Vehicle service store vehicle receiving management method and device based on AI and VR, and electronic equipment

Abstract

The invention discloses an AI and VR-based vehicle service store vehicle receiving management method and device and electronic equipment, and relates to the field of data processing. The method comprises the following steps: acquiring a vehicle image collected by a camera; identifying the vehicle image to obtain vehicle basic information and appearance damage information; determining a target maintenance case corresponding to the to-be-maintained vehicle from a preset historical maintenance case library; the target maintenance case is sent to maintenance personnel, so that the maintenance personnel can maintain the to-be-maintained vehicle; generating a VR maintenance animation based on the vehicle interior image and the vehicle exterior image; and according to the target maintenance case, generating a maintenance operation instruction text and a voice explanation, and synchronously presenting the maintenance operation instruction text and the voice explanation in the VR maintenance animation to obtain a target VR maintenance animation, so that a user can check the vehicle state through the target VR maintenance animation. By implementing the technical scheme provided by the invention, the experience feeling of the customer of the automobile repair shop is improved.

Description

Technical Field

[0001] This application relates to the field of data processing, and particularly to a vehicle reception management method, device, and electronic device for auto repair shops based on AI and VR. Background Art

[0002] In recent years, the continuous increase in the number of motor vehicles has brought about the rapid development of the auto repair industry. As consumers' expectations for the quality and efficiency of auto repair services continue to rise, traditional auto repair shops are facing challenges in improving service efficiency and customer satisfaction. Against this backdrop, it is particularly important to utilize high-tech to improve and innovate auto repair services. Especially the rapid development of artificial intelligence (AI) and virtual reality (VR) technologies has brought new opportunities to the auto repair industry, making repair services more intelligent, visual, and personalized.

[0003] Currently, many auto repair shops still rely on traditional manual vehicle reception and repair management methods, which are often inefficient, lack transparency, and cannot provide sufficient customer interaction and participation. For example, it is difficult for customers to obtain real-time vehicle repair status and detailed information, and the repair process lacks visual support, resulting in a low customer experience.

[0004] Therefore, there is an urgent need for a vehicle reception management method, device, and electronic device for auto repair shops based on AI and VR. Summary of the Invention

[0005] This application provides a vehicle reception management method, device, and electronic device for auto repair shops based on AI and VR, which improves the customer experience of auto repair shops.

[0006] In the first aspect of this application, a vehicle reception management method for auto repair shops based on AI and VR is provided. The method includes: when a vehicle to be repaired enters an auto repair shop, obtaining a vehicle image collected by a camera; identifying the vehicle image to obtain vehicle basic information and appearance damage information, where the vehicle basic information includes the license plate number and vehicle model, and the appearance damage information includes the fault location; based on the vehicle basic information and the appearance damage information, determining a target repair case corresponding to the vehicle to be repaired from a preset historical repair case library; sending the target repair case to the repair personnel for the repair personnel to repair the vehicle to be repaired; during the repair process, obtaining vehicle interior and exterior images, where the vehicle interior and exterior images include vehicle interior images and vehicle exterior images; generating a VR repair animation based on the vehicle interior images and the vehicle exterior images; generating a repair operation instruction text and voice commentary according to the target repair case, and synchronously presenting the repair operation instruction text and voice commentary to the VR repair animation to obtain a target VR repair animation, so that the user can view the vehicle status through the target VR repair animation.

[0007] By adopting the above technical solutions, by obtaining the image of the vehicle to be repaired and identifying the image, the basic vehicle information and appearance damage information are automatically obtained, reducing the operation of manual information entry and improving the vehicle receiving efficiency and accuracy. At the same time, based on the identified vehicle information, the most relevant target repair case is intelligently matched from the historical repair case database, providing a reference standard solution for the maintenance personnel, reducing the uncertainty and decision-making difficulty in the repair process, and ensuring the repair quality and consistency. In addition, the method also automatically generates vivid VR repair animations through the collection and analysis of the images inside and outside the vehicle, and superimposes the text descriptions and voice commentaries of the repair steps, making the repair process more intuitive. Finally, users can also intuitively understand the repair situation of the vehicle to be repaired through the VR animation, improving the transparency and credibility of the repair service and enhancing the user experience.

