Insurance service loss assessment method and device, server and storage medium

Abstract

The invention relates to the field of financial science and technology, and discloses an insurance service loss assessment method and device, a server and a storage medium, and the method comprises the steps: receiving a service loss assessment image corresponding to a target insurance service to be subjected to loss assessment, and determining the object attribute of a target object in each service loss assessment image based on the image features of the service loss assessment image; obtaining an image tag corresponding to each business loss assessment image and first loss assessment information according to the object attribute; annotating the service loss assessment image according to the image label to obtain a service annotation image, and preprocessing the service annotation image to obtain a preprocessed image; obtaining an image compression strategy of the preprocessed image at least according to label information corresponding to the image label so as to compress the preprocessed image; and at least transmitting the target compressed image obtained by compression to the service terminal, receiving second loss assessment information obtained by the service terminal, and obtaining insurance loss assessment information based on the first loss assessment information and the second loss assessment information.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and in particular to a method, device, server and storage medium for loss assessment of insurance business. Background Art

[0002] As the scale of agricultural insurance expands, agricultural insurance operations are facing a paradox of efficiency and risk management, which is particularly evident in information asymmetry and low claims efficiency, which seriously restricts the performance of agricultural insurance's economic compensation, financing and social management functions.

[0003] At present, when it comes to claims caused by natural disasters, traditional agricultural insurance often uses random field sampling to conduct inspections and loss assessments during the crop growth period and after a natural disaster occurs. However, if the insured object is affected by a large area due to natural disasters in certain regions, the traditional inspection and loss assessment model often has problems such as information asymmetry, low claims efficiency, and many claims disputes. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to provide a method, device, server and storage medium for loss assessment of insurance business, aiming to realize claim assessment of insurance business efficiently and accurately.

[0005] In a first aspect, the present application provides a method for loss assessment of insurance business, the method comprising:

[0006] Receiving a business damage assessment image corresponding to a target insurance business to be assessed, and determining object attributes of a target object in each of the business damage assessment images based on image features of the business damage assessment images, wherein the object attributes include an object type of the target object and a degree of damage to the target object;

[0007] Acquire the image label and the first damage assessment information corresponding to each of the business damage assessment images according to the object type and the damage degree;

[0008] Annotating the business damage assessment image according to the image label to obtain a business annotated image, and preprocessing the business annotated image to obtain a preprocessed image, wherein the preprocessed image has label information corresponding to the image label;

[0009] At least acquiring an image compression strategy of the preprocessed image according to the label information, and compressing the preprocessed image according to the image compression strategy to obtain a target compressed image;

[0010] At least transmitting the target compressed image to a service terminal so that the service terminal displays the target compressed image on a preset interface;

[0011] The second damage assessment information acquired by the service terminal is received, and the insurance damage assessment information of the target insurance business is obtained based on the first damage assessment information and the second damage assessment information.

[0012] In a second aspect, the present application provides a loss assessment device for insurance business, comprising:

[0013] An image receiving module, used to receive a business damage assessment image corresponding to a target insurance business to be assessed, and determine object attributes of a target object in each of the business damage assessment images based on image features of the business damage assessment images, wherein the object attributes include an object type of the target object and a degree of damage to the target object;

[0014] An image analysis module, used for obtaining image labels and first damage assessment information corresponding to each of the business damage assessment images according to the object type and the damage degree;

[0015] an image processing module, which annotates the business damage assessment image according to the image label to obtain a business annotated image, and preprocesses the business annotated image to obtain a preprocessed image, wherein the preprocessed image has label information corresponding to the image label;

[0016] An image compression module, configured to obtain an image compression strategy for the preprocessed image at least according to the tag information, and compress the preprocessed image according to the image compression strategy to obtain a target compressed image;

[0017] An image transmission module, used for transmitting at least the target compressed image to a service terminal, so that the service terminal displays the target compressed image on a preset interface;

[0018] The damage assessment module receives the second damage assessment information obtained by the service terminal, and obtains the insurance damage assessment information of the target insurance business based on the first damage assessment information and the second damage assessment information.

[0019] In a third aspect, the present application provides a server, including a processor and a memory;

[0020] The memory is used to store computer programs;

[0021] The processor is used to execute the computer program and implement the aforementioned loss assessment method for insurance business when executing the computer program.

[0022] In a fourth aspect, the present application provides a computer-readable storage medium, which, when executed by one or more processors, enables the one or more processors to execute the steps of the aforementioned method for loss assessment of insurance business.

[0023] The embodiment of the present application provides a method, device, equipment and storage medium for assessing the loss of insurance business. The method receives the business loss assessment image corresponding to the target insurance business to be assessed, and determines the object attributes of the target object in each business loss assessment image based on the image features of the business loss assessment image, the object attributes including the object type of the target object and the degree of damage of the target object; obtains the image label and the first loss assessment information corresponding to each business loss assessment image according to the object type and the degree of damage; annotates the business loss assessment image according to the image label to obtain the business annotated image, and preprocesses the business annotated image to obtain the preprocessed image; obtains the image compression strategy of the preprocessed image according to at least the label information corresponding to the image label of the preprocessed image, and compresses the preprocessed image according to the image compression strategy to obtain the target compressed image; transmits at least the target compressed image to the service terminal, so that the service terminal displays the target compressed image in a preset interface; receives the second loss assessment information obtained by the service terminal, and obtains the insurance loss assessment information of the target insurance business based on the first loss assessment information and the second loss assessment information.

