may 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 a microprocessor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital processor (DSP), an embedded device, etc.

[0156] 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.

[0157] The memory 9a 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, disk, optical disk, etc. In some embodiments, the memory 9a can be an internal storage unit of the computer device 9, such as a hard disk or memory of the computer device 9. In other embodiments, the memory 9a can also be an external storage device of the computer device 9, such as a plug-in hard disk equipped on the computer device 9, a smart memory card (Smart Med i aCard, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 9a can also include both the internal storage unit of the computer device 9 and its external storage device. In this embodiment, the memory 9a is generally used to store the operating system and various application software installed on the computer device 9, such as a computer-readable instruction of an object screening method, etc. In addition, the memory 9a can also be used to temporarily store various types of data that have been output or are to be output.

[0158] The processor 9b may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chips in some embodiments. The processor 9b is generally used to control the overall operation of the computer device 9. In the present embodiment, the processor 9b is used to run the computer-readable instructions or process data stored in the memory 9a, such as the computer-readable instructions for running the object screening method.

[0159] The network interface 9c may include a wireless network interface or a wired network interface. The network interface 9c is generally used to establish a communication connection between the computer device 9 and other electronic devices.

[0160] The computer device proposed in this embodiment belongs to the field of artificial intelligence technology and is applied to vehicle maintenance recommendation scenarios. This application obtains the image to be identified and the priority screening conditions uploaded by the user end; inputs the image to be identified into the first prediction model to obtain the first list; extracts the vehicle parts information that needs to be replaced or / and repaired in the first list; inputs the vehicle parts information and the priority screening conditions into the second prediction model to obtain the second list; extracts the comprehensive information of the maintenance manufacturer from the second list, and pushes the comprehensive information of the maintenance manufacturer to the user end, wherein the comprehensive information includes maintenance technology level information, maintenance service attitude information, maintenance total price information, maintenance time information, maintenance plant address information, and maintenance plant name information. The object screening method described in this application combines the previous and subsequent predictions to sequentially predict the vehicle parts information that needs to be replaced or / and repaired during vehicle maintenance, and predicts the maintenance manufacturers that meet the priority screening conditions, so as to facilitate the user end to select the maintenance manufacturer, be more intelligent and automated, and provide car owners with better quality and more satisfactory vehicle maintenance recommendation services.

[0161] The present application also provides another embodiment, that is, 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 a processor to enable the processor to perform the steps of the object screening method as described above.

[0162] The computer-readable storage medium proposed in this embodiment belongs to the field of artificial intelligence technology and is applied to vehicle maintenance recommendation scenarios. This application obtains the image to be identified and the priority screening conditions uploaded by the user end; inputs the image to be identified into the first prediction model to obtain the first list; extracts the vehicle parts information that needs to be replaced or / and repaired in the first list; inputs the vehicle parts information and the priority screening conditions into the second prediction model to obtain the second list; extracts the comprehensive information of the maintenance manufacturer from the second list, and pushes the comprehensive information of the maintenance manufacturer to the user end, wherein the comprehensive information includes maintenance technology level information, maintenance service attitude information, maintenance total price information, maintenance time information, maintenance plant address information, and maintenance plant name information. The object screening method described in this application combines the previous and subsequent predictions to sequentially predict the vehicle parts information that needs to be replaced or / and repaired during vehicle maintenance, and predicts the maintenance manufacturers that meet the priority screening conditions, so as to facilitate the user end to select the maintenance manufacturer, be more intelligent and automated, and provide car owners with better quality and more satisfactory vehicle maintenance recommendation services.

[0163] 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.

[0164] 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 method for object screening, characterized in that: The steps include: Obtain the image to be identified and the priority screening conditions uploaded by the user; Inputting the image to be recognized into a pre-trained first prediction model to obtain a first list output by the first prediction model; Extracting text content from the first list to extract information about vehicle parts that need to be replaced or / and repaired in the first list; Inputting the vehicle accessory information and the priority screening condition into a pre-trained second prediction model to obtain a second list output by the second prediction model; The comprehensive information of the target repair manufacturer is extracted from the second list, and the comprehensive information of the target repair manufacturer is pushed to the user end, wherein the comprehensive information includes repair technology level information, repair service attitude information, total repair price information, repair time information, repair shop address information, and repair shop name information.

2. The object screening method according to claim 1, characterized in that: The step of obtaining the image to be identified and the priority screening conditions uploaded by the user terminal specifically includes: Acquire the image to be identified and the priority screening conditions uploaded by the user according to the preset prompt content, wherein the preset prompt content includes an image clarity compliance field, an image import position indication field, and screening condition information for prompting the user to check, the image to be identified includes an image of a damaged or destroyed part of a vehicle, and the screening condition information includes maintenance technology level information, maintenance service attitude information, total maintenance price information, maintenance time information, and maintenance shop address information.

3. The object screening method according to claim 1, characterized in that: The first prediction model includes a vehicle parts prediction model, the first list includes information about vehicle parts that need to be replaced or / and repaired, and the vehicle parts information includes a name of a part and a model of a part. Before executing the step of inputting the image to be recognized into the pre-trained first prediction model to obtain a first list output by the first prediction model, the method further includes: Acquire a batch of vehicle accident images collected in advance and a list of vehicle repair parts corresponding to the batch of vehicle accident images, and construct a training data set with a one-to-one mapping between the images and the list; Inputting the training data set into a vehicle parts prediction model to be trained, performing learning training on the vehicle parts prediction model, and obtaining a pre-trained vehicle parts prediction model, wherein the vehicle parts prediction model includes a multimodal graph-text relationship prediction model based on LLM; The step of inputting the image to be recognized into a pre-trained first prediction model to obtain a first list output by the first prediction model specifically includes: Inputting the image of the damaged or destroyed part of the vehicle into a pre-trained vehicle parts prediction model to predict the name and model of the parts that need to be replaced or / and repaired by the vehicle parts prediction model; The names and models of the parts that need to be replaced and / or repaired are output in the form of a list.

