Image editing method and device, computer equipment and storage medium

Abstract

The invention provides an image editing method and apparatus, a computer device and a storage medium. The method comprises the steps of obtaining a to-be-edited document image and corresponding insurance type information in a preset claim settlement system; detecting a claim settlement related area in the document image according to the insurance type information; detecting a privacy area in the document image; determining a region identification type according to a region contour corresponding to the claim settlement related region; the region identifier type comprises one or more of a circular identifier, a square identifier and an arrow identifier; acquiring color distribution information corresponding to the privacy area in the document image; the color distribution information comprises a color corresponding to each pixel point in the privacy area; generating mosaic information corresponding to the privacy area according to the color distribution information; the mosaic information comprises a superposition color corresponding to each pixel point in the privacy area; and performing color superposition on each pixel point in the privacy area according to the mosaic information, completing marking of the area contour according to the identification type, and completing editing of the document image.

Description

Technical Field

[0001] The present application relates to the field of image processing, which is applied to the financial field, and in particular to an image editing method, device, computer equipment and storage medium. Background Art

[0002] In the financial field, after recording information during the claims process, claims workers usually upload photos of documents as subsequent verification evidence. Document photos usually refer to images of various documents and evidence related to the claims application. However, there are many types of document photos, and it is often necessary to mark some details and annotate some relevant information. The traditional way is to use special image editing software and then upload the processed photos, which is cumbersome and lengthy, and the operator's work experience is poor, which greatly affects the progress of claims work.

[0003] Therefore, there is an urgent need for an image editing method for document photos in the financial field, especially in the insurance claims field, to improve the processing efficiency of claims personnel. Summary of the invention

[0004] The present application provides an image editing method, apparatus, computer equipment and storage medium, which are intended to solve the problem that in the process of financial claims, due to the wide variety of document photos, it is often necessary to mark some details and annotate some relevant information. The traditional method is to use special image editing software and then upload the processed photos, which is cumbersome and lengthy, and the operator's operating experience is poor, which greatly affects the progress of the claims operation.

[0005] In a first aspect, the present application provides an image editing method, comprising:

[0006] Obtaining the document image to be edited and the insurance type information corresponding to the document image in a preset claim settlement system;

[0007] Detect claim-related areas in the document image based on the insurance type information;

[0008] Detecting privacy areas in areas other than claim-related areas in the document image;

[0009] Determine the type of area identification according to the area outline corresponding to the claim-related area in the document image; wherein the type of area identification includes at least one or more of a circular identification, a square identification, and an arrow identification;

[0010] Obtaining color distribution information corresponding to the privacy area in the document image; the color distribution information includes the color corresponding to each pixel in the privacy area;

[0011] Generate mosaic information corresponding to the privacy area according to the color distribution information; the mosaic information includes the superimposed color corresponding to each pixel in the privacy area;

[0012] Color is overlaid on each pixel in the privacy area according to the mosaic information, the area outline is marked according to the identification type, and the document image is edited.

[0013] In some embodiments, a claim-related area is detected in a document image based on insurance type information, including: obtaining a pre-built image recognition model library, the image recognition model library including multiple image recognition models and a recognition category corresponding to each image recognition model; matching the insurance type information with the recognition category, and obtaining an image recognition model corresponding to a successfully matched recognition category; inputting the document image into the image recognition model, the image recognition model parsing the document image, and outputting the claim-related area.

[0014] Exemplarily, before obtaining the pre-built image recognition model library, it also includes: obtaining multiple historical document images of claims that have been settled and the insurance type information and historical claim area corresponding to each historical document image in the claims system; classifying the multiple historical document images according to the insurance type information, and obtaining multiple historical image groups, each historical image group including at least one historical document image; generating multiple image recognition models to be trained according to the number of historical image groups; inputting the historical document images of each historical image group into an image recognition model respectively, and the image recognition model parses each historical document image and outputs the claim prediction area corresponding to each historical document image; completing the training of each corresponding image recognition model according to the claim prediction area and historical claim area corresponding to each historical document image; and generating the recognition category of the corresponding image recognition model according to the insurance type information corresponding to each historical image group.

[0015] In some embodiments, a privacy area is detected in other areas outside the claim-related area in a document image, including: performing optical character recognition in other areas outside the claim-related area to obtain text information corresponding to the document image and position information corresponding to the text information; parsing the text information based on natural language processing technology to obtain semantic information corresponding to the text information; matching the semantic information with a preset privacy type, and generating a privacy area based on the position information corresponding to the successfully matched semantic information.

[0016] In some embodiments, the area identification type is determined based on the area outline corresponding to the claim-related area in the document image, including: parsing the area outline, obtaining geometric feature information corresponding to the area outline; the geometric feature information includes at least one or more of aspect ratio, roundness and corner features; if the regularity corresponding to the area outline determined according to the geometric feature information is greater than the preset regularity, the identification type is determined to be a circular identification; if the regularity corresponding to the area outline determined according to the geometric feature information is less than or equal to the preset regularity, the identification type is determined to be a square identification; if the area corresponding to the area outline determined according to the geometric feature information is less than the preset area, the identification type is determined to be an arrow identification.

[0017] In some embodiments, before completing the editing of the document image, it also includes: obtaining the shooting location information corresponding to the document image; detecting the claims-related area to obtain the element type corresponding to the claims-related area; generating watermark information according to the element type, shooting location information and insurance type information, adding the watermark information to the document image, and completing the editing of the document image.

[0018] In some embodiments, color is overlaid on each pixel in the privacy area according to the mosaic information, and the area outline is marked according to the identification type to complete the editing of the document image, including: obtaining an initialized preset container; converting the document image into a preset document object model format; setting the converted document image as a background image in the preset container; color is overlaid on each pixel in the privacy area corresponding to the background image in the preset container according to the mosaic information, and the area outline is marked according to the identification type to complete the editing of the background image, and uploading the edited background image to the claims system.

