Analysis Method, Device, Equipment and Medium for Vehicle Damage Images Based on Image Detection

Abstract

The present invention relates to the field of artificial intelligence, and discloses an analysis method for vehicle damage images based on image detection, including: obtaining a set of vehicle damage images to be analyzed and analysis method information corresponding to the set of vehicle damage images, analyzing each vehicle damage image in the set of vehicle damage images according to the analysis method corresponding to the analysis method information, and obtaining a score value corresponding to each vehicle damage image, wherein the analysis method corresponding to the analysis method information at least includes a spare part integrity analysis method for determining the spare part integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image, and finally outputting an analysis result corresponding to the set of vehicle damage images based on the score value corresponding to each vehicle damage image. It can be seen that the present invention can realize the analysis of the integrity of spare parts in vehicle damage images based on the contour pixels and edge pixels in the vehicle damage images, and improve the analysis accuracy of the analysis method for vehicle damage images.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and particularly to an analysis method, device, computer device, and storage medium for vehicle damage images based on image detection. Background Art

[0002] Vehicle damage assessment refers to the process of evaluating the damage suffered by the involved vehicles in a traffic accident after the accident occurs, in order to estimate the cost required to repair these damages. In the actual vehicle damage assessment process, it is usually necessary for the damage assessors to take hundreds of photos of the accident scene at the accident site (such as the traffic environment at the accident scene, the damaged parts of the involved vehicles, the overall vehicle condition of the involved vehicles, etc.), and then select the photos that are of reference significance for vehicle damage assessment from the hundreds of photos taken, and finally evaluate the cost required to repair these damages by observing these photos.

[0003] Currently, when screening out the images that are of reference significance for vehicle damage assessment, it has been possible to use computer technology to automatically analyze vehicle damage images to assist users in screening. However, when using computer technology to automatically analyze vehicle damage images, specifically how to use a computer to analyze vehicle damage images to ensure the accuracy of the analysis results is a technical problem that urgently needs to be solved. It can be seen that there is still room for further improvement in the analysis accuracy of the current vehicle damage image analysis method. Summary of the Invention

[0004] The technical problem to be solved by the present invention is that the analysis accuracy of the current vehicle damage image analysis method is relatively low.

[0005] To solve the above technical problem, a first aspect of the present invention discloses an analysis method for vehicle damage images based on image detection, and the method includes:

[0006] Obtain a set of vehicle damage images to be analyzed, where the set of vehicle damage images includes multiple vehicle damage images;

[0007] Obtain the analysis method information corresponding to the set of vehicle damage images;

[0008] Analyze each vehicle damage image in the set of vehicle damage images according to the first analysis method corresponding to the analysis method information to obtain the contour pixels and edge pixels in the vehicle damage image;

[0009] Determine the component integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image;

[0010] Determine the score value corresponding to the vehicle damage image based on the component integrity value corresponding to each vehicle damage image;

[0011] Output the vehicle damage image analysis result corresponding to the set of vehicle damage images based on the score value corresponding to each vehicle damage image to screen out the target vehicle damage image.

[0012] The second aspect of the present invention discloses an analysis device for vehicle damage images based on image detection, and the device includes:

[0013] An acquisition module, configured to acquire a set of vehicle damage images to be analyzed, where the set of vehicle damage images includes a plurality of vehicle damage images;

[0014] The acquisition module is further configured to acquire the analysis method information corresponding to the set of vehicle damage images;

[0015] An analysis module, configured to analyze each vehicle damage image in the set of vehicle damage images according to a first analysis method corresponding to the analysis method information, so as to obtain the contour pixels and edge pixels in the vehicle damage image;

[0016] A determination module, configured to determine the accessory integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image;

[0017] The determination module is further configured to determine the score value corresponding to the vehicle damage image based on the accessory integrity value corresponding to each vehicle damage image;

[0018] An output module, configured to output the vehicle damage image analysis result corresponding to the set of vehicle damage images based on the score value corresponding to each vehicle damage image to screen out the target vehicle damage image. The third aspect of the present invention discloses a computer device, and the computer device includes:

[0019] A memory storing executable program code;

[0020] A processor connected to the memory;

[0021] The processor calls the executable program code stored in the memory and executes some or all of the steps in the analysis method of vehicle damage images based on image detection disclosed in the first aspect of the present invention.

[0022] The fourth aspect of the present invention discloses a computer storage medium, and the computer storage medium stores computer instructions, which are used to execute some or all of the steps in the analysis method of vehicle damage images based on image detection disclosed in the first aspect of the present invention when called.