[0024] Image analysis is used to obtain the first damage assessment information corresponding to the target object in the business damage assessment image and the second damage assessment information obtained by the auditor through the service terminal to evaluate the target object in the business damage assessment image. A comprehensive assessment is performed, so that the target insurance business can be accurately and efficiently assessed relatively efficiently and quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1 A schematic diagram of the structure of a loss assessment system for insurance business provided in an embodiment of the present application;

[0027] Figure 2 A schematic diagram of the steps of a method for assessing loss of insurance business provided in an embodiment of the present application;

[0028] Figure 3 A schematic diagram of a scenario for labeling an image using a method for loss assessment of insurance business provided in an embodiment of the present application;

[0029] Figure 4 A schematic diagram of the structure of a damage assessment device for insurance business provided in an embodiment of the present application;

[0030] Figure 5 A schematic block diagram of the structure of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0032] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0033] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0034] As the scale of agricultural insurance expands, agricultural insurance operations are facing a paradox of efficiency and risk management, which is particularly evident in information asymmetry and low claims efficiency, which seriously restricts the performance of agricultural insurance's economic compensation, financing and social management functions.

[0035] At present, when it comes to claims caused by natural disasters, traditional agricultural insurance often uses random field sampling to conduct inspections and loss assessments during the crop growth period and after a natural disaster occurs. However, if the insured object is affected by a large area due to natural disasters in certain regions, the traditional inspection and loss assessment model often has problems such as information asymmetry, low claims efficiency, and many claims disputes.

[0036] Based on this, the present application provides a method, device, server and storage medium for loss assessment of insurance business, aiming to realize claim assessment of insurance business efficiently and accurately.

[0037] Some embodiments of the present application are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0038] See also Figure 1 , Figure 1 A loss assessment system for insurance business is provided in an embodiment of the present application.

[0039] like Figure 1As shown, the insurance business loss assessment system 100 includes a client terminal 10, a server 20 and a service terminal 30, wherein the client terminal 10 is installed with an insurance business loss assessment client program, and the service terminal 30 is installed with an insurance business loss assessment service program. Generally, insurance business includes but is not limited to vehicle insurance, agricultural insurance, property insurance, etc. The client terminal 10 and the service terminal 30 may include but are not limited to: smart phones, tablet computers, laptop computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, smart wearable devices, etc.

[0040] Server 20 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be 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, content delivery networks (CDN), as well as big data and artificial intelligence platforms.

[0041] For ease of explanation, in the embodiments of the present application, an example is given in which the insurance business purchased by the user is agricultural insurance.

[0042] After the user purchases agricultural insurance, if the insured crops are damaged, and in compliance with the insurance claim terms, the user can operate the client terminal 10 to upload crop damage images (also called business damage assessment images) to the server 20, so that the server 20 can conduct a preliminary assessment of the damaged crops in the received images, obtain a preliminary damage assessment (also called first damage assessment information), and transmit the corresponding image to the corresponding service terminal 30, so that the reviewer can further assess the degree of damage to the current insured crops by referring to the images displayed by the service terminal 30, and obtain a secondary damage assessment (also called second damage assessment information), so as to conduct a comprehensive assessment based on the preliminary damage assessment and the secondary damage assessment, and obtain the final damage assessment of the current insured crops, so as to make a claim based on the final damage assessment.

[0043] Furthermore, the images uploaded by the client terminal 10 may have different image quality issues, for example, some images are not clear enough and some images are too large. Therefore, in order to improve the image quality, after receiving the images uploaded by the client terminal 10, the server 20 needs to process the images and then transmit the processed images to the corresponding service terminal 30, so that the reviewer can further evaluate the degree of damage to the currently insured crops by checking the images displayed by the service terminal 30.

[0044] Please refer to Figure 2 , Figure 2 A schematic flowchart of the steps of a method for loss assessment of insurance business provided in an embodiment of the present application.

[0045] like Figure 2 As shown, the loss assessment method for the insurance business includes steps S101 to S106.

[0046] Step S101: receiving a business damage assessment image corresponding to a target insurance business to be assessed, and determining object attributes of a target object in each of the business damage assessment images based on image features of the business damage assessment images, wherein the object attributes include an object type of the target object and a degree of damage to the target object.

[0047] For example, in the embodiments of the present application, the damage assessment method for insurance business is applied to the aforementioned damage assessment system 100 for insurance business, and is executed by the server 20 in the damage assessment system for insurance business as an example for description.

[0048] After the user purchases agricultural insurance, if the insured crops are damaged, the user can control the client terminal 10 to upload the business damage assessment image to the server 20 if the insurance claim terms are met. The business damage assessment image includes at least one target object, which is the insured target crop. After receiving the business damage assessment image transmitted by the client terminal 10, the server 20 extracts the image features of the business damage assessment image through a preset feature extraction model or feature extraction algorithm, and analyzes the acquired image features. Among them, the image features include but are not limited to the color features, texture features, and shape features of the image.