[0019] In a second aspect, the present application provides an image editing device, comprising:

[0020] An information acquisition unit, used to acquire the document image to be edited and the insurance type information corresponding to the document image in a preset claim settlement system;

[0021] An area detection unit, used to detect claim-related areas in the document image according to the insurance type information;

[0022] A privacy detection unit, for detecting a privacy area in other areas outside the claim settlement related area in the document image;

[0023] An identification determination unit, used to determine the type of regional identification according to the regional contour corresponding to the claim-related region in the document image; wherein the regional identification type includes at least one or more of a text identification, a circular identification, a square identification, and an arrow identification;

[0024] A distribution determination unit, used to obtain color distribution information corresponding to the privacy area in the document image; the color distribution information includes the color corresponding to each pixel in the privacy area;

[0025] An overlay determination unit, configured to generate mosaic information corresponding to the privacy area according to the color distribution information; the mosaic information includes an overlay color corresponding to each pixel in the privacy area;

[0026] The editing completion unit is used to superimpose the color of each pixel in the privacy area according to the mosaic information, mark the area outline according to the identification type, and complete the editing of the document image.

[0027] In a third aspect, the present application further provides a computer device, including:

[0028] Memory and processor;

[0029] The memory is used to store computer programs;

[0030] The processor is used to execute the computer program and implement the steps of the image editing method described in the first aspect when executing the computer program.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the steps of the image editing method described in the first aspect above.

[0032] The present application provides an image editing method, device, computer equipment and storage medium. The method obtains a document image to be edited from a preset claim system. According to the obtained insurance information, the document image is classified and the area related to the claim is identified. The claim-related area is automatically detected in the image. The area outside the claim-related area is identified as an area that may contain privacy information (i.e., a privacy area). Then the outline of the claim-related area is determined, and its area identification type is defined, and the area is marked using identification styles such as a circle, a square or an arrow. The color distribution information in the privacy area is obtained, and the color of each pixel is analyzed. A mosaic effect is generated based on this information to conceal the privacy information. The generated mosaic information is applied to the privacy area, and each pixel is subjected to color superposition processing. The document image finally edited contains the identified claim-related area, and the privacy information is effectively blocked. In summary, the method improves the image processing efficiency and privacy protection issues in the financial claims process through automated and intelligent image editing technology, making the overall operation smoother and safer.

[0033] The provided method has the following beneficial effects:

[0034] 1. Improve work efficiency: Automated area detection and labeling reduces manual operations and improves the efficiency of claims review and processing.

[0035] 2. User experience optimization: The operation process is simplified, the additional step of using image editing software is eliminated, and the operator's operating experience is significantly improved.

[0036] 3. Privacy protection: It effectively protects sensitive information that may be involved in the document image and ensures information security by automatically adding mosaics to the privacy area.

[0037] 4. Accuracy and consistency: Detect and mark relevant areas based on insurance information and preset rules, improving the accuracy and consistency of image processing.

[0038] 5. Flexibility and scalability: Providing multiple identification types makes this method flexible and able to adapt to different claims needs and insurance characteristics.

[0039] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] 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.

[0041] Figure 1 is a schematic flow chart of steps of an image editing method provided by an embodiment of the present application;

[0042] Figure 2 is a flowchart showing the steps of a method for detecting a claim-related area provided in one embodiment of the present application;

[0043] Figure 3 is a schematic flow chart of the steps of a privacy area detection method provided in an embodiment of the present application;

[0044] Figure 4 is a structural schematic diagram of an image editing device provided by an embodiment of the present application;

[0045] Figure 5 It is a schematic block diagram of the structure of a computer device provided in one embodiment of the present application.

[0046] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0047] 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.

[0048] 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.

[0049] It should be understood that, in order to facilitate the clear description of the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second", etc. are used to distinguish the same items or similar items with substantially the same functions and effects. Those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not necessarily limit the difference.

[0050] 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.

[0051] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0052] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0053] In the financial field, after recording information during the claims process, claims workers usually upload photos of documents as subsequent verification evidence. Document photos usually refer to images of various documents and evidence related to the claims application. However, there are many types of document photos, and it is often necessary to mark some details and annotate some relevant information. The traditional way is to use special image editing software and then upload the processed photos, which is cumbersome and lengthy, and the operator's work experience is poor, which greatly affects the progress of claims work.

[0054] Therefore, there is an urgent need for an image editing method for document photos in the financial field, especially in the insurance claims field, to improve the processing efficiency of claims personnel.

[0055] To solve the above problems, please refer to Figure 1 , Figure 1 1 is a schematic flow chart of an image editing method provided by an embodiment of the present application. The image editing method can be implemented by a computer device, which can be deployed on a single server or a server cluster. It can also be deployed on a handheld terminal, a laptop computer, a wearable device or a robot, etc.

[0056] It should be noted that the acquisition of relevant information involved in the acquisition process of any information such as insurance type information by the provided method complies with relevant regulations and is carried out with the user's consent, and will not infringe on the user's privacy or violate relevant laws and regulations.

[0057] Specifically, Figure 1 As shown, the provided image editing method includes steps S101 to S107, which are described in detail as follows:

[0058] Step S101. Obtain the document image to be edited and the insurance type information corresponding to the document image in a preset claim settlement system.

[0059] Specifically, in the set claims system, the document image to be edited is captured through the API interface or database query function. The image is usually stored in JPEG, PNG and other formats, and the document image to be processed may include various types of insurance documents, such as policy pictures, accident scene photos, etc. In addition to obtaining the image, the method also needs to capture the insurance information related to the image, such as the insurance type (auto insurance, property insurance, health insurance, etc.), policy number, policyholder name and contact information, etc. For example, the computer device obtains a customer's auto insurance policy photo and related information from the database (insurance type: auto insurance, policy number: 123456, policyholder: Zhang San). This ensures that the processed image and data are consistent and reduces the error of information entry. At the same time, the processing speed is increased, so that the claims work can quickly enter the subsequent steps and improve operational efficiency.