[0008] Optionally, the preset historical repair case database includes multiple repair cases. The determining of the target repair case corresponding to the vehicle to be repaired from the preset historical repair case database based on the basic vehicle information and the appearance damage information specifically includes: constructing a vehicle fault feature vector based on the basic vehicle information and the appearance damage information; extracting keywords from each of the repair cases to obtain a plurality of keywords, and constructing corresponding repair case feature vectors based on the plurality of keywords; calculating the similarity between the vehicle fault feature vector and each of the repair case feature vectors, and determining the target repair case according to the similarity, where the target repair case is the repair case corresponding to the maximum similarity among the plurality of similarities.

[0009] By adopting the above technical solutions, when determining the target repair case, a matching algorithm based on the similarity of feature vectors is adopted, making the matching result more accurate and reliable. First, the method constructs a vehicle fault feature vector representing the fault characteristics according to the identified vehicle fault information. Then, keywords are extracted from the historical repair cases and vectorized feature representation is performed. By calculating the similarity between the fault feature vector and each of the repair case feature vectors, the most similar repair case is found as the recommended result. This quantitative matching method synthesizes multiple key attributes of the vehicle fault, reducing the limitations brought by matching based on a single attribute. At the same time, the feature representation based on keywords highlights the core content of the repair case, reduces the interference of irrelevant information, and improves the pertinence of the matching.

[0010] Optionally, generating a VR maintenance animation based on the vehicle interior image and the vehicle exterior image specifically includes: extracting key frames from the vehicle interior image and the vehicle exterior image respectively to obtain key step images under different perspectives; segmenting the key step images to extract various components of the vehicle to be inspected in the key step images; generating a three-dimensional scene model corresponding to the vehicle to be inspected based on each of the components; and inputting the three-dimensional scene model into a VR engine for rendering to obtain the VR maintenance animation.

[0011] By adopting the above technical solutions, when generating VR maintenance animations, key frame extraction, image segmentation, 3D modeling, VR rendering and other technologies are used, making the production process of VR animations more automated and intelligent, and the generated VR scenes more realistic and immersive. First, by extracting key frames from images inside and outside the vehicle, the key steps in the maintenance process are automatically screened out, reducing data redundancy and highlighting core information. Then, the image segmentation technology is used to accurately identify the vehicle components in the key step images, providing important clues for subsequent 3D reconstruction. Based on the segmented component information and the preset 3D model, this method constructs a 3D scene model consistent with the real vehicle, and then uses a professional VR engine to render the virtual scene to generate a VR maintenance animation. When watching the animation in a VR device, users can observe the internal and external structure of the vehicle, as well as the details of the disassembly and assembly of parts during the maintenance process, and obtain an immersive maintenance experience.

[0012] Optionally, the segmentation of the key step image to extract various components of the vehicle to be inspected in the key step image specifically includes: performing edge detection on the key step image to obtain an edge contour map of the key step image; performing region segmentation on the key step image based on the edge contour map to obtain a segmentation result map of the key step image, wherein different regions in the segmentation result map correspond to different components of the vehicle to be inspected; and identifying each region in the segmentation result map according to a preset component model to obtain each of the components.

[0013] By adopting the above technical solution, when segmenting the key step image, the contour information of the vehicle parts is first obtained through edge detection, and then the key step image is segmented according to the edge contour map, and finally the segmented area is identified based on the preset component model, thereby achieving accurate positioning and extraction of vehicle parts, laying a solid foundation for subsequent 3D modeling. The preset 3D component model is used to match and identify the segmented area, eliminating the interference of other irrelevant objects, and further improving the accuracy and completeness of vehicle component extraction.

[0014] Optionally, generating the three-dimensional scene model corresponding to the vehicle to be repaired based on each of the components specifically includes: obtaining the three-dimensional dimension information of each of the components; constructing a three-dimensional skeleton model of the vehicle to be repaired based on the three-dimensional dimension information of each of the components; performing texture mapping on the surfaces of each of the components to generate three-dimensional solid models of each of the components; and attaching the three-dimensional solid models of each of the components to corresponding positions on the three-dimensional skeleton model to obtain the complete three-dimensional scene model.

