Vehicle damage photo shooting method and device, electronic equipment and storage medium

Abstract

The invention provides a vehicle damage photo shooting method and device, electronic equipment and a storage medium, and relates to the technical field of image processing and the field of financial science and technology. The method comprises the following steps: shooting a target damaged vehicle to obtain an initial vehicle damage picture; performing area identification on the initial vehicle damage picture to obtain vehicle damage area information and background area information; wherein the background region information comprises at least one of the following information: background region light information, background region disorder degree information and background region size information; according to the vehicle damage area information and the background area information, generating target vehicle damage photo shooting guidance; shooting the target damaged vehicle according to the target vehicle damage picture shooting guidance to obtain candidate vehicle damage pictures; and performing image enhancement on the candidate vehicle damage picture to obtain a target vehicle damage picture. According to the invention, the quality of the vehicle damage picture can be improved, and the dependence on photographing knowledge and experience of photographers is reduced.

Description

Technical Field

[0001] The present application relates to the field of image processing technology and financial technology, and in particular to a method and device for taking photos of vehicle damage, an electronic device, and a storage medium. Background Art

[0002] In the field of financial technology, photos of vehicle damage (hereinafter referred to as vehicle damage photos) can play an important role. For example, in the insurance industry, vehicle damage photos are an important basis for evaluating and processing insurance claims. Currently, most insurance companies rely on on-site surveyors or car owners to take photos of vehicle damage using their mobile phones and then upload them to the insurance company through an application.

[0003] There are some problems with existing photo shooting methods and tools, resulting in low quality of vehicle damage photos. For example, existing photo shooting methods mainly rely on the subjective judgment and shooting skills of surveyors or car owners, while most novice surveyors and car owners lack professional photography knowledge and experience. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a method and device for taking vehicle damage photos, an electronic device, and a storage medium, which can improve the quality of vehicle damage photos and reduce the reliance on the photographer's knowledge and experience in photography.

[0005] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application provides a method for taking a vehicle damage photo, the method comprising:

[0006] Take photos of the target damaged vehicle to obtain initial vehicle damage photos;

[0007] Performing region recognition on the initial vehicle damage photo to obtain vehicle damaged area information and background area information; wherein the background area information includes at least one of the following: background area light information, background area clutter information, and background area size information;

[0008] Generate a target vehicle damage photo shooting guide based on the vehicle damaged area information and the background area information;

[0009] Photographing the target damaged vehicle according to the target vehicle damage photo shooting guide to obtain candidate vehicle damage photos;

[0010] Perform image enhancement on the candidate vehicle damage photo to obtain a target vehicle damage photo.

[0011] In some embodiments, after performing image enhancement on the candidate vehicle damage photo to obtain the target vehicle damage photo, the method further includes:

[0012] Perform damage type recognition on the target vehicle damage photo to obtain a target damage type;

[0013] Obtaining the claim number of the target damaged vehicle;

[0014] A claim is made for the target damaged vehicle according to the basic claim information corresponding to the claim number and the target damage type.

[0015] In some embodiments, the step of identifying the damage type of the target vehicle damage photo to obtain the target damage type includes:

[0016] Performing damaged area detection on the target vehicle damage photo to obtain a target damaged area;

[0017] Extracting features of the target damaged area to obtain target damaged features;

[0018] The target damage type is determined according to the target damage characteristics.

[0019] In some embodiments, performing damaged area detection on the target vehicle damage photo to obtain the target damaged area includes:

[0020] Performing edge strength recognition on the target vehicle damage photo to obtain target edge strength features;

[0021] Performing color difference recognition on the target vehicle damage photo to obtain target color difference features;

[0022] Performing texture change recognition on the target vehicle damage photo to obtain target texture change features;

[0023] The target damaged area is obtained by performing region division according to the target edge intensity feature, the target color difference feature and the target texture change feature.

[0024] In some embodiments, performing region division according to the target edge intensity feature, the target color difference feature, and the target texture change feature to obtain the target damaged area includes:

[0025] Performing attention processing on the target edge intensity feature, the target color difference feature and the target texture change feature to obtain a damaged area attention feature;

[0026] Regional mapping is performed according to the attention features of the damaged area to obtain the target damaged area.

[0027] In some embodiments, the target vehicle damage photo shooting guidance includes target shooting light guidance;

[0028] The generating of a target vehicle damage photo shooting guide according to the vehicle damaged area information and the background area information includes:

[0029] Extracting the visual information of the damaged area of ​​the vehicle from the damaged area information of the vehicle;

[0030] The target shooting light guide is generated according to the visual information of the damaged area of ​​the vehicle and the light information of the background area, wherein the target shooting light guide is used to guide the shooting of photos with light meeting predetermined light requirements.