[0023] In an embodiment of the present invention, first, a set of vehicle damage images to be analyzed and information on the corresponding analysis method for the set of vehicle damage images are obtained. Then, each vehicle damage image in the set of vehicle damage images is analyzed according to the analysis method corresponding to the analysis method information to obtain a score value corresponding to each vehicle damage image. Among them, the analysis method corresponding to the analysis method information at least includes a parts integrity analysis method for determining the parts integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image. Finally, an analysis result corresponding to the set of vehicle damage images is output based on the score value corresponding to each vehicle damage image to assist the user in screening the target vehicle damage image, so as to be able to analyze the integrity of the parts in the vehicle damage image based on the contour pixels and edge pixels in the vehicle damage image, and realize the analysis of the vehicle damage image from the dimension of the integrity of the parts in the vehicle damage image, thereby making the analysis result of the vehicle damage image more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0025] Figure 1 is a schematic flowchart of an analysis method for vehicle damage images based on image detection disclosed in an embodiment of the present invention;

[0026] Figure 2 is a schematic structural diagram of an analysis device for vehicle damage images based on image detection disclosed in an embodiment of the present invention;

[0027] Figure 3 is a schematic structural diagram of a computer device disclosed in an embodiment of the present invention;

[0028] Figure 4 is a schematic structural diagram of a computer storage medium disclosed in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0030] In the description, claims and the above drawings of the present invention, terms such as "first", "second", etc. are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product or terminal that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include unlisted steps or units, or may optionally further include other steps or units inherent to these processes, methods, products or terminals.

[0031] Reference to "embodiment" herein means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0032] Embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is a theory, method, technology and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results.

[0033] Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. Artificial intelligence software technologies mainly include several major directions such as computer vision technology, robotics, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0034] The present invention discloses an analysis method, device, computer device, and storage medium for vehicle damage images based on image detection. First, a set of vehicle damage images to be analyzed and the analysis method information corresponding to the set of vehicle damage images are obtained. Then, each vehicle damage image in the set of vehicle damage images is analyzed according to the analysis method corresponding to the analysis method information to obtain a score value corresponding to each vehicle damage image. Among them, the analysis method corresponding to the analysis method information at least includes a parts integrity analysis method for determining the parts integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image. Finally, based on the score value corresponding to each vehicle damage image, the analysis result corresponding to the set of vehicle damage images is output to assist the user in screening the target vehicle damage image, so as to be able to analyze the integrity of the parts in the vehicle damage image based on the contour pixels and edge pixels in the vehicle damage image, and realize the analysis of the vehicle damage image from the dimension of the integrity of the parts in the vehicle damage image, so as to make the analysis result of the vehicle damage image more accurate. The following will be described in detail respectively.

[0035] Embodiment 1

[0036] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of an analysis method for vehicle damage images based on image detection disclosed in an embodiment of the present invention. As Figure 1 shown, the analysis method for vehicle damage images based on image detection may include the following operations:

[0037] 101. Obtain a set of vehicle damage images to be analyzed, where the set of vehicle damage images includes multiple vehicle damage images.

[0038] In the above step 101, the set of vehicle damage images to be analyzed may be pre-stored in the vehicle insurance system. Specifically, the loss adjuster can upload the images taken during a traffic accident as vehicle damage images to the vehicle insurance system. Usually, during a traffic accident, the loss adjuster will arrive at the scene of the traffic accident and take hundreds of images, and then upload them to the vehicle insurance system as the set of vehicle damage images for this traffic accident. Subsequently, when the loss adjuster needs to determine the loss for this traffic accident, the set of vehicle damage images for this traffic accident can be directly extracted from the vehicle insurance system for vehicle loss determination.

[0039] 102. Obtain the analysis method information corresponding to the set of vehicle damage images.

[0040] In the above step 102, the analysis method information corresponding to the set of vehicle damage images is used to record the analysis method to be performed on the set of vehicle damage images. Specifically, the analysis method information corresponding to the set of vehicle damage images can be determined by the following two methods:

[0041] (1) When the loss assessor uploads the vehicle damage image set to the vehicle insurance system, set the analysis method information corresponding to the vehicle damage image set. For example, in a certain traffic accident, if it is necessary to analyze the image clarity of the vehicle damage images of this traffic accident later, when the loss assessor uploads the vehicle damage image set to the vehicle insurance system, the analysis method information corresponding to this traffic accident can be set to require image clarity analysis.

[0042] (2) When the loss assessor uploads the vehicle damage image set to the vehicle insurance system, set the traffic accident type corresponding to the vehicle damage image set, and then query the analysis method information corresponding to this traffic accident type in the preset analysis method information table as the analysis method information corresponding to the vehicle damage image set. Among them, the analysis method information table can pre-store the analysis method information corresponding to different traffic accident types.

[0043] 103. Analyze each vehicle damage image in the vehicle damage image set according to the first analysis method corresponding to the analysis method information to obtain the contour pixels and edge pixels in the vehicle damage image.