[0015] By adopting the above technical solution, when constructing the three-dimensional scene model of the vehicle, first obtain the three-dimensional dimension information of the segmented vehicle components, then construct the three-dimensional skeleton model of the vehicle to be repaired based on the three-dimensional dimension information of each component, then perform texture mapping on each component to generate a three-dimensional solid model, and finally accurately attach the three-dimensional solid model of the component to the three-dimensional skeleton model, thereby generating the three-dimensional scene model, providing high-quality three-dimensional materials for the production of VR repair animations.

[0016] Optionally, generating the maintenance operation instruction text and voice commentary according to the target maintenance case specifically includes: extracting key information from the target maintenance case, where the key information includes maintenance steps and the maintenance components and maintenance tools corresponding to each of the maintenance steps; filling the key information into a preset maintenance template to obtain the maintenance operation instruction text; and using a preset text-to-speech model to convert the maintenance operation instruction text into the voice commentary.

[0017] By adopting the above technical solution, when generating the maintenance operation instruction text and voice commentary, first automatically extract key information such as maintenance steps, maintenance components, and maintenance tools from the target maintenance case, ensuring the integrity and accuracy of information extraction, reducing the tediousness and error rate of manual entry. Then, automatically fill the extracted key information into a preset maintenance template to generate a standardized and normalized maintenance operation text. At the same time, this method also uses text-to-speech technology to automatically convert the text content into smooth and natural voice commentary, saving the time and cost of manual dubbing. The converted voice is completely synchronized with the original text in terms of content and timeline, facilitating synchronous display with the text in the VR animation.

[0018] Optionally, the synchronously presenting the maintenance operation instruction text and the voice commentary in the VR maintenance animation specifically includes: identifying and marking each animation scene in the VR maintenance animation to obtain animation scene identification information, wherein the animation scene identification information includes the name and time period of each animation scene; dividing the maintenance operation instruction text into multiple text paragraphs, each text paragraph corresponding to a maintenance operation step; matching the name of each animation scene with the maintenance operation step, and determining the display time point of each text paragraph in the VR maintenance animation; dividing the voice commentary into multiple audio clips, each audio clip corresponding to a maintenance operation step; and based on the display time point, superimposing each of the text paragraphs and each of the audio clips in the VR maintenance animation to obtain the target VR maintenance animation.

[0019] By adopting the above technical solution, when the maintenance operation instruction text and voice commentary are synchronously presented in the VR maintenance animation, the VR animation is firstly subjected to scene recognition and timestamp marking, which provides a time positioning basis for the insertion of text and voice. Then, the maintenance operation instruction text is divided into several text paragraphs according to the maintenance steps, and the voice commentary is also divided into audio segments corresponding to the text paragraphs. By semantically matching the text paragraphs and the animation scenes, the optimal display position of each text paragraph on the animation timeline is automatically determined, ensuring the synchronization of the graphic information in time and content. Similarly, the voice segment is also inserted into the audio track of the VR animation at the same time point as the text paragraph to achieve audio and video synchronization. During the playback of the VR animation, the explanatory text appears synchronously with the picture in the form of subtitles or annotation boxes, and the voice commentary is also played synchronously with the picture audio track. Users can instantly see the text description of each operation step and hear the matching voice explanation while watching the VR animation, truly realizing the seamless connection of the text, image and audio in time and content.

[0020] In the second aspect of the present application, a vehicle repair shop vehicle reception management device based on AI and VR is provided. The device includes an acquisition module and a processing module, where: the acquisition module is used to acquire the vehicle image collected by the camera when the vehicle to be repaired enters the vehicle repair shop; the processing module is used to identify the vehicle image to obtain the basic vehicle information and appearance damage information. The basic vehicle information includes the license plate number and vehicle model, and the appearance damage information includes the fault location; the processing module is further used to determine the target repair case corresponding to the vehicle to be repaired from the preset historical repair case library based on the basic vehicle information and the appearance damage information; the processing module is further used to send the target repair case to the repair personnel so that the repair personnel can repair the vehicle to be repaired; the processing module is further used to acquire the internal and external vehicle images during the repair process. The internal and external vehicle images include the internal vehicle image and the external vehicle image; the processing module is further used to generate a VR repair animation based on the internal vehicle image and the external vehicle image; the processing module is further used to generate a repair operation instruction text and a voice commentary according to the target repair case, and synchronously present the repair operation instruction text and the voice commentary to the VR repair animation to obtain a target VR repair animation, so that the user can view the vehicle status through the target VR repair animation.