[0031] In some embodiments, the target vehicle damage photo shooting guidance includes a target shooting angle guidance;

[0032] The generating of a target vehicle damage photo shooting guide according to the vehicle damaged area information and the background area information includes:

[0033] Extracting vehicle damaged area location information from the vehicle damaged area information;

[0034] Extracting the clutter level information of at least two background sub-areas from the clutter level information of the background area; wherein each of the background sub-areas constitutes the background area of ​​the initial vehicle damage photo;

[0035] Selecting the background sub-region with the lowest clutter level from the clutter level information of each background sub-region to obtain a target background sub-region;

[0036] The target shooting angle guidance is generated according to the position information of the damaged area of ​​the vehicle and the position information corresponding to the target background sub-area, wherein the target shooting angle guidance is used to guide the taking of photos whose clutter level of the background area meets the predetermined clutter level requirement.

[0037] In some embodiments, the target vehicle damage photo shooting guide includes a target shooting distance guide;

[0038] The generating of a target vehicle damage photo shooting guide according to the vehicle damaged area information and the background area information includes:

[0039] Extracting damaged area size information from the damaged area information of the vehicle;

[0040] The target shooting distance guide is generated according to the damaged area size information and the background area size information; wherein the target shooting distance guide is used to guide the taking of photos in which the damaged area and the background area meet a predetermined size ratio.

[0041] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a vehicle damage photo shooting device, the device comprising:

[0042] The first shooting module is used to shoot the target damaged vehicle to obtain an initial vehicle damage photo;

[0043] A region recognition module is used to perform region recognition on the initial vehicle damage photo to obtain vehicle damaged region information and background region information; wherein the background region information includes at least one of the following: background region light information, background region clutter information, and background region size information;

[0044] A guide generation module, used for generating a guide for taking a target vehicle damage photo according to the vehicle damaged area information and the background area information;

[0045] A second shooting module is used to shoot the target damaged vehicle according to the target vehicle damage photo shooting guide to obtain candidate vehicle damage photos;

[0046] The image enhancement module is used to perform image enhancement on the candidate vehicle damage photo to obtain a target vehicle damage photo.

[0047] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the method for taking vehicle damage photos described in the first aspect when executing the computer program.

[0048] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method for taking vehicle damage photos described in the first aspect above.

[0049] The vehicle damage photo shooting method and device, electronic device, and storage medium proposed in the present application first shoot the target damaged vehicle to obtain an initial vehicle damage photo. The quality of the initial vehicle damage photo may be poor, such as poor light, a cluttered background area, or a large background area size. Secondly, the initial vehicle damage photo is region identified to obtain vehicle damage area information and background area information. The vehicle damage area information represents the information of the damaged area of ​​the target damaged vehicle. The background area information represents the information of the environment where the target damaged vehicle is located. The background area information includes at least one of the following: background area light information, background area clutter degree information, and background area size information. Then, according to the vehicle damage area information and the background area information, a target vehicle damage photo shooting guide is generated. The target vehicle damage photo shooting guide can guide the shooting of photos with better quality. Then, according to the target vehicle damage photo shooting guide, the target damaged vehicle is shot to obtain a candidate vehicle damage photo, which has good performance at least in terms of light, cluttered background area, or background area size. Finally, the candidate vehicle damage photo is image enhanced to obtain a target vehicle damage photo, so that the target vehicle damage photo can highlight the damaged area of ​​the target damaged vehicle. In summary, the embodiments of the present application can improve the quality of vehicle damage photos and reduce the reliance on the photographer's knowledge and experience in photography.

[0050] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flow chart of a method for taking vehicle damage photos provided in an embodiment of the present application;

[0052] Figure 2 yes Figure 1 The first flow chart of step 103 in;

[0053] Figure 3 yes Figure 1 A second flow chart of step 103 in FIG. 1 ;

[0054] Figure 4 yes Figure 1 A third flow chart of step 103 in FIG. 1 ;

[0055] Figure 5 is a flow chart of a method for taking vehicle damage photos provided by another embodiment of the present application;

[0056] Figure 6 yes Figure 5 Flow chart of step 501 in;

[0057] Figure 7 yes Figure 6 Flowchart of step 601 in FIG.

[0058] Figure 8 It is a block diagram of the module structure of the vehicle damage photo shooting device provided in an embodiment of the present application;

[0059] Fig. 9 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0061] It should be noted that, although the functional modules are divided in the device schematic diagram and the 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 above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0063] First, some nouns involved in this application are analyzed:

[0064] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a 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.

[0065] Natural language processing (NLP): NLP uses computers to process, understand and apply human languages ​​(such as Chinese, English, etc.). NLP is a branch of artificial intelligence and an interdisciplinary subject between computer science and linguistics. It is often referred to as computational linguistics. Natural language processing includes grammatical analysis, semantic analysis, and text comprehension. Natural language processing is commonly used in technical fields such as machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, information image processing, information extraction and filtering, text classification and clustering, public opinion analysis and opinion mining. It involves data mining related to language processing, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computing.