[0060] Step S102: Detect claim-related areas in the document image according to the insurance type information.

[0061] Specifically, computer vision technology is used in combination with deep learning algorithms (such as YOLO, Faster R-CNN, etc.) to identify and locate areas related to claims in document images. This step involves semantic segmentation and target detection, and the key content required for insurance claims is identified through training models. Areas related to claims may include loss photos, cost details, accident descriptions, etc. For example, in a photo of a car insurance claim scene, the system identifies the collision car, the damaged area, and the annotations of the accident location. Automated area recognition reduces manual labeling errors and improves processing accuracy. On the basis of high efficiency, it can process multiple images at the same time to speed up claims.

[0062] In some embodiments, Figure 2 As shown, the claim settlement related area is detected in the document image according to the insurance type information, including steps S102a to S102c.

[0063] Step S102a. Obtain a pre-built image recognition model library, where the image recognition model library includes multiple image recognition models and a recognition category corresponding to each image recognition model.

[0064] Step S102b. Match the insurance type information with the identification category, and obtain the image recognition model corresponding to the identification category that has been successfully matched.

[0065] Step S102c: Input the document image into the image recognition model, the image recognition model analyzes the document image, and outputs the claim-related area.

[0066] The image recognition model library is a collection of pre-trained models, each of which is optimized for a specific recognition category. These models may be built using technologies such as convolutional neural networks (CNNs) and deep learning. Each model corresponds to one or more recognition categories and may cover image features in different insurance scenarios, such as vehicle damage, medical bills, accident scenes, etc. The provided insurance information (such as auto insurance, health insurance) is matched with the recognition categories in the model library. This matching process may be achieved with the help of information retrieval technology or simple rule matching to ensure that the image recognition model that best meets the current insurance type is selected. Based on the matching results, a suitable image recognition model is selected for the current document image processing. This process can optimize the efficiency of model selection and prevent unnecessary computational overhead. The document image is input into the selected recognition model. The model parses the image content and outputs the claim-related area. The parsing process involves steps such as feature extraction, feature matching, and boundary detection of the image.

[0067] When processing a car insurance claim, identify a photo of a damaged vehicle. Obtain an image recognition model library, which contains a model specifically for car insurance damage recognition. Match the insurance type "car insurance" and select a recognition model related to damage detection. Input the vehicle photo into the model. The model outputs information about the damaged area of ​​the vehicle in the image for labeling and further processing.

[0068] Accordingly, each image recognition model is optimized for a specific category, providing higher accuracy and precision, which is crucial in the claims process. Through automatic model selection and image analysis, human intervention is reduced, and claim-related areas can be quickly identified in real time, accelerating the entire claims process. Due to the model library structure, new models can be continuously added to support more insurance types and different types of claims needs. The flexible matching mechanism enables the system to adapt to a variety of scenarios, helping to maintain the system's advancement and competitiveness. The automated selection and recognition process reduces human decision-making and operational complexity, and improves the user experience and consistency of processing.

[0069] Through the image recognition mechanism described in the above embodiment, the insurance system can handle various claims-related issues in a more intelligent and automated manner, ensuring efficient operation and user privacy protection. This method not only improves the system's processing capabilities, but also significantly improves user satisfaction and trust.

[0070] Exemplarily, before obtaining the pre-built image recognition model library, it also includes: obtaining multiple historical document images of claims that have been settled and the insurance type information and historical claim area corresponding to each historical document image in the claims system; classifying the multiple historical document images according to the insurance type information, and obtaining multiple historical image groups, each historical image group including at least one historical document image; generating multiple image recognition models to be trained according to the number of historical image groups; inputting the historical document images of each historical image group into an image recognition model respectively, and the image recognition model parses each historical document image and outputs the claim prediction area corresponding to each historical document image; completing the training of each corresponding image recognition model according to the claim prediction area and historical claim area corresponding to each historical document image; and generating the recognition category of the corresponding image recognition model according to the insurance type information corresponding to each historical image group.

[0071] By extracting multiple historical document images of completed claims from the claims system, as well as the insurance type information corresponding to each image and the accurately marked historical claims area. This step requires database operations and data extraction technology. According to the extracted insurance type information, multiple historical document images are classified to form multiple historical image groups. The images in each group belong to the same insurance type or are related to the same type of claims. The classification step can use insurance type labels or document metadata for grouping processing. For each historical image group, initialize an image recognition model to be trained. The initial structure and parameters of these models can be set based on deep learning frameworks such as TensorFlow or PyTorch.

[0072] Input the images in each historical image group into the corresponding image recognition model, parse the image content, and output the claim prediction area. Use the historical claim area as a label to compare and adjust with the prediction area to train the model to improve its recognition accuracy. Use the loss function in supervised learning (such as cross entropy loss) to optimize the model parameters. Define the recognition category of the model based on the insurance type information corresponding to each historical image group. This process ensures that each model focuses on a specific recognition task, improving processing efficiency and accuracy.

[0073] For example, an insurance company has a large number of auto insurance and health insurance claims cases. Extract historical images of auto insurance claims and health insurance claims images from the system. Separately group auto insurance and health insurance into auto insurance groups and health insurance groups. Initialize a recognition model for each group, and train the recognition models for the auto insurance group and health insurance group separately. Input the auto insurance group images into the model for training, and the model learns to recognize the damaged area of ​​the vehicle body; input the health insurance images into another model to learn the recognition of medical expense details. During the training process, the model is calibrated and optimized through the labeled data of the historical claim area. Finally, two recognition models are generated, focusing on the recognition of claim areas for auto insurance and health insurance respectively.