[0044] In step 103 above, in a vehicle damage image, the contour refers to a connected and closed curve in the image, while the edge usually refers to an unconnected and unclosed curve in the image. The contour curve and edge curve in the vehicle damage image can be extracted through the opencv tool. The pixels in the extracted contour curve are the contour pixels, and the pixels in the extracted edge curve are the edge pixels. Among them, the analysis methods for vehicle damage images can include the first analysis method, the second analysis method, and the third analysis method, which are respectively used to analyze the accessory integrity value, image size matching value, and image clarity value of the image. Users can set the analysis of vehicle damage images by setting the analysis method information.

[0045] Currently, when using artificial intelligence technology to analyze vehicle damage images, the analysis methods are usually fixed and single. For example, only the clarity of the vehicle damage images is analyzed, or only the image size of the vehicle damage images is analyzed, etc. However, in actual damage assessment, the analysis requirements for the set of vehicle damage images are usually complex and variable. For example, the analysis requirements for damage assessment photos in different types of traffic accidents are usually different. For example, for a traffic accident with the license plate number of the responsible party already determined, the clarity of the damage assessment photos is usually not required to be analyzed, and only the integrity of the parts in the damage assessment photos needs to be analyzed. For a traffic accident without the license plate number of the responsible party determined, the clarity of the damage assessment photos is required to be analyzed (because the license plate number of the responsible party needs to be extracted from the damage assessment photos, so the damage assessment photos need to have sufficient clarity to identify the license plate number). At this time, if a fixed and single analysis method is used for the sets of vehicle damage images of these two types of traffic accidents, it is obviously inappropriate. It can be seen that the current analysis method for vehicle damage images still has the problem that the analysis method is fixed and single and cannot flexibly adapt to the actual application scenarios. In the embodiments of the present invention, by setting different analysis methods for different sets of vehicle damage images and then analyzing different sets of vehicle damage images using different analysis methods (the specific analysis process will be described later), a method for analyzing vehicle damage images with rich analysis methods can be provided, which can more flexibly adapt to diverse actual application scenarios. In addition, in the method for analyzing vehicle damage images in the embodiments of the present invention, an analysis method for determining the integrity of parts in a vehicle damage image based on the contour pixels and edge pixels in the vehicle damage image is also proposed, and the score value of the vehicle damage image is determined based on the integrity value of the parts in the vehicle damage image. In the prior art, from which dimension to analyze vehicle damage images to ensure the accuracy of the analysis results is a technical problem that urgently needs to be solved. In the embodiments of the present invention, by realizing the analysis of the integrity of parts in a vehicle damage image based on the contour pixels and edge pixels in the vehicle damage image, the vehicle damage image is analyzed from the dimension of the integrity of the parts in the vehicle damage image, so that the analysis results of the vehicle damage image can be more accurate. The specific analysis process of the integrity of parts in the vehicle damage image will be described later.

[0046] 104. Determine the integrity value of the parts corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image.

[0047] In step 104 above, when analyzing the vehicle damage image using the first analysis method, the contour pixels and edge pixels extracted from the vehicle damage image can be used to calculate the integrity value of the parts corresponding to the vehicle damage image, and the specific calculation process will be described later.

[0048] 105. Determine the score value of the vehicle damage image based on the integrity value of the parts corresponding to each vehicle damage image.

[0049] In the above step 105, after calculating the accessory integrity value, the accessory integrity value can be used as a factor to calculate the score value corresponding to the vehicle damage image.

[0050] 106. Output the vehicle damage image analysis result corresponding to the vehicle damage image set based on the score value corresponding to each vehicle damage image to screen out the target vehicle damage image.

[0051] In the above step 106, after completing the analysis of the vehicle damage image set, the score value corresponding to each vehicle damage image in the vehicle damage image set can be obtained. At this time, each vehicle damage image and the score value corresponding to the vehicle damage image can be displayed to the user together in the interaction interface, and the score value corresponding to the vehicle damage image can be marked on the vehicle damage image in the interaction interface (that is, output the analysis result corresponding to the vehicle damage image set). In this way, the user can intuitively observe the score of each vehicle damage image, and then can select a suitable target vehicle damage image with reference to the score of the vehicle damage image for subsequent vehicle damage assessment work. Optionally, all vehicle damage images can also be sorted in descending order of the score value and displayed to the user in the interaction interface according to the sorting (that is, output the analysis result corresponding to the vehicle damage image set). In this way, the user can quickly and accurately select a suitable target vehicle damage image according to the score value sorting of the vehicle damage images for subsequent vehicle damage assessment work.