[0066] There are some problems with existing photo shooting methods and tools, resulting in low quality of vehicle damage photos. Specifically, existing photos often have the following problems: (1) Photos taken by non-professionals are not taken at the right angles and are not accurate, which cannot fully show the damaged parts and extent of the vehicle. (2) Insufficient or excessive light, poor lighting conditions, resulting in blurred or overexposed photos, insufficient photo resolution, unclear details, and other factors that affect the quality of photos and further reduce the accuracy of damage assessment. (3) Lack of standardized shooting guidance and appropriate reference objects make it difficult to accurately judge the size and location of damaged parts, making it difficult to effectively compare damage to different parts of the same vehicle. These problems not only increase the difficulty of insurance companies' audits, but may also lead to delays in claims processing and affect customer experience.

[0067] Based on the above problems, the embodiments of the present application provide a method and device for taking photos of vehicle damage, an electronic device, and a storage medium. The embodiments of the present application aim to propose a solution for improving the quality of taking photos of vehicle damage based on AI assistance. The core of this solution is to guide users to take high-quality photos of vehicle damage through intelligent algorithms to ensure that the photos can accurately reflect the damage to the vehicle, thereby improving the efficiency and accuracy of damage assessment. Specifically, photo shooting instructions are given during the shooting process, and after obtaining photos taken based on the photo shooting instructions, image enhancement processing is also performed on the photos. In this way, the embodiments of the present application can improve the quality of vehicle damage photos and reduce the reliance on the photographer's knowledge and experience in photography.

[0068] The method for taking photos of damaged vehicles provided in the embodiment of the present application can be applied to the terminal and the server side, and can also be software running on the server side. The server side can be configured as an independent physical server, or as a server cluster or distributed system consisting of multiple physical servers, or as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the method for taking photos of damaged vehicles, etc., but is not limited to the above forms.

[0069] The present application can be used in many general or special computer system environments or configurations. For example: server computers, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, distributed computing environments including any of the above systems or devices, etc. The present 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. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0070] The embodiments of the present application provide a method for taking vehicle damage photos, a device for taking vehicle damage photos, an electronic device, and a computer-readable storage medium, which are specifically illustrated by the following embodiments. First, the method for taking vehicle damage photos in the embodiments of the present application is described.

[0071] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0072] Reference Figure 1 , Figure 1 It is an optional flowchart of the method for taking vehicle damage photos provided in an embodiment of the present application, which may include but is not limited to steps 101 to 105.

[0073] Step 101, photographing the target damaged vehicle to obtain an initial vehicle damage photo;

[0074] Step 102, performing region recognition on the initial vehicle damage photo to obtain vehicle damaged region information and background region information; wherein the background region information includes at least one of the following: background region light information, background region clutter information, and background region size information;

[0075] Step 103, generating a target vehicle damage photo shooting guide based on the vehicle damaged area information and background area information;

[0076] Step 104, photographing the target damaged vehicle according to the target vehicle damage photo photographing guide to obtain candidate vehicle damage photos;

[0077] Step 105, performing image enhancement on the candidate vehicle damage photo to obtain the target vehicle damage photo.

[0078] In the steps 101 to 105 shown in the embodiment of the present application, the target damaged vehicle is first photographed for the first time to obtain an initial vehicle damage photo. The quality of the initial vehicle damage photo may be poor, such as poor light, a cluttered background area, or a large background area size. Secondly, the initial vehicle damage photo is region identified to obtain vehicle damaged area information and background area information. The vehicle damaged area information represents information about the damaged area of ​​the target damaged vehicle. The background area information represents information about the environment where the target damaged vehicle is located. The background area information includes at least one of the following: background area light information, background area clutter degree information, and background area size information. Then, according to the vehicle damaged area information and the background area information, a target vehicle damage photo shooting guide is generated. The target vehicle damage photo shooting guide can guide the shooting of photos with better quality. Then, according to the target vehicle damage photo shooting guide, the target damaged vehicle is photographed to obtain a candidate vehicle damage photo, which has good performance at least in terms of light, cluttered background area, or background area size. Finally, the candidate vehicle damage photo is image enhanced to obtain a target vehicle damage photo, so that the target vehicle damage photo can highlight the damaged area of ​​the target damaged vehicle. In summary, the embodiments of the present application can improve the quality of vehicle damage photos and reduce the reliance on the photographer's knowledge and experience in photography.

[0079] In step 101 of some embodiments, the target damaged vehicle is photographed to obtain an initial vehicle damage photo. The target damaged vehicle refers to a damaged vehicle, specifically a vehicle with damage on the vehicle surface. The target damaged vehicle can be photographed by a camera to obtain the initial vehicle damage photo.

[0080] In one example, a user opens a target application and selects the AI ​​vehicle damage photo taking function in the target application; the terminal / server responds to the touch operation of the AI ​​vehicle damage photo taking function, starts the camera to take a photo of the target damaged vehicle, and obtains an initial vehicle damage photo.

[0081] In step 102 of some embodiments, the initial vehicle damage photo is subjected to region recognition to obtain vehicle damaged area information and background area information. The vehicle damaged area information includes at least one of the following: vehicle damaged area visual information, vehicle damaged area position information, and vehicle damaged area size information. The background area information includes at least one of the following: background area light information, background area clutter level information, and background area size information.

[0082] The visual information of the damaged area of ​​the vehicle includes at least one of the following: exposure, color, and shadow.