[0049] Different target objects in the business damage assessment image have different image features, and the same target object has different corresponding image features under different degrees of damage. Therefore, after extracting the image features of the business damage assessment image, the acquired image features can be analyzed, and the object attributes of the target objects in each business damage assessment image can be obtained according to the analysis results. The object attributes include the object type of the target object and the degree of damage of the target object.

[0050] In some implementations, determining the object attributes of the target objects in each of the business damage assessment images based on the image features of the business damage assessment images includes:

[0051] Determine the object identification area where the target object is located from the business damage assessment image, and extract image features of the object identification area;

[0052] The target object in the business damage assessment image is analyzed based on the image features to obtain object attributes of the target object.

[0053] Optionally, determining the object identification area where the target object is located from the business damage assessment image includes: selecting the target object of the business damage assessment image with a labeling box using an image labeling model, and determining the object identification area where the target object is located in the business damage assessment image based on the labeling box.

[0054] Optionally, determining the object identification area where the target object is located from the business damage assessment image includes: identifying the object contour of the target object in the business damage assessment image, and determining the object identification area where the target object is located in the business damage assessment image based on the object contour, wherein the object identification area is larger than the area where the object contour is located.

[0055] like Figure 3 As shown, exemplarily, in order to improve the efficiency of image feature extraction, an object identification area with a target object is selected from a business damage assessment image, and the image features of the object identification area are extracted, thereby enabling feature extraction of the corresponding business damage assessment image to be achieved at a faster rate.

[0056] Specifically, the target object 201 in the business damage assessment image 200 is annotated with a labeling box using a trained image labeling model. For example, the target object in the business damage assessment image 200 is selected with a dotted labeling box. After the labeling of the target object 201 is completed, the minimum area 200a including all the labeling boxes is segmented from the business damage assessment image 200, and the area is used as the object identification area where the target object is located.

[0057] Alternatively, the business damage assessment image is gray-scaled, and the gray-scaled business damage assessment image is binarized to obtain a binarized damage assessment image. Then, the object contour of the target object in the binarized business damage assessment image (binarized damage assessment image) is identified by a preset contour recognition algorithm, and the minimum area including all object contours is segmented from the business damage assessment image, and the area is used as the object recognition area where the target object is located, wherein the object recognition area is larger than the area where the object contour is located, and the contour recognition algorithm includes but is not limited to a contour recognition algorithm based on connectivity and a contour recognition algorithm based on edge detection.

[0058] After the object identification area where the target object is located is identified from the business damage assessment image, the image features of the object identification area are extracted, so that the feature extraction of the corresponding business damage assessment image can be achieved at a faster level.

[0059] It can be understood that the method of extracting the image features of the object recognition area can be to extract the image features in the object recognition area through a preset feature extraction algorithm, or to cut out the object recognition area from the business damage assessment image and input the object recognition area as input information to a preset feature extraction model, thereby extracting the image features of the object recognition area, and using the image features as the image features of the business damage assessment image, and finally realizing the feature extraction of the corresponding business damage assessment image at a faster level.

[0060] After the image features are acquired, the acquired image features are analyzed so that the object attributes of the target object in the corresponding business damage assessment image can be acquired based on the image features. The object attributes include the object type of the target object and the degree of damage of the target object.

[0061] For example, after extracting image feature A of business damage assessment image A and analyzing image feature A, it is found that business damage assessment image A includes target objects of type 1 and target objects of type 2, among which target objects of type 1 are not damaged, target objects of type 2 are seriously damaged, and the damaged area is about 50 square meters, accounting for 49% of the total insured area.

[0062] Image feature B of business damage assessment image B is extracted, and after analyzing image feature B, it is found that business damage assessment image B includes 1 type of target objects and 3 types of target objects, among which type 1 target objects are not damaged, type B target objects are seriously damaged, and the damaged area is about 100 square meters, accounting for 98% of the total insured area.

[0063] Step S102: Obtain image labels and first damage assessment information corresponding to each of the business damage assessment images according to the object type and the damage degree.

[0064] Exemplarily, the server 20 stores a first association relationship between the degree of damage and the image label. After obtaining the degree of damage of the target object in each business damage assessment image, the server 20 determines an image label adapted to the degree of damage according to the degree of damage and the first association relationship. The image label is used to characterize the degree of damage of the target object in the corresponding business damage assessment image. For example, the image label includes a first type of label that characterizes that the target object is not damaged or slightly damaged, and a second type of label that characterizes that the target object is severely damaged.

[0065] At the same time, the server 20 also stores a damage assessment function, which is related to the object type of the target object and the degree of damage of the target object. After obtaining the object type of the target object and the degree of damage of the target object, the first damage assessment information corresponding to each business damage image uploaded by the client terminal 10 can be obtained through the preset damage assessment function.

[0066] Step S103: annotating the business damage assessment image according to the image label to obtain a business annotated image, and preprocessing the business annotated image to obtain a preprocessed image, wherein the preprocessed image has label information corresponding to the image label.