[0052] It can be seen that the implementation Figure 1 The described analysis method of vehicle damage images based on image detection first obtains the vehicle damage image set to be analyzed and the analysis method information corresponding to the vehicle damage image set, then analyzes each vehicle damage image in the vehicle damage image set according to the analysis method corresponding to the analysis method information to obtain the score value corresponding to each vehicle damage image, and finally outputs the analysis result corresponding to the vehicle damage image set based on the score value corresponding to each vehicle damage image to assist the user in screening the target vehicle damage image, so that when the user screens out the photos with reference significance for vehicle damage assessment from the vehicle damage image set, the computer technology can be used to realize the automatic analysis of the vehicle damage images to assist the user in screening the vehicle damage images, and improve the analysis efficiency and analysis quality of the analysis method of the vehicle damage images when screening the vehicle damage images. In addition, by setting different analysis methods for different vehicle damage image sets and then analyzing different vehicle damage image sets with different analysis methods, an analysis method of vehicle damage images with rich analysis methods can be provided, which can more flexibly adapt to diverse actual application scenarios and improve the effect of automatic analysis of vehicle damage images. It also realizes the analysis of the integrity of accessories in the vehicle damage image based on the contour pixels and edge pixels in the vehicle damage image, and realizes the analysis of the vehicle damage image from the dimension of the integrity of accessories in the vehicle damage image, so that the analysis result of the vehicle damage image can be more accurate.

[0053] In an optional embodiment, determining the accessory integrity value corresponding to each vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image includes:

[0054] Detecting whether there are target contour pixels corresponding to the vehicle damage image among the contour pixels of each vehicle damage image in the vehicle damage image set;

[0055] When it is detected that there are target contour pixels corresponding to the vehicle damage image among the contour pixels of the vehicle damage image, setting a preset target accessory integrity value as the accessory integrity value corresponding to the vehicle damage image;

[0056] When it is detected that there are no such target contour pixels among the contour pixels of the vehicle damage image, calculating the ratio of the number of pixels of the edge pixels in the vehicle damage image to the preset number of complete pixels corresponding to the vehicle damage image, as the accessory integrity value corresponding to the vehicle damage image.

[0057] In this optional embodiment, the target contour pixels corresponding to each vehicle damage image can be preset. For example, when a vehicle damage image is an image for recording the damage condition of an automobile wheel, the target contour pixels corresponding to the vehicle damage image can be set as the pixels within a circular ring shape. When it is detected that there are pixels (i.e., target contour pixels) within the same circular ring shape in the vehicle damage image (the specific detection process will be described in detail later), it indicates that there is a complete wheel automotive accessory in the vehicle damage image, and thus the accessory integrity value corresponding to the vehicle damage image can be set to the full value (i.e., the target accessory integrity value), for example, setting the accessory integrity value corresponding to the vehicle damage image to 1. When it is detected that there are no pixels within the same circular ring shape (i.e., target contour pixels) in the vehicle damage image, it indicates that there is no complete wheel automotive accessory in the vehicle damage image. At this time, the edge curve in the vehicle damage image can be extracted, and then the ratio of the number of pixels in the edge curve (i.e., the number of pixels of the edge pixels) to the preset number of complete pixels corresponding to the vehicle damage image is calculated as the accessory integrity value corresponding to the vehicle damage image. For example, if there is only one-third of a wheel in the vehicle damage image, the finally calculated accessory integrity value corresponding to the vehicle damage image can be one-third. Optionally, the accessory integrity value corresponding to the vehicle damage image can be directly used as the scoring value corresponding to the vehicle damage image.

[0058] It can be seen that when implementing this optional embodiment, when there are target contour pixels corresponding to the vehicle damage image in the vehicle damage image, the preset target part integrity value is set to the part integrity value corresponding to the vehicle damage image. When there are no target contour pixels corresponding to the vehicle damage image in the vehicle damage image, the ratio of the number of edge pixels in the vehicle damage image to the number of complete pixels corresponding to the preset vehicle damage image is calculated as the part integrity value corresponding to the vehicle damage image. Thus, by detecting the contour pixels or edge pixels in the vehicle damage image and then calculating the part integrity value corresponding to the vehicle damage image, the analysis of the integrity of the vehicle parts in the vehicle damage image can be realized.

[0059] In an optional embodiment, detecting whether there are target contour pixels corresponding to the vehicle damage image among the contour pixels of each vehicle damage image in the vehicle damage image set includes:

[0060] Calculating the ratio of the number of contour pixels of the vehicle damage image to the number of target contour pixels corresponding to the preset vehicle damage image;

[0061] Judging whether the ratio is within a preset ratio range;

[0062] When it is judged that the ratio is within the ratio range, it is determined that there are target contour pixels corresponding to the vehicle damage image in the vehicle damage image;

[0063] When it is judged that the ratio is not within the ratio range, it is determined that there are no target contour pixels corresponding to the vehicle damage image in the vehicle damage image.