[0083] The light information of the background area includes at least one of the following: light type, light intensity, light quality, and light color temperature. Light type refers to the type of light, such as natural light (such as sunlight) and artificial light (such as lights and flash). Light intensity refers to the strength of the light, which affects the brightness and contrast of the photo. Light quality refers to whether the light is soft or hard. Soft light usually produces fewer shadows, while hard light produces obvious shadows and high contrast. Light color temperature refers to the color tendency of the light, such as cold light (blue) or warm light (yellow).

[0084] The clutter level information of the background area includes the clutter level information of at least two background sub-areas. Each background sub-area constitutes the background area of ​​the initial vehicle damage photo. The clutter level information of the background sub-area refers to the complexity and disorder of the background elements. The clutter level information includes at least one of the following: the number of elements, visual interference, contrast, and spatial organization. The number of elements refers to the number of objects, textures, or patterns in the background. The more elements there are, the more cluttered the background appears. Visual interference refers to whether the elements in the background will distract the model from the target damaged vehicle. If the background elements are too complicated, it may cause visual confusion. Contrast refers to whether the color and pattern of the background form a clear contrast with the target damaged vehicle. Too similar colors and complex patterns will increase the sense of clutter. Spatial organization refers to whether the arrangement of background elements is orderly. Good spatial organization will make the background look neat, while irregular arrangement will create a sense of clutter.

[0085] In step 103 of some embodiments, a target vehicle damage photo shooting guide is generated based on the vehicle damaged area information and the background area information. The target vehicle damage photo shooting guide is used to guide the shooting of photos whose quality meets the quality requirements. The target vehicle damage photo shooting guide includes at least one of the following: target shooting light guide, target shooting angle guide, and target shooting distance guide.

[0086] In one embodiment, referring to Figure 2 , step 103 may include:

[0087] Step 201, extracting the visual information of the damaged area of ​​the vehicle from the damaged area information of the vehicle;

[0088] Step 202: Generate target shooting light guidance based on the visual information of the damaged area of ​​the vehicle and the light information of the background area.

[0089] Specifically, the target shooting light guidance is used to guide the shooting of photos with light that meets the predetermined light requirements. A light guidance generation model can be pre-trained, and the light guidance generation model is used to generate guidance for the visual information of the damaged area of ​​the vehicle and the light information of the background area to obtain the target shooting light guidance. The light guidance generation model can use a common large language model, such as a GPT model.

[0090] The benefit of the above-mentioned embodiment of step 201 to step 202 is that it can generate target shooting light guidance, which helps to improve the visual quality of the vehicle damage photos.

[0091] In one embodiment, referring to Figure 3 , step 103 may include:

[0092] Step 301, extracting the location information of the damaged area of ​​the vehicle from the damaged area information of the vehicle;

[0093] Step 302, extracting clutter level information of at least two background sub-areas from the clutter level information of the background area; wherein each background sub-area constitutes the background area of ​​the initial vehicle damage photo;

[0094] Step 303, selecting the background sub-region with the lowest clutter level from the clutter level information of each background sub-region, and obtaining the target background sub-region;

[0095] Step 304: Generate target shooting angle guidance according to the position information of the damaged area of ​​the vehicle and the position information corresponding to the target background sub-area.

[0096] Specifically, the target shooting angle guidance is used to guide the shooting of photos whose clutter level of the background area meets the predetermined clutter level requirement. An angle guidance generation algorithm can be pre-constructed, and the position information of the damaged area of ​​the vehicle and the position information corresponding to the target background sub-area are guided and generated by the angle guidance generation algorithm to obtain the target shooting angle guidance. An angle guidance generation model can also be pre-trained, and the position information of the damaged area of ​​the vehicle and the position information corresponding to the target background sub-area are guided and generated by the angle guidance generation model to obtain the target shooting angle guidance. The angle guidance generation model can use a common large language model, such as a GPT model.

[0097] The benefit of the above-mentioned embodiment of step 301 to step 304 is that it can generate target shooting angle guidance, so as to facilitate the target vehicle damage photo to fully display the damaged parts and degree of damage of the vehicle.

[0098] In one embodiment, referring to Figure 4 , step 103 may include:

[0099] Step 401, extracting damaged area size information from the damaged area information of the vehicle;

[0100] Step 402: Generate target shooting distance guidance according to the damaged area size information and the background area size information.

[0101] Specifically, the target shooting distance guidance is used to guide the shooting of photos in which the damaged area and the background area meet the predetermined size ratio. A size guidance generation algorithm can be pre-constructed, and the damaged area size information and the background area size information are guided and generated by the size guidance generation algorithm to obtain the target shooting distance guidance. A distance guidance generation model can also be pre-trained, and the damaged area size information and the background area size information are guided and generated by the distance guidance generation model to obtain the target shooting distance guidance. The distance guidance generation model can use a common large language model, such as a GPT model.

[0102] The benefit of the above-mentioned embodiment of steps 401 to 402 is that it can generate a target shooting distance guide, facilitate the determination of the size and location of the damaged parts of the vehicle based on the target vehicle damage photos, and enable effective comparison of the damage to different parts of the same target damaged vehicle.