[0067] Exemplarily, in order to enhance the reliability of the first damage assessment information obtained by the server 20, the server 20 also needs to send the received business damage assessment image to the corresponding service terminal 30 so that the user can perform a secondary assessment at the service terminal 30.

[0068] The server 30 labels each business damage assessment image according to the acquired image label, and preprocesses the labeled image to obtain a preprocessed image, and the obtained preprocessed image has label information corresponding to the image label. The preprocessing of the image includes but is not limited to at least one of image compression and image denoising.

[0069] The pre-processed image obtained by compressing the business annotated image can effectively reduce the time required for transmitting the image from the service area 20 to the service terminal 30, thereby reducing image transmission time and data traffic consumption, and ultimately improving the insurance processing speed.

[0070] The preprocessed image obtained by denoising the business annotated image can effectively improve the image quality, so as to improve the image effect displayed on the service terminal 30 .

[0071] Optionally, preprocessing the business annotated image to obtain a preprocessed image includes:

[0072] The business annotated image is compressed using a preset image processing model to obtain a preprocessed image. The image processing model includes an image coding network, a quantization network and an image decoding network, wherein the image coding network is used to convert the business annotated image into a corresponding image feature vector, the quantization network is used to discretize the image feature vector output by the image coding network, and the image coding network is used to restore the discretized image feature vector to obtain the preprocessed image. The image coding network includes at least a residual network or a U-Net network, and the network structures of the image coding network and the image decoding network are symmetrical.

[0073] Exemplarily, a deep convolutional neural network is used as an image coding network, which is responsible for extracting features from the original business annotated image and compressing information to obtain an image feature vector. The image coding network can be based on a residual network (ResNet), U-Net or other efficient architectures to gradually reduce the spatial dimension of the feature map extracted from the original business annotated image while maintaining important image details.

[0074] The image coding network adds the eigenvalues ​​corresponding to the image feature vector obtained from the business annotated image to the quantization layer, which is used to convert the continuous eigenvalues ​​into discrete representations through a preset quantization strategy, where the quantization strategy includes uniform quantization, vector quantization (VQ) or nearest neighbor quantization.

[0075] The image decoding network is symmetrical with the image encoding network in structure, and is used to restore the quantized eigenvalues ​​to an image close to the original image (business annotated image), thereby reconstructing a high-quality image with as few bits as possible to obtain a preprocessed image.

[0076] Optionally, the image processing model uses a combination of mean square error (MSE) and structural similarity index (SSIM) as a loss function to ensure that the reconstructed image (preprocessed image) is close to the original image (business annotated image) at the pixel level and maintains structural similarity.

[0077] Step S104: obtaining an image compression strategy for the pre-processed image at least according to the tag information, and compressing the pre-processed image according to the image compression strategy to obtain a target compressed image.

[0078] Exemplarily, in order to take into account both the efficiency of image transmission and the reliability of damage assessment analysis based on the image, the image compression strategy of the target compressed image to be transmitted to the service terminal 30 is determined according to the tag information. For example, when it is known from the tag information that the damage degree of the target object in the pre-processed image is relatively serious, at this time, in order to ensure that the damaged details of the target object can be seen in the image received by the service terminal 30, the pre-processed image should be compressed with a lower image compression rate, so that the damaged details of the target object in the obtained target compressed image have higher clarity, which is convenient for the auditor to review the damage degree of the target object.

[0079] When it is known from the label information that the target object in the preprocessed image is not damaged or is less damaged, the preprocessed image has little auxiliary effect on the damage assessment of the target object. A higher image compression rate should be sampled to compress the preprocessed image, which can effectively reduce the file size of the image and speed up the transmission of the image.

[0080] In some implementations, obtaining the image compression strategy of the preprocessed image at least according to the tag information includes:

[0081] Determine the damage degree of the target object in the preprocessed image based on the tag information, and obtain device performance parameters and network environment parameters of the service terminal connected to the server for communication;

[0082] An image compression strategy for the preprocessed image is determined according to at least one of the device performance parameter, the network environment parameter, and the damage degree.

[0083] Optionally, determining the image compression strategy of the pre-processed image according to at least one of the device performance parameter, the network environment parameter and the damage degree includes:

[0084] When the device performance parameter meets the preset performance requirement, the network environment parameter meets the preset data transmission requirement, and the damage degree of the target object is a type of damage, determining that the image compression strategy of the preprocessed image is a first compression strategy;

[0085] When the device performance parameters do not meet the preset performance requirements, or the network environment parameters do not meet the preset data transmission requirements, or the damage degree of the target object is Class II damage, the image compression strategy of the preprocessed image is determined to be the second compression strategy, wherein the damage degree of the target object corresponding to Class II damage is less than the damage degree of the target object corresponding to Class I damage, and the image compression rate of the second compression strategy is less than the image compression rate of the first compression strategy.