[0074] By utilizing historical data, the system can quickly build and train multiple recognition models, greatly shortening the development and deployment cycle. The training model has high specificity and accuracy for specific insurance types. At the same time, by comparing historical annotated data with predicted output, the model parameters can be accurately adjusted to improve the accuracy and consistency of the recognition results. With a flexible model building and training mechanism, the system can easily expand its functions and adapt to more insurance types by adding new historical data sets and corresponding models. High-precision automated recognition reduces the workload of manual inspection and annotation, accelerates the claims process, and improves overall processing efficiency. Converting large-scale, available historical data into training resources improves the accuracy and practicality of existing technologies and further improves the economic benefits of data use. Through the technical path of the above embodiment, insurance companies can effectively utilize historical claims data, which not only improves the accuracy and efficiency of image recognition, but also optimizes the overall claims process and enhances the value of data acquisition.

[0075] Step S103: Detect privacy areas in other areas outside the claim settlement related area in the document image.

[0076] Specifically, through image processing algorithms (such as edge detection-based algorithms or image segmentation technology), privacy areas that need to be protected are identified and marked in the document image. These privacy areas generally include personal identity information (such as ID number, bank account information), address and contact information, etc. Use OCR (optical character recognition) technology to automatically identify text information in the image to help determine the location of the privacy area. For example, in the image of the auto insurance policy, the insured's name, ID number and other information are identified and marked as privacy areas that need to be protected. Ensure data security by clearly defining the scope of personal privacy information. Comply with current data protection laws and regulations (such as GDPR, CCPA, etc.) to maintain the image and reputation of the company.

[0077] In some embodiments, Figure 3 As shown, a privacy area is detected in other areas outside the claim settlement related area in the document image, including steps S103a to S103c.

[0078] Step S103a. Perform optical character recognition in other areas outside the claim settlement related area to obtain text information corresponding to the document image and position information corresponding to the text information.

[0079] Step S103b. Analyze the text information based on natural language processing technology to obtain semantic information corresponding to the text information.

[0080] Step S103c: Match the semantic information with the preset privacy type, and generate a privacy area for the location information corresponding to the successfully matched semantic information.

[0081] Apply OCR technology to the non-claims-related areas of the document image to extract text information and corresponding location information. OCR technology involves image preprocessing (such as noise reduction and binarization) and character recognition (such as using OCR engines such as Tesseract). The processing results include text content and the coordinates or boundary information of each text block in the image. The extracted text information is parsed using NLP technology to obtain the semantic information of the text. NLP technology may involve operations such as word segmentation, part-of-speech tagging, named entity recognition (NER), and syntactic analysis in this process. The parsing results form a set of semantic tags, which may identify the specific meaning of the text (such as name, address, ID number, etc.). Match the above-obtained semantic information with the preset privacy type. The privacy type can be defined in advance, such as keywords or patterns containing sensitive information such as name, ID number, address, contact information, etc. According to the matching results, determine which text semantics meet the privacy type standards and identify the privacy information that needs to be protected. For the successfully matched privacy information, a privacy area is generated according to its location information. This step usually frames the identified sensitive information location as the area to be masked or processed. These areas can then be blurred or mosaicked to ensure that the information cannot be visually discerned.

[0082] For example, when processing a customer's insurance application form, you want to identify possible personal information. In the application form image, apply OCR to detect and extract text information, such as "Name: Zhang San, Phone: 1234567890". Use NLP to analyze the extracted information and identify the semantics of "Zhang San" and "1234567890" as "name" and "phone number" respectively. Match the semantics of "name" and "phone number" with the preset privacy type to confirm that these two pieces of information belong to the privacy content to be protected. Based on the matching results, determine their location in the image and generate an area for subsequent privacy processing.

[0083] By utilizing OCR and NLP technologies, privacy information can be automatically detected and identified without manual item-by-item inspection, thus improving efficiency and accuracy. Matching based on semantic information is more flexible and adaptable than relying solely on keywords, and has better response capabilities to diverse text expressions and language changes. Automatic identification and labeling of privacy areas can ensure that sensitive data is not leaked, helping companies comply with privacy protection regulations such as GDPR and reduce legal risks. Rapid identification and processing of privacy information makes customers more assured when providing materials and enhances the trust relationship between customers and companies. This technology is not limited to the insurance field, but can also be applied to industries such as banking, medical care and government agencies that need to process large amounts of personal information, and has a wide range of applicability. Through the above embodiments, insurance companies and related companies can more effectively and intelligently protect customers' privacy information, improve processing efficiency, reduce the risk of human error, and provide strong technical support for companies to meet regulatory compliance requirements.

[0084] Step S104. Determine the area identification type according to the area contour corresponding to the claim-related area in the document image; wherein the area identification type includes at least one or more of a circular identification, a square identification, and an arrow identification.

[0085] Specifically, the most appropriate identification type is determined based on the shape and features of the detected claims-related areas. This step is mainly based on the analysis of the geometric features of the image area, and the selection of the identification type is determined by the rule engine and the model. There are many options for the type of area identification, including circles, squares, arrows, etc. Such diversity can adapt to different usage scenarios. For example, for vehicle loss photos, choose to use square identification to surround the damaged area, and use arrow identification in the accident description area to point to the cause of the accident. Flexible identification methods allow users to clearly identify important information and improve the readability of the image. Improving the accuracy of annotations will help claims auditors quickly understand the course of the accident and the loss situation.

[0086] Step S105. Obtain color distribution information corresponding to the privacy area in the document image; the color distribution information includes the color corresponding to each pixel in the privacy area.

[0087] Specifically, the color information of each pixel in the privacy area is obtained through a pixel-level processing algorithm, which is mainly achieved by traversing all the pixels in the area. This information provides basic data for the subsequent generation of mosaics. Methods such as color histograms can be used to analyze and summarize the color distribution in the privacy area. For example, the RGB value of each pixel in the customer ID number area is extracted to form a detailed color data table. By providing the necessary color information for the processing of the privacy area, the naturalness and consistency of the mosaic effect are ensured. Avoid visual abruptness after processing and make the final result more beautiful.