[0103] In one embodiment, step 103 may include at least one of the following:

[0104] (1) Step 201, extracting the visual information of the damaged area of ​​the vehicle from the damaged area information of the vehicle; Step 202, generating the target shooting light guide according to the visual information of the damaged area of ​​the vehicle and the light information of the background area.

[0105] (2) Step 301, extracting the location information of the damaged area of ​​the vehicle from the information of the damaged area of ​​the vehicle; Step 302, extracting the clutter level information of at least two background sub-areas from the clutter level information of the background area; wherein each background sub-area constitutes the background area of ​​the initial vehicle damage photo; Step 303, selecting the background sub-area with the lowest clutter level from the clutter level information of each background sub-area to obtain a target background sub-area; Step 304, generating a target shooting angle guide according to the location information of the damaged area of ​​the vehicle and the location information corresponding to the target background sub-area.

[0106] (3) Step 401, extracting damaged area size information from the damaged area information of the vehicle; Step 402, generating target shooting distance guidance according to the damaged area size information and the background area size information.

[0107] In step 104 of some embodiments, the target damaged vehicle is photographed according to the target vehicle damage photo shooting instructions to obtain candidate vehicle damage photos.

[0108] In one embodiment, the target damaged vehicle is photographed according to at least one of the target shooting light guidance, the target shooting angle guidance, and the target shooting distance guidance to obtain a candidate damaged vehicle photo. The target photographed damaged vehicle photo is different from the initial photographed damaged vehicle photo in at least one of the following: light, background area clutter, and background area size.

[0109] In step 105 of some embodiments, the candidate vehicle damage photo is enhanced to obtain the target vehicle damage photo. In one example, adaptive histogram equalization and local contrast enhancement techniques are used to enhance the candidate vehicle damage photo to improve the clarity and contrast of the target vehicle damage photo. In this way, by integrating the image processing algorithm, the contrast, brightness and color of the photo can be automatically adjusted to ensure the photo quality even under unfavorable shooting conditions.

[0110] In one embodiment, after step 105, refer to Figure 5 The vehicle damage photo shooting method provided in this embodiment may also include:

[0111] Step 501, identifying the damage type of the target vehicle damage photo to obtain the target damage type;

[0112] Step 502, obtaining the claim number of the target damaged vehicle;

[0113] Step 503: Make a claim for the target damaged vehicle according to the basic claim information corresponding to the claim number and the target damage type.

[0114] In step 501, the target damage type refers to the type of damage of the target damaged vehicle. For example, the target damage type includes scratches, dents, cracks, etc. In one example, a damage type recognition model can be constructed based on a variant of a convolutional neural network (CNN), namely U-Net, and the damage type recognition model is used to identify the damage type of the target vehicle damage photo. U-Net is an efficient network structure for image segmentation, which combines encoder and decoder modules to accurately locate and identify specific areas in an image. In this embodiment, U-Net is trained to identify various types of damage on the surface of a vehicle.

[0115] The loss function of the U-Net model mainly uses cross entropy loss, and its mathematical expression is:

[0116] L=-∑cy c log(p c ).

[0117] Among them, L is the cross entropy loss data, c is the regularization term, and y c is the true label, p cis the probability predicted by the model. Adding a regularization term can prevent the model from overfitting and improve the generalization ability of the model. Specifically, during the model training process, the cross entropy loss data is compared with the preset loss threshold. If the cross entropy loss data is greater than the loss threshold, the parameters of the damage type recognition model are adjusted until the cross entropy loss data is less than or equal to the loss threshold, and then the parameter adjustment is stopped. The Adam optimizer can be used for parameter update, setting the learning rate to 0.001, the batch size to 32, and the number of training rounds to 50.

[0118] In one embodiment, referring to Figure 6 , step 501 may include:

[0119] Step 601, performing damaged area detection on the target vehicle damage photo to obtain the target damaged area;

[0120] Step 602, extracting features of the target damaged area to obtain target damaged features;

[0121] Step 603: Determine the target damage type according to the target damage characteristics.

[0122] In step 601, the target damaged area represents the photo area where the damaged parts of the target damaged vehicle are located.

[0123] In one embodiment, referring to Figure 7 , step 601 may include:

[0124] Step 701, performing edge strength recognition on the target vehicle damage photo to obtain target edge strength features;

[0125] Step 702, performing color difference recognition on the target vehicle damage photo to obtain target color difference features;

[0126] Step 703, performing texture change recognition on the target vehicle damage photo to obtain target texture change features;

[0127] Step 704 , performing region division according to the target edge intensity feature, the target color difference feature and the target texture change feature to obtain the target damaged area.

[0128] Specifically, the target edge intensity feature can be used to distinguish the damaged area from the normal vehicle body surface. The target color difference feature can be used to indicate that there is a significant difference in color between the damaged area and the surrounding area of ​​the damaged area. The target texture change feature can indicate that the damaged area has a different texture than the surrounding area of ​​the damaged area, such as being rougher or more irregular. The edge intensity identifier, color difference identifier, and texture change identifier can be pre-trained and used for edge intensity recognition, color difference recognition, and texture change recognition, respectively.