[0064] In this alternative embodiment, in the vehicle damage image, the outlines of different types of accessories are usually different. For example, the outline of a wheel is a circular ring, and the outline of a rearview mirror is similar to a rectangle. Since the outlines of the accessories are different, the number of pixels occupied by the outline of a complete accessory in the vehicle damage image is also different. Therefore, different numbers of pixels of the target outline pixels can be preset for the vehicle damage images of different types of accessories, and then by determining whether the number of pixels of the outline pixels of the vehicle damage image matches the number of pixels of the target outline pixels corresponding to the preset vehicle damage image, it can be determined whether there are target outline pixels corresponding to the vehicle damage image in the vehicle damage image (that is, whether there is a complete accessory in the vehicle damage image). For example, for the vehicle damage image of a wheel, the number of pixels of the target outline pixels corresponding to the vehicle damage image can be preset to 1000, and the ratio range can be set to 0.9 - 1.1. If there is a wheel in the vehicle damage image, the number of pixels of the extracted outline pixels will also be approximately distributed around 1000. For example, if the number of pixels of the extracted outline pixels is 960, at this time, the ratio of the number of pixels of the extracted outline pixels to the number of pixels of the target outline pixels corresponding to the vehicle damage image is calculated to be 0.96, which is within the ratio range, so it can be determined that there are target outline pixels corresponding to the vehicle damage image in the vehicle damage image (that is, there is a complete wheel). If only one-third of a wheel exists in the vehicle damage image, the number of pixels of the extracted outline pixels may be 300. At this time, the calculated ratio is 0.3, which is not within the ratio range, so it can be determined that there are no target outline pixels corresponding to the vehicle damage image in the vehicle damage image (that is, there is no complete wheel).

[0065] It can be seen that by implementing this alternative embodiment, by extracting the outline pixels of the vehicle damage image, then calculating the ratio of the number of pixels of the outline pixels of the vehicle damage image to the number of pixels of the target outline pixels corresponding to the vehicle damage image, and finally determining whether there are target outline pixels in the vehicle damage image according to whether the ratio is within the ratio range, the detection of the target outline pixels in the vehicle damage image is realized, providing a basis for analyzing the integrity of the vehicle accessories in the vehicle damage image.

[0066] In an alternative embodiment, before determining the score value corresponding to the vehicle damage image based on the integrity value of the accessory corresponding to each vehicle damage image, the method further includes:

[0067] According to the second analysis method corresponding to the analysis method information, calculate the image size matching value of each vehicle damage image in the vehicle damage image set through the following formula:

[0068]

[0069] Wherein, a is the image size matching value of the vehicle damage image, b is the width pixel value of the vehicle damage image, c is the preset standard width pixel value, d is the length pixel value of the vehicle damage image, and e is the preset standard length pixel value;

[0070] And, determining the score value corresponding to the vehicle damage image based on the integrity value of the accessories corresponding to each vehicle damage image includes:

[0071] Determining the score value corresponding to the vehicle damage image based on the integrity value of the accessories corresponding to each vehicle damage image and the image size matching value.

[0072] In this optional embodiment, when analyzing the vehicle damage image, it is also possible to analyze whether the image size of the vehicle damage image conforms to the preset image size, that is, the second analysis method. For example, the width pixel value of the vehicle damage image is 90, the length pixel value is 110, and the preset standard width pixel value and standard length pixel value are both 100. Then the calculated image size matching value is:

[0073]

[0074] The image size matching value calculated by the above formula can represent the matching degree between the image size of the vehicle damage image and the preset image size. Optionally, the weighted sum of the integrity value of the accessories corresponding to the vehicle damage image and the image size matching value can be used as the score value corresponding to the vehicle damage image.

[0075] It can be seen that when implementing this optional embodiment, when analyzing the vehicle damage image, the image size matching value that can represent the matching degree between the image size of the vehicle damage image and the preset image size is also calculated according to the preset formula, so as to realize the analysis of the image size of the vehicle damage image, increase the analysis dimension of the analysis method of the vehicle damage image, and improve the analysis effect.

[0076] In an optional embodiment, before determining the score value corresponding to the vehicle damage image based on the integrity value of the accessories corresponding to each vehicle damage image and the image size matching value, the method further includes:

[0077] According to the third analysis method corresponding to the analysis method information, analyzing each vehicle damage image in the vehicle damage image set based on a preset image clarity analysis model to obtain the image clarity value corresponding to the vehicle damage image;

[0078] And, determining the score value corresponding to the vehicle damage image based on the integrity value of the accessories corresponding to each vehicle damage image and the image size matching value includes:

[0079] Determine the score value corresponding to each vehicle damage image based on the component integrity value, image size matching value, and image clarity value corresponding to each vehicle damage image.

[0080] In this optional embodiment, the image clarity analysis model can be the RankIQA model. By performing image processing transformation on the determined clarity pictures, different levels and types of sorted distorted images can be generated, thus obtaining a large dataset, and then a wider and deeper network can be selected for training. Specifically, the Siamese network can be first selected to learn the representation features of the generated data sorting relationship, and then the knowledge represented in the trained Siamese network can be transferred to the traditional CNN, so as to predict the image clarity value of a single image.