[0086] Optionally, determining the image compression strategy of the pre-processed image according to at least one of the device performance parameter, the network environment parameter and the damage degree includes:

[0087] When the device performance parameter meets the preset performance requirement and the damage degree of the target object is a type of damage, determining that the image compression strategy of the preprocessed image is a first compression strategy;

[0088] When the device performance parameters do not meet the preset performance requirements or the damage degree of the target object is Class II damage, the image compression strategy of the preprocessed image is determined to be the second compression strategy, wherein the damage degree of the target object corresponding to Class II damage is less than the damage degree of the target object corresponding to Class I damage, and the image compression rate of the second compression strategy is less than the image compression rate of the first compression strategy.

[0089] Optionally, determining the image compression strategy of the pre-processed image according to at least one of the device performance parameter, the network environment parameter and the damage degree includes:

[0090] When the network environment parameters meet the preset data transmission requirements and the damage degree of the target object is a type of damage, determining that the image compression strategy of the pre-processed image is a first compression strategy;

[0091] When the network environment parameters do not meet the preset data transmission requirements or the damage degree of the target object is Class II damage, the image compression strategy of the preprocessed image is determined to be the second compression strategy, wherein the damage degree of the target object corresponding to Class II damage is less than the damage degree of the target object corresponding to Class I damage, and the image compression rate of the second compression strategy is less than the image compression rate of the first compression strategy.

[0092] Optionally, determining the image compression strategy of the pre-processed image according to at least one of the device performance parameter, the network environment parameter and the damage degree includes:

[0093] When the damage degree of the target object is type I damage, the image compression strategy of the preprocessed image is determined to be the first compression strategy; when the damage degree of the target object is type II damage, the image compression strategy of the preprocessed image is determined to be the second compression strategy, wherein the damage degree of the target object corresponding to type II damage is less than the damage degree of the target object corresponding to type I damage, and the image compression rate of the second compression strategy is less than the image compression rate of the first compression strategy.

[0094] Exemplarily, the device performance parameter includes at least one of CPU performance and GPU performance. The network environment parameter includes at least network bandwidth.

[0095] If it is detected that the network environment parameters of the service terminal 30 cannot meet the preset data transmission requirements, indicating that the network bandwidth of the service terminal 30 is less than the preset network bandwidth value, then the image compression strategy of the pre-processed image is determined to be the second compression strategy, that is, the pre-processed image is compressed using the second image compression rate to reduce the image file size and speed up the transmission speed.

[0096] On the contrary, if it is detected that the network environment parameters of the service terminal 30 can meet the preset data transmission requirements, indicating that the network bandwidth of the service terminal 30 is greater than or equal to the preset network bandwidth value, then the image compression strategy of the preprocessed image is determined to be the first compression strategy, and a lower image compression rate can be selected to provide a higher quality picture.

[0097] Alternatively, if it is detected that the device performance parameters of the service terminal 30 cannot meet the preset device performance parameters, indicating that the service terminal 30 has a weak image processing capability, then the image compression strategy of the pre-processed image is determined to be the second compression strategy, that is, the pre-processed image is compressed using the second image compression rate to reduce the image file size and speed up the transmission speed so that the compressed image can be loaded and displayed faster.

[0098] On the contrary, if it is detected that the device performance parameters of the service terminal 30 meet the preset device performance parameters, indicating that the service terminal 30 has a strong image processing capability, then the image compression strategy of the preprocessed image is determined to be the first compression strategy, and a lower image compression rate can be selected to provide a higher quality picture.

[0099] Alternatively, when the damage degree of the target object in the preprocessed image is Class II damage, it indicates that the damage degree of the target object in the preprocessed image is relatively serious. At this time, in order to ensure that the damaged details of the target object can be seen in the image received by the service terminal 30, the preprocessed image should be compressed with a lower image compression rate, so that the damaged details of the target object in the obtained target compressed image have high clarity, which is convenient for the auditor to review and determine the damage degree of the target object. That is, the image compression strategy of the preprocessed image is determined to be the second compression strategy.

[0100] That is, if it is detected that the network environment parameters of the service terminal 30 meet the preset data transmission requirements, the device performance parameters meet the preset device performance parameters, and the damage degree of the target object in the pre-processed image is Class I damage, it indicates that the service terminal 30 has a strong image processing capability and sufficient bandwidth performance, and the damage degree of the target object in the pre-processed image is light or undamaged, the image compression strategy of the pre-processed image is determined to be the first compression strategy, and a lower image compression rate can be selected to provide a higher quality picture.

[0101] Step S105: at least transmitting the target compressed image to the service terminal, so that the service terminal displays the target compressed image on a preset interface.

[0102] Exemplarily, after obtaining the target compressed image, the target compressed image is transmitted to the service terminal 30 so that the target compressed image can be displayed on a preset interface of the service terminal 30, so that the auditor can read the corresponding target compressed image and conduct a worker review and assessment of the damage to the target object based on the image to obtain a second damage assessment information of the damage to the target object.

[0103] Alternatively, after obtaining the target compressed image, the target compressed image and the first damage assessment information are transmitted to the service terminal 30 so that the target compressed image can be displayed on a preset interface of the service terminal 30, so that the reviewer can read the corresponding target compressed image and the first damage assessment information, and then conduct a worker review and assessment of the damage to the target object based on the image to obtain second damage assessment information on the damage to the target object.