[0129] In one embodiment, step 704 may include: performing attention processing on the target edge intensity feature, the target color difference feature and the target texture change feature to obtain the damaged area attention feature; performing regional mapping according to the damaged area attention feature to obtain the target damaged area. Attention processing is a feature extraction method based on the attention mechanism. The attention mechanism is a neural network mechanism that imitates human attention and is used to improve the efficiency and effectiveness of the model when processing a large amount of information. It dynamically focuses on the important parts of the input data, thereby enhancing the memory and processing capabilities of key information. This embodiment uses attention processing to determine the features related to the damaged area from the target edge intensity feature, the target color difference feature and the target texture change feature, and obtains the damaged area attention feature, thereby determining the target damaged area. This embodiment has good universality.

[0130] The benefit of the above-mentioned embodiment of step 701 to step 704 is that the target damaged area can be accurately identified.

[0131] In step 602, an encoder may be used to extract features of the target damaged area to obtain target damaged features. In step 603, a classifier may be used to classify the target damaged features to obtain target damage types.

[0132] The benefit of the above-mentioned embodiment of step 601 to step 602 is that the damage of the damaged area can be accurately classified.

[0133] In step 502, the claim number of the target damaged vehicle is obtained. The claim number refers to the number of the claim associated with the target damaged vehicle.

[0134] In one embodiment, step 502 may include: displaying a claim form binding interface, wherein a claim form number binding control is displayed in the claim form binding interface, and the claim form number binding control is used to bind a claim form number to a target damaged vehicle; in response to a touch operation on the claim form number binding control, photographing the license plate number of the target damaged vehicle to obtain a target license plate number; and determining the claim form number corresponding to the target damaged vehicle according to the target license plate number. This embodiment obtains the claim form number through a visual interface, and has high interactivity.

[0135] In one embodiment, the above process of determining the claim number corresponding to the target damaged vehicle based on the target license plate number may include: the terminal uploads the target license plate number to the server, and the server searches a preset license plate claim number mapping table based on the license plate number to obtain the claim number; and receives the claim number sent by the server.

[0136] Finally, in step 503, a claim is made for the target damaged vehicle according to the basic claim information and the target damage type corresponding to the claim number. In one embodiment, step 503 may include generating an initial vehicle damage document according to the basic claim information and the target damage type corresponding to the claim number; updating the initial vehicle damage document according to the compensation information corresponding to the target damage type to obtain a target vehicle damage document; and making compensation for the target damaged vehicle according to the target vehicle damage document. The basic claim information includes basic information of the target damaged vehicle, such as the license plate number, vehicle model, and vehicle age. The compensation information includes the compensation amount, compensation conditions, etc. The target vehicle damage document can indicate information such as the damage to the vehicle and the compensation amount. The target vehicle damage document is an important basis for compensation for vehicle damage.

[0137] The benefit of the above-mentioned embodiment of steps 501 to 503 is that claims can be settled for the target damaged vehicle efficiently and with high accuracy.

[0138] In summary, the present application has at least the following beneficial effects: (1) The present application can generate high-quality vehicle damage photos under various shooting conditions and accurately identify the damaged area, thereby greatly improving the efficiency and accuracy of vehicle damage assessment. (2) Not only does it reduce the uncertainty in the damage assessment process, but clear and comprehensive vehicle damage photos also reduce the assessment errors caused by photo quality issues. (3) High-quality photos enable insurance companies to assess vehicle damage more quickly and accurately, shorten the claims processing cycle, and greatly improve claims efficiency. (4) Standardized shooting processes and automatic damage identification functions reduce dependence on professional surveyors and users can easily take photos that meet the requirements, reducing the number of on-site re-inspections conducted by insurance companies due to photo quality issues, reducing operating costs, and also improving customer satisfaction.

[0139] See also Figure 8 The embodiment of the present application also provides a vehicle damage photo shooting device, which can implement the above-mentioned vehicle damage photo shooting method. Figure 8The module structure block diagram of the vehicle damage photo shooting device provided in the embodiment of the present application includes: a first shooting module 801, a region recognition module 802, a guide generation module 803, a second shooting module 804 and an image enhancement module 805. Among them, the first shooting module 801 is used to shoot the target damaged vehicle to obtain an initial vehicle damage photo; the region recognition module 802 is used to perform region recognition on the initial vehicle damage photo to obtain vehicle damaged area information and background area information; wherein the background area information includes at least one of the following: background area light information, background area clutter information and background area size information; the guide generation module 803 is used to generate a target vehicle damage photo shooting guide according to the vehicle damaged area information and the background area information; the second shooting module 804 is used to shoot the target damaged vehicle according to the target vehicle damage photo shooting guide to obtain a candidate vehicle damage photo; the image enhancement module 805 is used to perform image enhancement on the candidate vehicle damage photo to obtain the target vehicle damage photo.