[0081] It can be seen that when implementing this optional embodiment, when analyzing the vehicle damage image, the image clarity value corresponding to the vehicle damage image is also analyzed through a preset image clarity analysis model, so as to realize the analysis of the image clarity of the vehicle damage image, further increase the analysis dimension of the analysis method of the vehicle damage image, and improve the analysis effect.

[0082] In an optional embodiment, the determining the score value corresponding to each vehicle damage image based on the component integrity value, image size matching value, and image clarity value corresponding to each vehicle damage image includes:

[0083] Calculate the weighted sum of the component integrity value, image size matching value, and image clarity value corresponding to the vehicle damage image based on a preset weight as the original score value corresponding to the vehicle damage image;

[0084] Based on the following formula, calculate the score value corresponding to the vehicle damage image based on the original score value corresponding to the vehicle damage image:

[0085]

[0086] Where f is the score value corresponding to the vehicle damage image, g is the original score value corresponding to the vehicle damage image, max is the largest original score value among all the original score values corresponding to the vehicle damage images, and min is the smallest original score value among all the original score values corresponding to the vehicle damage images.

[0087] In this alternative embodiment, when using the weighted sum of the accessory integrity value, image size matching value, and image clarity value corresponding to the damaged image as the original score value corresponding to the damaged vehicle image, if the original score value is directly used as the score value of the damaged vehicle image, since the distribution of the values of the original score value is usually relatively scattered (for example, one original score value may be 10, while another original score value may be 1000), it is not conducive to the user observing the analysis results, and thus not conducive to the user screening the damaged vehicle images according to the analysis results. Through the above formula, the original score value can be mapped to the interval [0, 1] as the score value, and at the same time, the mapped score value still retains the function of the original score value to represent the score level of the damaged vehicle image, so that the finally obtained score value is more convenient for the user to directly observe and is more conducive to the user screening the damaged vehicle images according to the analysis results.

[0088] It can be seen that by implementing this alternative embodiment, after using the weighted sum of the accessory integrity value, image size matching value, and image clarity value corresponding to the damaged image as the original score value corresponding to the damaged vehicle image, the original score value is mapped to a specified numerical interval through a preset formula as the score value, so that the finally obtained score value is more convenient for the user to directly observe and is more conducive to the user screening the damaged vehicle images according to the analysis results.

[0089] In an alternative embodiment, before obtaining the set of damaged vehicle images to be analyzed, the method further includes:

[0090] Obtain the original damaged vehicle images uploaded by the user;

[0091] Analyze the original damaged vehicle images according to all the analysis methods corresponding to the analysis method information, and obtain the score value corresponding to the original damaged vehicle images;

[0092] Judge whether the score value corresponding to the original damaged vehicle image is greater than a preset score value threshold;

[0093] When it is judged that the score value corresponding to the original damaged vehicle image is greater than the score value threshold, add the original damaged vehicle image to the set of damaged vehicle images;

[0094] When it is judged that the score value corresponding to the original damaged vehicle image is not greater than the score value threshold, output a warning prompt to the user, where the warning prompt is used to prompt the user to re-upload the original damaged vehicle image.

[0095] In this optional embodiment, the vehicle damage images in the vehicle damage image set may be uploaded by the loss adjusters. Specifically, the loss adjusters can use a mobile terminal to capture vehicle damage images (i.e., original vehicle damage images) at the scene of a traffic accident and then upload them to the vehicle insurance system. After the user uploads the original vehicle damage images, the original vehicle damage images can be analyzed once to obtain the score values of the original vehicle damage images. If the score value of an original vehicle damage image is higher than the preset score value threshold, the original vehicle damage image is added to the vehicle damage image set. If the score value of the original vehicle damage image is not higher than the preset score value threshold, it indicates that the image quality of the original vehicle damage image is unqualified, and a warning prompt is output to the user to prompt the user to re-upload the original vehicle damage image. In this way, the vehicle damage images uploaded to the vehicle damage image set can be screened according to the score values, which is beneficial to improving the image quality of the vehicle damage image set.

[0096] It can be seen that by implementing this optional embodiment, after the user uploads the original vehicle damage images, the uploaded original vehicle damage images are analyzed to obtain the scores of the original vehicle damage images. If the score of an original vehicle damage image is greater than the score value threshold, the original vehicle damage image is added to the vehicle damage image set. If the score of the original vehicle damage image is not greater than the score value threshold, the user is prompted to re-upload the original vehicle damage images, so that the vehicle damage images uploaded to the vehicle damage image set can be screened according to the score values, which is beneficial to improving the image quality of the vehicle damage image set.