[0088] Step S106: Generate mosaic information corresponding to the privacy area according to the color distribution information; the mosaic information includes the superimposed color corresponding to each pixel in the privacy area.

[0089] Specifically, the color distribution information of the privacy area is used to generate a corresponding mosaic effect. This is usually achieved by color overlay processing on each pixel to ensure that the privacy information cannot be identified. Blurring or blocking technology is used to render the privacy area into a mosaic pattern composed of small blocks of color, effectively blocking its real information. For example, by applying mosaics to the customer's ID card area, a blurred area is generated according to the color distribution of pixels to conceal the sensitive information therein. It can effectively prevent the leakage of sensitive information and protect customer privacy. The digitally processed image still maintains good visual consistency without affecting the basic functions of the document.

[0090] Step S107: Color each pixel in the privacy area is superimposed according to the mosaic information, the area outline is marked according to the identification type, and the document image is edited.

[0091] Specifically, the mosaic information generated in the previous step is used to overlay the corresponding color on each pixel in the privacy area. According to the logo type (circle, square, arrow, etc.), add logos and explanatory text to the claim-related area. Finally, a new edited document image is formed and is ready to be uploaded to the claims system or sent to relevant reviewers. For example, the computer equipment processes the customer's ID card area into a mosaic, and adds a square logo and the descriptive text "accident damage" around the vehicle loss area to generate the final edited image. Automated editing significantly improves the efficiency of claims processing and reduces the probability of manual changes. The professionalization and refinement of the finished image improves the satisfaction of customers and reviewers with the claims process and enhances the user experience of the entire claims system.

[0092] Through the above steps, the provided image editing method greatly improves the image processing efficiency in the financial claims process, protects customer privacy, and ensures the accuracy of claims information, thereby promoting the optimization of the entire claims process.

[0093] In some embodiments, the area identification type is determined based on the area outline corresponding to the claim-related area in the document image, including: parsing the area outline, obtaining geometric feature information corresponding to the area outline; the geometric feature information includes at least one or more of aspect ratio, roundness and corner features; if the regularity corresponding to the area outline determined according to the geometric feature information is greater than the preset regularity, the identification type is determined to be a circular identification; if the regularity corresponding to the area outline determined according to the geometric feature information is less than or equal to the preset regularity, the identification type is determined to be a square identification; if the area corresponding to the area outline determined according to the geometric feature information is less than the preset area, the identification type is determined to be an arrow identification.

[0094] Perform contour detection on the claim-related area in the document image, which may use image processing techniques such as edge detection algorithms (such as Canny edge detection) and contour extraction techniques (such as findContours in OpenCV). The detection results provide boundary information of the area for further analysis. Analyze the geometric characteristics of the contour, including the following: Aspect ratio: Calculate the ratio of the width to the height of the contour. Roundness: Calculate the degree of deviation from a perfect circle, which may be approximated based on the ratio of the calculated area to the perimeter. Corner features: Analyze whether the contour presents obvious corners, such as the sharp corners of a rectangle.

[0095] Regularity is determined through geometric feature information, such as judging the symmetry and uniformity of the area. According to the preset regularity threshold comparison, if it is greater than this value, the area is considered regular (such as a circle); if it is less than or equal to this value, it is considered irregular (such as a square). If the regularity of the area is greater than the preset value, the area is marked as a circle. This usually indicates high symmetry and uniform boundaries of the area. If the regularity of the area is less than or equal to the preset value, the area is marked as a square, indicating that it is relatively irregular but still has certain boundary features. If the area of ​​the area is less than the preset value, the area is marked as an arrow, which is usually used to indicate direction or emphasize certain areas.

[0096] For example, the image of the damaged part marked in the car insurance claim form is parsed to automatically determine the type of mark. The contour of the damaged part image is collected and parsed. Its aspect ratio, roundness and corner features are calculated. For example, if the aspect ratio of the contour of an area is close to 1 and the roundness is high, it is a circle. The regularity is calculated to be greater than the preset threshold, and the area is marked as a circular mark. If the regularity of another area is low and the boundary is square, it is marked as a square. If some small area protruding marks meet the characteristics of arrows, they are marked as arrows to indicate the start or direction of the damage. By automatically identifying different types of marks based on image features, human errors and subjective judgment deviations are reduced. Marks of different shapes can be used to visualize different information, which helps to understand and analyze the content in the document image.

[0097] The processing time is greatly reduced through automated geometric analysis, which is suitable for large-scale document image batch processing. Pre-identification of logo types can provide a clear starting point and conditions for subsequent processing (such as data extraction, analysis and pattern recognition). The technology is not limited to the field of insurance claims, but can also be extended to other application scenarios that require image recognition and analysis, such as map annotation, architectural plan recognition, etc.

[0098] The above embodiment provides an efficient and scalable method for identifying logo types by analyzing the geometric features of image contours, combined with preset regularity and area standards. This not only improves the accuracy of image analysis, but also provides a basis for information visualization and data mining, which helps to improve overall system performance and user experience.

[0099] In some embodiments, before completing the editing of the document image, it also includes: obtaining the shooting location information corresponding to the document image; detecting the claims-related area to obtain the element type corresponding to the claims-related area; generating watermark information according to the element type, shooting location information and insurance type information, adding the watermark information to the document image, and completing the editing of the document image.