[0140] In one embodiment, after performing image enhancement on the candidate vehicle damage photo to obtain the target vehicle damage photo, the vehicle damage photo shooting device further includes:

[0141] The type recognition module is used to identify the damage type of the target vehicle damage photo and obtain the target damage type;

[0142] The claim number acquisition module is used to obtain the claim number of the target damaged vehicle;

[0143] The vehicle claims module is used to make claims for the target damaged vehicle based on the basic claims information and target damage type corresponding to the claim number.

[0144] In one embodiment, the type recognition module is specifically used to: detect the damaged area of ​​the target vehicle damage photo to obtain the target damaged area; extract features of the target damaged area to obtain target damaged features; and determine the target damage type based on the target damaged features.

[0145] In one embodiment, the type recognition module is specifically used to: perform edge strength recognition on the target vehicle damage photo to obtain the target edge strength feature; perform color difference recognition on the target vehicle damage photo to obtain the target color difference feature; perform texture change recognition on the target vehicle damage photo to obtain the target texture change feature; perform regional division according to the target edge strength feature, the target color difference feature and the target texture change feature to obtain the target damaged area; perform feature extraction on the target damaged area to obtain the target damaged feature; and determine the target damage type according to the target damaged feature.

[0146] In one embodiment, the type recognition module is specifically used to: perform edge strength recognition on the target vehicle damage photo to obtain the target edge strength feature; perform color difference recognition on the target vehicle damage photo to obtain the target color difference feature; perform texture change recognition on the target vehicle damage photo to obtain the target texture change feature; perform attention processing on the target edge strength feature, the target color difference feature and the target texture change feature to obtain the damaged area attention feature; perform region mapping based on the damaged area attention feature to obtain the target damaged area; perform feature extraction on the target damaged area to obtain the target damaged feature; and determine the target damage type based on the target damaged feature.

[0147] In one embodiment, the target vehicle damage photo shooting guidelines include target shooting light guidelines, and the guidelines generation module 803 is specifically used to: extract the visual information of the damaged area of ​​the vehicle from the damaged area information of the vehicle; generate the target shooting light guidelines according to the visual information of the damaged area of ​​the vehicle and the light information of the background area, wherein the target shooting light guidelines are used to guide the shooting of photos with light that meets the predetermined light requirements.

[0148] In one embodiment, the target vehicle damage photo shooting guidance includes a target shooting angle guidance, and the guidance generation module 803 is specifically used to: extract the vehicle damaged area position information from the vehicle damaged area information; extract the clutter level information of at least two background sub-areas from the background area clutter level information; wherein each background sub-area constitutes the background area of ​​the initial vehicle damage photo; select the background sub-area with the lowest clutter level from the clutter level information of each background sub-area to obtain the target background sub-area; generate the target shooting angle guidance according to the vehicle damaged area position information and the position information corresponding to the target background sub-area, wherein the target shooting angle guidance is used to guide the shooting of photos in which the clutter level of the background area meets the predetermined clutter level requirement.

[0149] In one embodiment, the target vehicle damage photo shooting guide includes a target shooting distance guide, and the guide generation module 803 is specifically used to: extract damaged area size information from the damaged area information of the vehicle; generate a target shooting distance guide based on the damaged area size information and the background area size information; wherein the target shooting distance guide is used to guide the taking of photos in which the damaged area and the background area meet a predetermined size ratio.

[0150] It should be noted that the specific implementation of the vehicle damage photo shooting device is basically the same as the specific implementation example of the above-mentioned vehicle damage photo shooting method, and will not be repeated here.

[0151] The embodiment of the present application also provides an electronic device, which includes: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory, and the program is executed by the processor to realize the above-mentioned method for taking photos of vehicle damage. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0152] See also Fig. 9 , Fig. 9 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0153] The processor 901 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an 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 the present application;

[0154] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 902, and the processor 901 calls and executes the vehicle damage photo shooting method of the embodiment of this application;

[0155] Input / output interface 903, used to implement information input and output;

[0156] Communication interface 904, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0157] A bus 905 that transmits information between the various components of the device (e.g., the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);

[0158] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0159] An embodiment of the present application also provides a computer-readable storage medium for computer-readable storage, the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the above-mentioned method for taking vehicle damage photos.

[0160] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0161] The vehicle damage photo shooting method and device, electronic device, and storage medium proposed in the present application first shoot the target damaged vehicle to obtain an initial vehicle damage photo. The quality of the initial vehicle damage photo may be poor, such as poor light, a cluttered background area, or a large background area size. Secondly, the initial vehicle damage photo is region identified to obtain vehicle damage area information and background area information. The vehicle damage area information represents the information of the damaged area of ​​the target damaged vehicle. The background area information represents the information of the environment where the target damaged vehicle is located. The background area information includes at least one of the following: background area light information, background area clutter degree information, and background area size information. Then, according to the vehicle damage area information and the background area information, a target vehicle damage photo shooting guide is generated. The target vehicle damage photo shooting guide can guide the shooting of photos with better quality. Then, according to the target vehicle damage photo shooting guide, the target damaged vehicle is shot to obtain a candidate vehicle damage photo, which has good performance at least in terms of light, cluttered background area, or background area size. Finally, the candidate vehicle damage photo is image enhanced to obtain a target vehicle damage photo, so that the target vehicle damage photo can highlight the damaged area of ​​the target damaged vehicle. In summary, the embodiments of the present application can improve the quality of vehicle damage photos and reduce the reliance on the photographer's knowledge and experience in photography.