[0097] Optionally, it can also: upload the analysis information of the vehicle damage image based on image detection of the analysis method of the vehicle damage image based on image detection to the blockchain.

[0098] Specifically, the analysis information of the vehicle damage image based on image detection is obtained by running the analysis method of the vehicle damage image based on image detection and is used to record the analysis situation of the vehicle damage image based on image detection. For example, the obtained vehicle damage image set, analysis method information, score value of the vehicle damage image, and so on. Uploading the analysis information of the vehicle damage image based on image detection to the blockchain can ensure its security and fairness and transparency to users. Users can download the analysis information of the vehicle damage image based on image detection from the blockchain to verify whether the analysis information of the vehicle damage image based on image detection of the analysis method of the vehicle damage image based on image detection has been tampered with. The blockchain referred to in this example is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithm. Blockchain, in essence, is a decentralized database, a string of data blocks generated by using cryptographic methods. Each data block contains information about a batch of network transactions, which is used to verify the validity of the information (anti-counterfeiting) and generate the next block. The blockchain can include the blockchain underlying platform, the platform product service layer, and the application service layer, etc.

[0099] Example Two

[0100] Please refer to Figure 2 , Figure 2 which is a schematic structural diagram of an analysis device for vehicle damage images based on image detection disclosed in an embodiment of the present invention. As Figure 2 shown, the analysis device for vehicle damage images based on image detection may include:

[0101] An acquisition module 201, configured to acquire a set of vehicle damage images to be analyzed, where the set of vehicle damage images includes multiple vehicle damage images;

[0102] The acquisition module 201 is further configured to acquire analysis method information corresponding to the set of vehicle damage images;

[0103] An analysis module 202, configured to analyze each vehicle damage image in the set of vehicle damage images according to a first analysis method corresponding to the analysis method information, so as to obtain contour pixels and edge pixels in the vehicle damage image;

[0104] A determination module 203, configured to determine a parts integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image;

[0105] The determination module 203 is further configured to determine a score value corresponding to the vehicle damage image based on the parts integrity value corresponding to each vehicle damage image;

[0106] An output module 204, configured to output an analysis result of the set of vehicle damage images corresponding to the set of vehicle damage images based on the score value corresponding to each vehicle damage image to screen out target vehicle damage images.

[0107] For the specific description of the above analysis device for vehicle damage images based on image detection, reference may be made to the specific description of the above analysis method for vehicle damage images based on image detection. To avoid repetition, it will not be elaborated here one by one.

[0108] Example Three

[0109] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of a computer device disclosed in an embodiment of the present invention. As Figure 3 shown, the computer device may include:

[0110] A memory 301 storing executable program code;

[0111] A processor 302 connected to the memory 301;

[0112] The processor 302 calls the executable program code stored in the memory 301 and executes the steps in the method for analyzing vehicle damage images based on image detection disclosed in Embodiment 1 of the present invention.

[0113] Embodiment 4

[0114] Please refer to Figure 4 , the present invention discloses a computer storage medium 401. The computer storage medium 401 stores computer instructions, which when called, are used to execute the steps in the method for analyzing vehicle damage images based on image detection disclosed in Embodiment 1 of the present invention.

[0115] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0116] Through the above specific descriptions of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solutions, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, and the storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc memories, magnetic disk memories, tape memories, or any other computer-readable medium capable of carrying or storing data.

[0117] Finally, it should be noted that: What is disclosed by an analysis method, device, computer device, and storage medium for vehicle damage images based on image detection in the embodiments of the present invention is only the preferred embodiments of the present invention, and is only used to illustrate the technical solutions of the present invention, rather than limiting it; Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for analyzing vehicle damage images based on image detection, characterized in that, the method includes: Obtain a set of vehicle damage images to be analyzed, where the set of vehicle damage images includes multiple vehicle damage images; Obtain the analysis method information corresponding to the set of vehicle damage images, and the analysis method information includes at least a first analysis method and a second analysis method; Analyze each vehicle damage image in the set of vehicle damage images according to the first analysis method corresponding to the analysis method information to obtain the contour pixels and edge pixels in the vehicle damage image; Determine the accessory integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image; According to the second analysis method corresponding to the analysis method information, calculate the image size matching value of each vehicle damage image in the set of vehicle damage images through the following formula: where a is the image size matching value of the vehicle damage image, b is the width pixel value of the vehicle damage image, c is the preset standard width pixel value, d is the length pixel value of the vehicle damage image, and e is the preset standard length pixel value; Determine the score value corresponding to the vehicle damage image based on the accessory integrity value and the image size matching value corresponding to each vehicle damage image; Output the vehicle damage image analysis result corresponding to the set of vehicle damage images based on the score value corresponding to each vehicle damage image to screen out the target vehicle damage image.