[0100] Extract the location information of the document image, which is usually obtained through the metadata of the image (such as EXIF ​​information), including geographical information such as longitude and latitude. Analyze the image to detect the claim-related area and identify the element type in the area. This may involve using image recognition technology to determine different object or text types, such as license plates, damaged parts, etc. Generate specific watermark information based on the element type, shooting location information and insurance type information. The watermark can be in text or graphic form, and the content may include: Element type description (such as "license plate" or "damage picture") Accurate geographical location (such as city name, region, etc.) Related insurance type information (such as "auto insurance" or "health insurance") The generation of watermark information may include formatting and encoding steps to ensure its readability for end users and compatibility with machine processing. Add the generated watermark information to the document image. This step can be achieved through image processing tools to ensure that the watermark does not affect the readability of the main content. The layout, transparency and position of the watermark can be customized according to the specific application scenario to achieve the purpose of both anti-counterfeiting and not interfering with user reading.

[0101] For example, when an insurance company processes photo files for auto insurance claims, it ensures that the true source and content of the submitted images have not been tampered with. By extracting the shooting location from the car damage photos in the claim application, such as "Chaoyang District, Beijing". The main element in the image is detected as the license plate, and the insurance type information is confirmed to be auto insurance. Generate watermark information: "License Plate-Chaoyang District, Beijing-Auto Insurance". Add the watermark in a semi-transparent way to the lower right corner of the image to ensure that the key damaged parts are not blocked.

[0102] By adding watermarks containing location, time, and insurance type, false information or tampering can be prevented, and the authenticity and traceability of evidence can be improved. It is clear that the entry information can be directly read from the image, which is convenient for auditors to quickly confirm and process. Through clearly visible watermarks, users can increase their trust in the application and provide verification references for the system. At the same time, in case of disputes or doubts, the detailed information in the watermark provides additional evidence support, which helps reduce potential fraud risks. While simplifying the review and verification process, it ensures the reliability and quick retrieval of information.

[0103] Through the above embodiments, insurance companies or related organizations can introduce accurate metadata and element descriptions in each link of image information processing to prevent tampering, while improving the transparency and security of data processing. This not only has a significant effect on optimizing internal operating processes, but also provides strong guarantees for privacy protection, compliance requirements, etc.

[0104] In some embodiments, color is overlaid on each pixel in the privacy area according to the mosaic information, and the area outline is marked according to the identification type to complete the editing of the document image, including: obtaining an initialized preset container; converting the document image into a preset document object model format; setting the converted document image as a background image in the preset container; color is overlaid on each pixel in the privacy area corresponding to the background image in the preset container according to the mosaic information, and the area outline is marked according to the identification type to complete the editing of the background image, and uploading the edited background image to the claims system.

[0105] By creating a preset container for holding and processing images. This container can be a memory space in the image processing environment or a virtual canvas in which subsequent image processing operations are performed. Convert the document image to a format suitable for the document object model. This step involves converting traditional image data into a more structured format to facilitate flexible subsequent operations. The DOM format facilitates access to and manipulation of different image elements such as background images, foreground objects, and text in technical processes. In the preset container, set the converted document image as the background image. This is equivalent to laying a layer of base image on the canvas, providing a basis for subsequent overlay and marking operations. Use mosaic information to color overlay the pixels in the privacy area of ​​the background image. The specific operation usually involves applying blurring or pixelation to the privacy area to interfere with its clarity and make it unrecognizable. This step can protect privacy information without affecting the structural logic of the overall image.

[0106] Mark the relevant areas in the background image according to the logo type. Different logo types (such as circles, squares or arrows) determine different marking methods. The marking step may include drawing boundary lines or color filling to enhance the clarity of identification and information communication. Upload the edited background image to the relevant claims system for further review and processing.

[0107] For example, an insurance company needs to perform privacy processing on documents in auto insurance claims and mark the damage type or location. Create a preset processing container. Convert the vehicle damage document image to DOM format for easy operation. Set the image as the background image of the container. Apply mosaic processing to the license plate or personal information privacy area to perform color overlay to ensure that it is unrecognizable. Draw boundary marks for the damage location area on the image according to the damage type, such as marking the impact area with a red circle. After editing, upload it to the internal claims system.

[0108] Through color overlay and mosaic processing, the privacy information in the image is effectively protected, and the security of image data use is improved. Using different types of logos to significantly mark the area enhances the visualization of information and facilitates rapid communication and decision-making. At the same time, the use of automated processes reduces manual intervention and improves image processing efficiency, which is suitable for batch document processing scenarios. By uploading the processed images directly to the claims system, seamless docking and integration from image acquisition to review is achieved, improving overall operational efficiency.

[0109] The above-mentioned embodiments provide an efficient, safe and intelligent method for image management and review in the insurance and other industries through a reasonable image processing process combined with privacy protection and information labeling technology, greatly improving operational convenience and user trust.

[0110] The method obtains the document image to be edited from the preset claims system. According to the obtained insurance information, the document image is classified and the area related to the claim is identified. The claim-related area is automatically detected in the image. The area outside the claim-related area is identified as the area that may contain privacy information (i.e., the privacy area). Then the outline of the claim-related area is determined, and its area identification type is defined, and the area is marked with identification styles such as circle, square or arrow. The color distribution information in the privacy area is obtained, and the color of each pixel is analyzed. Based on this information, a mosaic effect is generated to conceal the privacy information. The generated mosaic information is applied to the privacy area, and each pixel is subjected to color superposition processing. The final edited document image contains the identified claim-related area, and the privacy information is effectively blocked. In summary, this method improves the image processing efficiency and privacy protection issues in the financial claims process through automated and intelligent image editing technology, making the overall operation smoother and safer.

[0111] The provided method has the following beneficial effects:

[0112] 1. Improve work efficiency: Automated area detection and labeling reduces manual operations and improves the efficiency of claims review and processing.

[0113] 2. User experience optimization: The operation process is simplified, the additional step of using image editing software is eliminated, and the operator's operating experience is significantly improved.

[0114] 3. Privacy protection: It effectively protects sensitive information that may be involved in the document image and ensures information security by automatically adding mosaics to the privacy area.

[0115] 4. Accuracy and consistency: Detect and mark relevant areas based on insurance information and preset rules, improving the accuracy and consistency of image processing.