4. The object screening method according to claim 3, characterized in that: The step of extracting text content from the first list to extract information about vehicle parts that need to be replaced or / and repaired in the first list specifically includes: The natural language content recognition component in the LLM-based multimodal image-text relationship prediction model is used to extract text content from the first list, and the names and models of the parts that need to be replaced or / and repaired in the first list are extracted as the vehicle parts information. or, Using natural language recognition technology based on NLP, the names and models of the parts that need to be replaced or / and repaired are extracted from the first list as the vehicle parts information, or, The OCR optical character extraction technology is used to identify and extract the names and models of the parts that need to be replaced and / or repaired from the first list as the vehicle parts information.

5. The object screening method according to claim 3, characterized in that: The second prediction model includes a maintenance object screening prediction model, and the maintenance object refers to a maintenance manufacturer that repairs the accident vehicle. Before executing the step of inputting the vehicle parts information and the priority screening condition into the pre-trained second prediction model to obtain a second list output by the second prediction model, the method further includes: Extracting identification information of the repair manufacturers one by one from the vehicle repair parts lists corresponding to the batch of vehicle accident images, wherein the identification information includes name information of the repair manufacturers; Using the identification information as an index field, multi-dimensional feature information corresponding to all maintenance vendors is obtained from a plurality of preset data sources, wherein the multi-dimensional feature information includes maintenance technology level information, maintenance service attitude information, maintenance total price information, maintenance time consumption information, and maintenance factory address information; The multi-dimensional feature information is input as multi-dimensional training data into the maintenance object screening prediction model to be trained, and the maintenance object screening prediction model to be trained is trained to learn the maintenance technology level information, maintenance service attitude information of all maintenance manufacturers, and maintenance price information and maintenance time information of all maintenance manufacturers when repairing different vehicle parts, so as to obtain a pre-trained maintenance object screening prediction model.

6. The object screening method according to claim 5, characterized in that: The step of training the maintenance object screening prediction model to be trained to learn the maintenance technology level information, maintenance service attitude information of all maintenance manufacturers, and maintenance price information and maintenance time consumption information of all maintenance manufacturers when repairing different vehicle parts specifically includes: Analyzing the multi-dimensional feature information corresponding to all maintenance manufacturers respectively obtained from the preset multiple data sources; Obtaining preset confidence weights of the plurality of preset data source terminals; Based on the confidence weights of the multiple data sources and the multi-dimensional feature information, a weighted average method is adopted for each dimensional feature to obtain the maintenance technology level score, maintenance service attitude score, maintenance price score, maintenance time consumption score and maintenance factory distance score corresponding to all maintenance manufacturers; The maintenance technology level scores, maintenance service attitude scores, maintenance price scores, maintenance time scores and maintenance factory distance scores corresponding to all maintenance manufacturers are used as reference knowledge and deployed into the maintenance object screening prediction model; and, According to the maintenance technology level score, maintenance service attitude score, maintenance price score, maintenance time score and maintenance plant distance score, score levels of different dimensions are divided respectively; The score levels of the different dimensions are set as checkable options in the priority screening conditions and sent to the user end interface, so that the user can select the maintenance screening conditions when uploading the image to be identified.

7. The object screening method according to claim 5, characterized in that: The step of inputting the vehicle accessory information and the priority screening condition into the pre-trained second prediction model to obtain the second list output by the second prediction model specifically includes: Inputting the vehicle parts information and the priority screening conditions into the pre-trained maintenance object screening prediction model; By parsing and identifying the vehicle accessory information, all vehicle accessory names and accessory models contained in the vehicle accessory information are determined; By analyzing and identifying the priority screening conditions, determining the priority screening conditions of the maintenance vendors selected by the user through the user terminal; According to the maintenance manufacturer screening priority condition, the maintenance manufacturers in the pre-trained maintenance object screening prediction model are initially screened to obtain an initial screening result; Generate a repair quotation result and comprehensive information of the target repair manufacturer based on the repair price information and repair time information of all repair manufacturers when repairing different vehicle parts in the initial screening results, and all vehicle parts names and parts models contained in the vehicle parts information; The maintenance quotation result and the comprehensive information of the target maintenance manufacturer are output in a list form from the output position of the pre-trained maintenance object screening prediction model to obtain the second list.

8. An object screening device, characterized in that: include: The user-end uploaded information acquisition module is used to obtain the image to be identified and the priority screening conditions uploaded by the user end; A first prediction module, used for inputting the image to be recognized into a pre-trained first prediction model to obtain a first list output by the first prediction model; A vehicle parts information extraction module, used to extract text content from the first list to extract information about vehicle parts that need to be replaced or / and repaired in the first list; A second prediction module, used for inputting the vehicle parts information and the priority screening condition into a pre-trained second prediction model to obtain a second list output by the second prediction model; The push module to the user terminal is used to extract the comprehensive information of the target repair manufacturer from the second list and push the comprehensive information of the target repair manufacturer to the user terminal, wherein the comprehensive information includes maintenance technology level information, maintenance service attitude information, total maintenance price information, maintenance time information, maintenance plant address information, and maintenance plant name information.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the object screening method according to any one of claims 1 to 7 when executing the computer-readable instructions.

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 object screening method according to any one of claims 1 to 7 are implemented.

Images