2. The method for analyzing vehicle damage images based on image detection according to claim 1, characterized in that, the determining the accessory integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image includes: Detect whether there are target contour pixels corresponding to the vehicle damage image in the contour pixels of each vehicle damage image in the set of vehicle damage images; When it is detected that there are target contour pixels corresponding to the vehicle damage image in the contour pixels of the vehicle damage image, set the preset target accessory integrity value as the accessory integrity value corresponding to the vehicle damage image; When it is detected that there are no such target contour pixels in the contour pixels of the vehicle damage image, calculate the ratio of the number of edge pixels in the vehicle damage image to the preset number of complete pixels corresponding to the vehicle damage image as the accessory integrity value corresponding to the vehicle damage image.

3. The method for analyzing vehicle damage images based on image detection according to claim 2, characterized in that, the detecting whether there are target contour pixels corresponding to the vehicle damage image in the contour pixels of each vehicle damage image in the set of vehicle damage images includes: Calculate the ratio of the number of contour pixels of the vehicle damage image to the number of target contour pixels corresponding to the vehicle damage image preset; Judge whether the ratio is within a preset ratio range; When it is judged that the ratio is within the ratio range, determine that there are target contour pixels corresponding to the vehicle damage image in the vehicle damage image; When it is judged that the ratio is not within the ratio range, determine that there are no target contour pixels corresponding to the vehicle damage image in the vehicle damage image.

4. The method for analyzing vehicle damage images based on image detection according to claim 1, characterized in that, Before determining the score value corresponding to each vehicle damage image based on the component integrity value and the image size matching value corresponding to the vehicle damage image, the method further includes: Analyzing each vehicle damage image in the vehicle damage image set based on a preset image clarity analysis model according to the third analysis method corresponding to the analysis method information, to obtain the image clarity value corresponding to the vehicle damage image; And, determining the score value corresponding to the vehicle damage image based on the component integrity value and the image size matching value corresponding to each vehicle damage image includes: Determining the score value corresponding to the vehicle damage image based on the component integrity value, the image size matching value, and the image clarity value corresponding to each vehicle damage image.

5. The analysis method of vehicle damage images based on image detection according to claim 4, wherein, Determining the score value corresponding to the vehicle damage image based on the component integrity value, the image size matching value, and the image clarity value corresponding to each vehicle damage image includes: Calculating the weighted sum of the component integrity value, the image size matching value, and the image clarity value corresponding to the vehicle damage image based on a preset weight, to be used as the original score value corresponding to the vehicle damage image; Calculating the score value corresponding to the vehicle damage image based on the original score value corresponding to the vehicle damage image through the following formula: where f is the score value corresponding to the vehicle damage image, g is the original score value corresponding to the vehicle damage image, max is the largest original score value among all the original score values corresponding to the vehicle damage images, and min is the smallest original score value among all the original score values corresponding to the vehicle damage images.

6. The analysis method of vehicle damage images based on image detection according to claim 1, wherein, Before obtaining the vehicle damage image set to be analyzed, the method further includes: Obtaining the original vehicle damage image uploaded by the user; Analyzing the original vehicle damage image according to all the analysis methods corresponding to the analysis method information, to obtain the score value corresponding to the original vehicle damage image; Judging whether the score value corresponding to the original vehicle damage image is greater than a preset score value threshold; When it is judged that the score value corresponding to the original vehicle damage image is greater than the score value threshold, adding the original vehicle damage image to the vehicle damage image set; When it is judged that the score value corresponding to the original vehicle damage image is not greater than the score value threshold, outputting a warning prompt to the user, where the warning prompt is used to prompt the user to re-upload the original vehicle damage image.

7. An analysis device for vehicle damage images based on image detection, the device is used to implement the steps of the analysis method of vehicle damage images based on image detection according to any one of claims 1-6, wherein, The device includes: An acquisition module, configured to acquire a vehicle damage image set to be analyzed, where the vehicle damage image set includes multiple vehicle damage images; The acquisition module is further configured to acquire the analysis method information corresponding to the vehicle damage image set; An analysis module, configured to analyze each vehicle damage image in the vehicle damage image set according to a first analysis method corresponding to the analysis method information, so as to obtain contour pixels and edge pixels in the vehicle damage image; A determination module, configured to determine a parts integrity value corresponding to the vehicle damage image based on the contour pixels and edge pixels in each vehicle damage image; The determination module is further configured to determine a score value corresponding to the vehicle damage image; An output module, configured to output an analysis result of the vehicle damage image set corresponding to the vehicle damage image based on the score value corresponding to each vehicle damage image, so as to screen out a target vehicle damage image.

8. A computer device, characterized in that the computer device includes: a memory storing executable program code; a processor connected to the memory; The processor calls the executable program code stored in the memory and executes the analysis method of the vehicle damage image based on image detection according to any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, it implements the analysis method of the vehicle damage image based on image detection according to any one of claims 1-6.

Images