[0116] 5. Flexibility and scalability: Providing multiple identification types makes this method flexible and able to adapt to different claims needs and insurance characteristics.

[0117] See also Figure 4 As shown, Figure 4 1 is a schematic diagram of the structure of an image editing device 200 provided in an embodiment of the present application. The image editing device 200 is used to execute the steps of the image editing method shown in the above embodiments. The image editing device 200 can be a single server or a server cluster, or the image editing device 200 can be a terminal, and the terminal can be a handheld terminal, a laptop computer, a wearable device, a robot, etc.

[0118] like Figure 4 As shown, the image editing device 200 includes:

[0119] The information acquisition unit 201 is used to acquire the document image to be edited and the insurance type information corresponding to the document image in a preset claim settlement system;

[0120] The area detection unit 202 is used to detect the claim-related area in the document image according to the insurance type information;

[0121] A privacy detection unit 203, for detecting a privacy area in other areas outside the claim settlement related area in the document image;

[0122] The identification determination unit 204 is used to determine the type of regional identification according to the regional contour corresponding to the claim-related area in the document image; wherein the regional identification type includes at least one or more of a text identification, a circular identification, a square identification, and an arrow identification;

[0123] The distribution determination unit 205 is used to obtain color distribution information corresponding to the privacy area in the document image; the color distribution information includes the color corresponding to each pixel in the privacy area;

[0124] The superposition determination unit 206 is used to generate mosaic information corresponding to the privacy area according to the color distribution information; the mosaic information includes the superposition color corresponding to each pixel in the privacy area;

[0125] The editing completion unit 207 is used to perform color superposition on each pixel in the privacy area according to the mosaic information, complete the marking of the area outline according to the identification type, and complete the editing of the document image.

[0126] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific working processes of the image editing device and each module described above can refer to the corresponding processes in the image editing method embodiments described in the above embodiments, and will not be repeated here.

[0127] The above-mentioned image editing method can be implemented in the form of a computer program. Figure 4 Run on the device shown.

[0128] See also Figure 5 , Figure 5 1 is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application. The computer device includes a processor, a memory and a network interface connected via a device bus, wherein the memory may include a storage medium and an internal memory.

[0129] The storage medium can store an operating device and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any image editing method.

[0130] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0131] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any image editing method.

[0132] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will appreciate that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the terminal to which the scheme of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0133] It should be understood that the processor may be a central processing unit (CPU), and the processor may 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 gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0134] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0135] Obtaining the document image to be edited and the insurance type information corresponding to the document image in a preset claim settlement system;

[0136] Detect claim-related areas in the document image based on the insurance type information;

[0137] Detecting privacy areas in areas other than claim-related areas in the document image;

[0138] Determine the type of area identification according to the area outline corresponding to the claim-related area in the document image; wherein the type of area identification includes at least one or more of a circular identification, a square identification, and an arrow identification;

[0139] Obtaining color distribution information corresponding to the privacy area in the document image; the color distribution information includes the color corresponding to each pixel in the privacy area;

[0140] Generate mosaic information corresponding to the privacy area according to the color distribution information; the mosaic information includes the superimposed color corresponding to each pixel in the privacy area;

[0141] Color is overlaid on each pixel in the privacy area according to the mosaic information, the area outline is marked according to the identification type, and the document image is edited.

[0142] In some embodiments, a claim-related area is detected in a document image based on insurance type information, including: obtaining a pre-built image recognition model library, the image recognition model library including multiple image recognition models and a recognition category corresponding to each image recognition model; matching the insurance type information with the recognition category, and obtaining an image recognition model corresponding to a successfully matched recognition category; inputting the document image into the image recognition model, the image recognition model parsing the document image, and outputting the claim-related area.

[0143] Exemplarily, before obtaining the pre-built image recognition model library, it also includes: obtaining multiple historical document images of claims that have been settled and the insurance type information and historical claim area corresponding to each historical document image in the claims system; classifying the multiple historical document images according to the insurance type information, and obtaining multiple historical image groups, each historical image group including at least one historical document image; generating multiple image recognition models to be trained according to the number of historical image groups; inputting the historical document images of each historical image group into an image recognition model respectively, and the image recognition model parses each historical document image and outputs the claim prediction area corresponding to each historical document image; completing the training of each corresponding image recognition model according to the claim prediction area and historical claim area corresponding to each historical document image; and generating the recognition category of the corresponding image recognition model according to the insurance type information corresponding to each historical image group.

[0144] In some embodiments, a privacy area is detected in other areas outside the claim-related area in a document image, including: performing optical character recognition in other areas outside the claim-related area to obtain text information corresponding to the document image and position information corresponding to the text information; parsing the text information based on natural language processing technology to obtain semantic information corresponding to the text information; matching the semantic information with a preset privacy type, and generating a privacy area based on the position information corresponding to the successfully matched semantic information.

[0145] In some embodiments, the area identification type is determined based on the area outline corresponding to the claim-related area in the document image, including: parsing the area outline, obtaining geometric feature information corresponding to the area outline; the geometric feature information includes at least one or more of aspect ratio, roundness and corner features; if the regularity corresponding to the area outline determined according to the geometric feature information is greater than the preset regularity, the identification type is determined to be a circular identification; if the regularity corresponding to the area outline determined according to the geometric feature information is less than or equal to the preset regularity, the identification type is determined to be a square identification; if the area corresponding to the area outline determined according to the geometric feature information is less than the preset area, the identification type is determined to be an arrow identification.

[0146] In some embodiments, before completing the editing of the document image, it also includes: obtaining the shooting location information corresponding to the document image; detecting the claims-related area to obtain the element type corresponding to the claims-related area; generating watermark information according to the element type, shooting location information and insurance type information, adding the watermark information to the document image, and completing the editing of the document image.