[0104] Step S106: receiving the second damage assessment information acquired by the service terminal, and obtaining the insurance damage assessment information of the target insurance business based on the first damage assessment information and the second damage assessment information.

[0105] Exemplarily, after reviewing the target compressed image or the target compressed image and the first damage assessment information through the service terminal 30, the reviewer obtains the second damage assessment information of the target object, and sends the second damage assessment information to the server 20, so that the server 20 is provided with a damage assessment program or algorithm, which obtains the final damage assessment information of the target object through the damage assessment program or algorithm based on the analysis of the first damage assessment information and the second damage assessment information, that is, obtains the insurance loss information of the target insurance business.

[0106] See also Figure 4 , Figure 4 A schematic block diagram of the structure of a damage assessment device for insurance business provided in an embodiment of the present application.

[0107] like Figure 4 As shown, the insurance business damage assessment device 300 includes an image receiving module 301, an image analysis module 302, an image processing module 303, an image compression module 304, an image transmission module 305 and a damage assessment module 306. Among them, the image receiving module 301 is used to receive the business damage assessment image corresponding to the target insurance business to be assessed, and determine the object attributes of the target object in each of the business damage assessment images based on the image features of the business damage assessment image, and the object attributes include the object type of the target object and the damage degree of the target object. The image analysis module 302 is used to obtain the image label and the first damage assessment information corresponding to each of the business damage assessment images according to the object type and the damage degree. The image processing module 303 is used to annotate the business damage assessment image according to the image label to obtain a business annotated image, and preprocess the business annotated image through a preset image processing model to obtain a preprocessed image. The image compression module 304 is used to obtain the image compression strategy of the preprocessed image, and compress the preprocessed image according to the image compression strategy to obtain a target compressed image. The image transmission module 305 is used to transmit at least the target compressed image to the service terminal so that the service terminal displays the target compressed image in a preset interface. The damage assessment module 306 receives the second damage assessment information obtained by the service terminal, and obtains the insurance damage assessment information of the target insurance business based on the first damage assessment information and the second damage assessment information.

[0108] It can be understood that the loss assessment device 300 for insurance business can be applied to a computer device and used to execute the steps of the loss assessment method for insurance business provided in any embodiment of the present application.

[0109] It should be noted that, those skilled in the art can clearly understand that, for the sake of convenience and brevity of description, the specific working process of the damage assessment device 300 for insurance business described above can refer to the corresponding process in the aforementioned embodiment of the damage assessment method for insurance business, and will not be repeated here.

[0110] In some implementations, determining the object attributes of the target objects in each of the business damage assessment images based on the image features of the business damage assessment images includes:

[0111] Determine the object identification area where the target object is located from the business damage assessment image, and extract image features of the object identification area;

[0112] The target object in the business damage assessment image is analyzed based on the image features to obtain object attributes of the target object.

[0113] In some implementations, determining the object identification area where the target object is located from the business damage assessment image includes:

[0114] Identify the object contour of the target object in the business damage assessment image, and determine the object identification area where the target object is located in the business damage assessment image based on the object contour, wherein the object identification area is larger than the area where the object contour is located.

[0115] In some embodiments, the preprocessing of the business annotated image to obtain the preprocessed image includes: using a preset image processing model to compress the business annotated image to obtain the preprocessed image, the image processing model includes an image coding network, a quantization network and an image decoding network, wherein the image coding network is used to convert the business annotated image into a corresponding image feature vector, the quantization network is used to discretize the image feature vector output by the image coding network, the image coding network is used to restore the discretized image feature vector to obtain the preprocessed image, the image coding network includes at least a residual network or a U-Net network, and the network structures of the image coding network and the image decoding network are symmetrical.

[0116] In some implementations, obtaining the image compression strategy of the preprocessed image at least according to the tag information includes:

[0117] Determine the damage degree of the target object in the preprocessed image based on the tag information, and obtain device performance parameters and network environment parameters of the service terminal connected to the server for communication;

[0118] An image compression strategy for the preprocessed image is determined according to at least one of the device performance parameter, the network environment parameter, and the damage degree.

[0119] In some implementations, determining the image compression strategy of the pre-processed image according to at least one of the device performance parameter, the network environment parameter, and the damage degree includes:

[0120] When the device performance parameter meets the preset performance requirement, the network environment parameter meets the preset data transmission requirement, and the damage degree of the target object is a type of damage, determining that the image compression strategy of the preprocessed image is a first compression strategy;

[0121] When the device performance parameters do not meet the preset performance requirements, or the network environment parameters do not meet the preset data transmission requirements, or the damage degree of the target object is Class II damage, the image compression strategy of the preprocessed image is determined to be the second compression strategy, wherein the damage degree of the target object corresponding to Class II damage is less than the damage degree of the target object corresponding to Class I damage, and the image compression rate of the second compression strategy is greater than the image compression rate of the first compression strategy.

[0122] In some implementations, the network environment parameter includes at least network bandwidth, and / or the device performance parameter includes at least one of CPU performance and GPU performance. Figure 5 , Figure 5 A schematic diagram of the structure of a server provided in an embodiment of the present application.