[0162] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0163] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0164] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0165] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0166] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0167] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0168] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0169] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0170] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0171] If the integrated unit is implemented in the form of 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 the present application is essentially 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, and the computer software product is stored in a storage medium, including multiple instructions to enable an electronic device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0172] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for taking photos of vehicle damage, characterized in that: The method comprises: Take photos of the target damaged vehicle to obtain initial vehicle damage photos; Performing region recognition on the initial vehicle damage photo to obtain vehicle damaged area information and background area information; wherein the background area information includes at least one of the following: background area light information, background area clutter information, and background area size information; Generate a target vehicle damage photo shooting guide based on the vehicle damaged area information and the background area information; Photographing the target damaged vehicle according to the target vehicle damage photo shooting guide to obtain candidate vehicle damage photos; Perform image enhancement on the candidate vehicle damage photo to obtain a target vehicle damage photo.

2. The method according to claim 1, characterized in that After performing image enhancement on the candidate vehicle damage photo to obtain the target vehicle damage photo, the method further includes: Perform damage type recognition on the target vehicle damage photo to obtain a target damage type; Obtaining the claim number of the target damaged vehicle; A claim is made for the target damaged vehicle according to the basic claim information corresponding to the claim number and the target damage type.

3. The method according to claim 2, characterized in that The step of identifying the damage type of the target vehicle damage photo to obtain the target damage type includes: Performing damaged area detection on the target vehicle damage photo to obtain a target damaged area; Extracting features of the target damaged area to obtain target damaged features; The target damage type is determined according to the target damage characteristics.

4. The method according to claim 3, characterized in that The detecting of the damaged area of ​​the target vehicle damage photo to obtain the target damaged area includes: Performing edge strength recognition on the target vehicle damage photo to obtain target edge strength features; Performing color difference recognition on the target vehicle damage photo to obtain target color difference features; Performing texture change recognition on the target vehicle damage photo to obtain target texture change features; The target damaged area is obtained by performing region division according to the target edge intensity feature, the target color difference feature and the target texture change feature.

5. The method according to any one of claims 1 to 4, characterized in that: The target vehicle damage photo shooting guidance includes target shooting light guidance; The generating of a target vehicle damage photo shooting guide according to the vehicle damaged area information and the background area information includes: Extracting the visual information of the damaged area of ​​the vehicle from the damaged area information of the vehicle; The target shooting light guide is generated according to the visual information of the damaged area of ​​the vehicle and the light information of the background area, wherein the target shooting light guide is used to guide the shooting of photos with light meeting predetermined light requirements.

6. The method according to any one of claims 1 to 4, characterized in that: The target vehicle damage photo shooting guidance includes target shooting angle guidance; The generating of a target vehicle damage photo shooting guide according to the vehicle damaged area information and the background area information includes: Extracting vehicle damaged area location information from the vehicle damaged area information; Extracting the clutter level information of at least two background sub-areas from the clutter level information of the background area; wherein each of the background sub-areas constitutes the background area of ​​the initial vehicle damage photo; Selecting the background sub-region with the lowest clutter level from the clutter level information of each background sub-region to obtain a target background sub-region; The target shooting angle guidance is generated according to the position information of the damaged area of ​​the vehicle and the position information corresponding to the target background sub-area, wherein the target shooting angle guidance is used to guide the taking of photos whose clutter level of the background area meets the predetermined clutter level requirement.

7. The method according to any one of claims 1 to 4, characterized in that: The target vehicle damage photo shooting guide includes a target shooting distance guide; The generating of a target vehicle damage photo shooting guide according to the vehicle damaged area information and the background area information includes: Extracting damaged area size information from the damaged area information of the vehicle; The target shooting distance guide is generated according to the damaged area size information and the background area size information; wherein the target shooting distance guide is used to guide the taking of photos in which the damaged area and the background area meet a predetermined size ratio.

8. A vehicle damage photo shooting device, characterized in that: The device comprises: The first shooting module is used to shoot the target damaged vehicle to obtain an initial vehicle damage photo; A region recognition module is used to perform region recognition on the initial vehicle damage photo to obtain vehicle damaged region information and background region information; wherein the background region information includes at least one of the following: background region light information, background region clutter information, and background region size information; A guide generation module, used for generating a guide for taking a target vehicle damage photo according to the vehicle damaged area information and the background area information; A second shooting module is used to shoot the target damaged vehicle according to the target vehicle damage photo shooting guide to obtain candidate vehicle damage photos; The image enhancement module is used to perform image enhancement on the candidate vehicle damage photo to obtain a target vehicle damage photo.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images