[0147] In some embodiments, color is overlaid on each pixel in the privacy area according to the mosaic information, and the area outline is marked according to the identification type to complete the editing of the document image, including: obtaining an initialized preset container; converting the document image into a preset document object model format; setting the converted document image as a background image in the preset container; color is overlaid on each pixel in the privacy area corresponding to the background image in the preset container according to the mosaic information, and the area outline is marked according to the identification type to complete the editing of the background image, and uploading the edited background image to the claims system.

[0148] A computer-readable storage medium is also provided in an embodiment of the present application, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and the processor executes the program instructions to implement the steps of the image editing method provided in the above embodiments of the present application.

[0149] The computer-readable storage medium may be an internal storage unit of the computer device described in the above embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device.

[0150] The above is only a specific implementation 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. An image editing method, characterized in that: include: Obtaining the document image to be edited and the insurance type information corresponding to the document image in a preset claims system; Detecting a claim-related area in the document image according to the insurance type information; Detecting a privacy area in other areas of the document image outside the claim settlement related area; Determine the type of area identification according to the area outline corresponding to the claim-related area in the document image; wherein the type of area identification includes at least one or more of a circular identification, a square identification, and an arrow identification; Acquire color distribution information corresponding to the privacy area in the document image; the color distribution information includes the color corresponding to each pixel in the privacy area; Generate mosaic information corresponding to the privacy area according to the color distribution information; the mosaic information includes the superimposed color corresponding to each pixel in the privacy area; Color is superimposed on each pixel in the privacy area according to the mosaic information, the outline of the area is marked according to the identification type, and the editing of the document image is completed.

2. The method according to claim 1, characterized in that: The detecting of a claim settlement related area in the document image according to the insurance type information includes: Obtain a pre-built image recognition model library, wherein the image recognition model library includes a plurality of image recognition models and a recognition category corresponding to each of the image recognition models; Matching the insurance type information with the identification category, and obtaining an image recognition model corresponding to the identification category that has been successfully matched; The document image is input into the image recognition model, and the image recognition model analyzes the document image and outputs the claim-related area.

3. The method according to claim 2, characterized in that Before obtaining the pre-built image recognition model library, the method further includes: Acquire, in the claim settlement system, a plurality of historical document images of claims settled and the insurance type information and historical claim settlement area corresponding to each of the historical document images; Classifying the plurality of historical document images according to the insurance type information to obtain a plurality of historical image groups, each of the historical image groups including at least one historical document image; generating a plurality of image recognition models to be trained according to the number of the historical image groups; Inputting the historical document images of each of the historical image groups into one of the image recognition models respectively, the image recognition model analyzing each of the historical document images and outputting the claim prediction area corresponding to each of the historical document images; Complete the training of each corresponding image recognition model according to the claim prediction area and the historical claim area corresponding to each historical document image; The recognition category of the corresponding image recognition model is generated according to the insurance type information corresponding to each historical image group.

4. The method according to claim 1, characterized in that The detecting of the privacy area in other areas outside the claim settlement related area in the document image includes: Perform optical character recognition in other areas outside the claim settlement related area to obtain text information corresponding to the document image and position information corresponding to the text information; Analyze the text information based on natural language processing technology to obtain semantic information corresponding to the text information; The semantic information is matched with a preset privacy type, and the privacy area is generated based on the location information corresponding to the successfully matched semantic information.

5. The method according to claim 1, characterized in that The determining of the area identification type according to the area contour corresponding to the claim settlement related area in the document image includes: Analyze the area outline to obtain geometric feature information corresponding to the area outline; the geometric feature information includes at least one or more of aspect ratio, roundness and corner features; If it is determined according to the geometric feature information that the regularity corresponding to the area outline is greater than a preset regularity, determining that the identification type is the circular identification; If it is determined according to the geometric feature information that the regularity corresponding to the area outline is less than or equal to a preset regularity, determining that the mark type is the square mark; If it is determined according to the geometric feature information that the area corresponding to the region outline is smaller than a preset area, the identification type is determined to be the arrow identification.

6. The method according to claim 1, characterized in that Before completing the editing of the document image, the method further includes: Obtaining shooting location information corresponding to the document image; Detecting the claim-related area to obtain the element type corresponding to the claim-related area; Watermark information is generated according to the element type, the shooting location information and the insurance type information, and the watermark information is added to the document image to complete the editing of the document image.

7. The method according to claim 1, characterized in that The step of superimposing colors on each pixel in the privacy area according to the mosaic information, marking the area outline according to the identification type, and editing the document image includes: Get the initialized preset container; Converting the document image into a preset document object model format; Setting the converted document image as a background image in the preset container; In the preset container, color is superimposed on each pixel in the privacy area corresponding to the background image according to the mosaic information, and the outline of the area corresponding to the background image is marked according to the identification type to complete the editing of the background image, and the edited background image is uploaded to the claims system.

8. An image editing device, characterized in that: include: An information acquisition unit, used to acquire the document image to be edited and the insurance type information corresponding to the document image in a preset claim settlement system; An area detection unit, used to detect a claim-related area in the document image according to the insurance type information; A privacy detection unit, configured to detect a privacy area in other areas outside the claim settlement related area in the document image; An identification determination unit, configured to determine a region identification type according to a region outline corresponding to the claim settlement related region in the document image; wherein the region identification type includes at least one or more of a text identification, a circular identification, a square identification, and an arrow identification; A distribution determination unit, configured to obtain color distribution information corresponding to the privacy area in the document image; the color distribution information includes a color corresponding to each pixel in the privacy area; A superposition determination unit, configured to generate mosaic information corresponding to the privacy area according to the color distribution information; the mosaic information includes a superposition color corresponding to each pixel in the privacy area; The editing completion unit is used to perform color superposition on each pixel in the privacy area according to the mosaic information, complete the marking of the area outline according to the identification type, and complete the editing of the document image.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and implement the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to implement the method according to any one of claims 1 to 7.

Images