[0123] like Figure 5 As shown, the server 20 includes a processor 21 and a memory 22, and the processor 21 and the memory 22 are connected via a bus 23, such as an I2C (Inter-integrated Circuit) bus.

[0124] Specifically, the processor 21 is used to provide computing and control capabilities to support the operation of the entire server 20. The processor 21 can be a central processing unit (CPU), and the processor 21 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0125] Specifically, the memory 22 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0126] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a partial structure related to the embodiment of the present application, and does not constitute a limitation on the computer device to which the embodiment of the present application is applied. The specific server may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0127] The processor 21 is used to run a computer program stored in the memory, and implement any one of the insurance business loss assessment methods provided in the embodiments of the present application when executing the computer program.

[0128] In some embodiments, the processor 21 is used to run a computer program stored in the memory, and implements the following steps when executing the computer program:

[0129] Receiving a business damage assessment image corresponding to a target insurance business to be assessed, and determining object attributes of a target object in each of the business damage assessment images based on image features of the business damage assessment images, wherein the object attributes include an object type of the target object and a degree of damage to the target object;

[0130] Acquire the image label and the first damage assessment information corresponding to each of the business damage assessment images according to the object type and the damage degree;

[0131] Annotating the business damage assessment image according to the image label to obtain a business annotated image, and preprocessing the business annotated image to obtain a preprocessed image, wherein the preprocessed image has label information corresponding to the image label;

[0132] At least acquiring an image compression strategy of the preprocessed image according to the label information, and compressing the preprocessed image according to the image compression strategy to obtain a target compressed image;

[0133] At least transmitting the target compressed image to a service terminal so that the service terminal displays the target compressed image on a preset interface;

[0134] The second damage assessment information acquired by the service terminal is received, and the insurance damage assessment information of the target insurance business is obtained based on the first damage assessment information and the second damage assessment information.

[0135] In some implementations, determining the object attributes of the target objects in each of the business damage assessment images based on the image features of the business damage assessment images includes:

[0136] Determine the object identification area where the target object is located from the business damage assessment image, and extract image features of the object identification area;

[0137] The target object in the business damage assessment image is analyzed based on the image features to obtain object attributes of the target object.

[0138] In some implementations, determining the object identification area where the target object is located from the business damage assessment image includes:

[0139] Identify the object contour of the target object in the business damage assessment image, and determine the object identification area where the target object is located in the business damage assessment image based on the object contour, wherein the object identification area is larger than the area where the object contour is located.

[0140] In some embodiments, the preprocessing of the business annotated image to obtain the preprocessed image includes: using a preset image processing model to compress the business annotated image to obtain the preprocessed image, the image processing model includes an image coding network, a quantization network and an image decoding network, wherein the image coding network is used to convert the business annotated image into a corresponding image feature vector, the quantization network is used to discretize the image feature vector output by the image coding network, the image coding network is used to restore the discretized image feature vector to obtain the preprocessed image, the image coding network includes at least a residual network or a U-Net network, and the network structures of the image coding network and the image decoding network are symmetrical.

[0141] In some implementations, obtaining the image compression strategy of the preprocessed image at least according to the tag information includes:

[0142] Determine the damage degree of the target object in the preprocessed image based on the tag information, and obtain device performance parameters and network environment parameters of the service terminal connected to the server for communication;

[0143] An image compression strategy for the preprocessed image is determined according to at least one of the device performance parameter, the network environment parameter, and the damage degree.

[0144] In some implementations, determining the image compression strategy of the pre-processed image according to at least one of the device performance parameter, the network environment parameter, and the damage degree includes:

[0145] When the device performance parameter meets the preset performance requirement, the network environment parameter meets the preset data transmission requirement, and the damage degree of the target object is a type of damage, determining that the image compression strategy of the preprocessed image is a first compression strategy;

[0146] When the device performance parameters do not meet the preset performance requirements, or the network environment parameters do not meet the preset data transmission requirements, or the damage degree of the target object is Class II damage, the image compression strategy of the preprocessed image is determined to be the second compression strategy, wherein the damage degree of the target object corresponding to Class II damage is less than the damage degree of the target object corresponding to Class I damage, and the image compression rate of the second compression strategy is greater than the image compression rate of the first compression strategy.

[0147] In some implementations, the network environment parameters include at least network bandwidth, and / or the device performance parameters include at least one of CPU performance and GPU performance.

[0148] It should be noted that, technical personnel in the relevant field can clearly understand that, for the convenience and brevity of description, the specific working process of the processor in the server 20 described above calling the program to execute the steps of the damage assessment method for insurance business can refer to the corresponding process in the aforementioned embodiment of the damage assessment method for insurance business, and will not be repeated here.

[0149] An embodiment of the present application also provides a storage medium for computer-readable storage, wherein 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 steps of a method for loss assessment of insurance business as provided in any embodiment of the present application specification.

[0150] The storage medium may be an internal storage unit of the server in the aforementioned embodiment, such as a hard disk or memory of the server. The storage medium may also be an external storage device of a computer device, such as a plug-in hard disk equipped on the server, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc.

[0151] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0152] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0153] The serial numbers of the embodiments of the present application are for description only and do not represent the advantages and disadvantages of the embodiments. The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for loss assessment of insurance business, characterized in that: The method comprises: Receiving a business damage assessment image corresponding to a target insurance business to be assessed, and determining object attributes of a target object in each of the business damage assessment images based on image features of the business damage assessment images, wherein the object attributes include an object type of the target object and a degree of damage to the target object; Acquire the image label and the first damage assessment information corresponding to each of the business damage assessment images according to the object type and the damage degree; Annotating the business damage assessment image according to the image label to obtain a business annotated image, and preprocessing the business annotated image to obtain a preprocessed image, wherein the preprocessed image has label information corresponding to the image label; At least acquiring an image compression strategy of the preprocessed image according to the label information, and compressing the preprocessed image according to the image compression strategy to obtain a target compressed image; At least transmitting the target compressed image to a service terminal so that the service terminal displays the target compressed image on a preset interface; The second damage assessment information acquired by the service terminal is received, and the insurance damage assessment information of the target insurance business is obtained based on the first damage assessment information and the second damage assessment information.

2. The method according to claim 1, characterized in that: The determining the object attribute of the target object in each of the business damage assessment images based on the image features of the business damage assessment images includes: Determine the object identification area where the target object is located from the business damage assessment image, and extract image features of the object identification area; The target object in the business damage assessment image is analyzed based on the image features to obtain object attributes of the target object.

3. The method according to claim 2, characterized in that The determining, from the business damage assessment image, the object identification area where the target object is located comprises: Identify the object contour of the target object in the business damage assessment image, and determine the object identification area where the target object is located in the business damage assessment image based on the object contour, wherein the object identification area is larger than the area where the object contour is located.

4. The method according to claim 1, characterized in that The preprocessing of the business annotated image to obtain a preprocessed image includes: The business annotated image is compressed using a preset image processing model to obtain a preprocessed image. The image processing model includes an image coding network, a quantization network and an image decoding network. The image coding network is used to convert the business annotated image into a corresponding image feature vector. The quantization network is used to discretize the image feature vector output by the image coding network. The image coding network is used to restore the discretized image feature vector to obtain the preprocessed image. The image coding network includes at least a residual network or a U-Net network. The network structures of the image coding network and the image decoding network are symmetrical.

5. The method according to claim 1, characterized in that The step of obtaining the image compression strategy of the pre-processed image at least according to the label information includes: Determine the damage degree of the target object in the preprocessed image based on the tag information, and obtain device performance parameters and network environment parameters of the service terminal connected to the server for communication; An image compression strategy for the preprocessed image is determined according to at least one of the device performance parameter, the network environment parameter, and the damage degree.

6. The method according to claim 5, characterized in that The determining of the image compression strategy of the pre-processed image according to at least one of the device performance parameter, the network environment parameter and the damage degree comprises: When the device performance parameter meets the preset performance requirement, the network environment parameter meets the preset data transmission requirement, and the damage degree of the target object is a type of damage, determining that the image compression strategy of the preprocessed image is a first compression strategy; When the device performance parameters do not meet the preset performance requirements, or the network environment parameters do not meet the preset data transmission requirements, or the damage degree of the target object is Class II damage, the image compression strategy of the preprocessed image is determined to be the second compression strategy, wherein the damage degree of the target object corresponding to Class II damage is less than the damage degree of the target object corresponding to Class I damage, and the image compression rate of the second compression strategy is greater than the image compression rate of the first compression strategy.

7. The method according to claim 5, characterized in that The network environment parameters include at least network bandwidth, and / or the device performance parameters include at least one of CPU performance and GPU performance.

8. A loss assessment device for insurance business, characterized in that: The method comprises: An image receiving module, used to receive a business damage assessment image corresponding to a target insurance business to be assessed, and determine object attributes of a target object in each of the business damage assessment images based on image features of the business damage assessment images, wherein the object attributes include an object type of the target object and a degree of damage to the target object; An image analysis module, used for obtaining image labels and first damage assessment information corresponding to each of the business damage assessment images according to the object type and the damage degree; an image processing module, which annotates the business damage assessment image according to the image label to obtain a business annotated image, and preprocesses the business annotated image to obtain a preprocessed image, wherein the preprocessed image has label information corresponding to the image label; An image compression module, configured to obtain an image compression strategy for the preprocessed image at least according to the tag information, and compress the preprocessed image according to the image compression strategy to obtain a target compressed image; An image transmission module, used for transmitting at least the target compressed image to a service terminal, so that the service terminal displays the target compressed image on a preset interface; The damage assessment module receives the second damage assessment information obtained by the service terminal, and obtains the insurance damage assessment information of the target insurance business based on the first damage assessment information and the second damage assessment information.

9. A server, characterized in that: Including processor and memory; The memory is used to store computer programs; The processor is used to execute the computer program and implement the loss assessment method for insurance business as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: When the computer-readable storage medium is executed by one or more processors, the one or more processors execute the steps of the loss assessment method for insurance business as described in any one of claims 1 to 7.

Images