[0021] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method described in any one of the above.

[0022] In the fourth aspect of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores instructions that, when executed, execute the method described in any one of the above.

[0023] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By obtaining the images of the vehicle to be repaired and recognizing the images, the basic vehicle information and appearance damage information are automatically obtained, reducing the operation of manual information entry, improving the vehicle receiving efficiency and accuracy. At the same time, based on the recognized vehicle information, the most relevant target repair cases are intelligently matched from the historical repair case database, providing a reference standard solution for maintenance personnel, reducing the uncertainty and decision-making difficulty during the repair process, and ensuring the repair quality and consistency. In addition, this method also automatically generates vivid VR repair animations through the collection and analysis of in-vehicle and out-of-vehicle images, and superimposes text descriptions and voice commentaries of the repair steps, making the repair process more intuitive. Finally, users can also intuitively understand the repair situation of the vehicle to be repaired through the VR animation, improving the transparency and credibility of the repair service and enhancing the user experience.

[0024] 2. When determining the target repair case, a matching algorithm based on the similarity of feature vectors is adopted, making the matching result more accurate and reliable. First, according to the recognized vehicle fault information, this method constructs a vehicle fault feature vector representing the fault characteristics. Then, keyword extraction and feature vector representation are performed on historical repair cases. By calculating the similarity between the fault feature vector and the feature vectors of each repair case, the most similar repair case is found as the recommended result. This quantitative matching method synthesizes multiple key attributes of vehicle faults, reducing the limitations brought by matching based on a single attribute. At the same time, the feature representation based on keywords highlights the core content of the repair case, reducing the interference of irrelevant information and improving the pertinence of the matching.

[0025] 3. When generating the VR repair animation, technologies such as key frame extraction, image segmentation, 3D modeling, and VR rendering are adopted, making the production process of the VR animation more automated and intelligent, and the generated VR scene more realistic and immersive. First, by performing key frame extraction on the in-vehicle and out-of-vehicle images, the key steps during the repair process are automatically screened out, reducing data redundancy and highlighting the core information. Then, the image segmentation technology is used to accurately identify the vehicle components in the key step images, providing important clues for subsequent 3D reconstruction. Based on the segmented component information and the preset 3D model, this method constructs a 3D scene model consistent with the real vehicle, and then uses a professional VR engine to render the virtual scene to generate the VR repair animation. When users watch the animation in the VR device, they can observe the internal and external structures of the vehicle and the disassembly and assembly details of the components during the repair process as if they were on the spot, obtaining an immersive repair experience. Brief Description of the Drawings

[0026] Figure 1 is a schematic flowchart of the vehicle repair shop receiving management method based on AI and VR disclosed in the embodiments of the present application; Figure 2It is a schematic diagram of modules of a vehicle reception management device for auto repair shops based on AI and VR disclosed in an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application.

[0027] Explanation of reference numerals: 201, acquisition module; 202, processing module; 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Specific embodiments

[0028] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0029] In the description of the embodiments of the present application, words such as "for example" or "for illustration" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "for example" or "for illustration" in the embodiments of the present application should not be construed as being more preferred or more advantageous than other embodiments or design solutions. Exactly speaking, the use of words such as "for example" or "for illustration" is intended to present relevant concepts in a specific manner.

[0030] In the description of the embodiments of the present application, the meaning of the term "plurality" refers to two or more. For example, a plurality of systems refers to two or more systems, and a plurality of screen terminals refers to two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the technical features indicated. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The terms "include", "comprise", "have" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0031] The present application provides a vehicle reception management method for auto repair shops based on AI and VR. Refer to Figure 1 , Figure 1 It is a flowchart of the vehicle reception management method for auto repair shops based on AI and VR provided in an embodiment of the present application. This method is applied to a server. The server is a server that executes a vehicle reception management program for auto repair shops based on AI and VR. The server can be a single server, a server cluster composed of multiple servers, or a cloud computing service center. This method includes steps S101 to S107, and the above steps are as follows: Step S101: When the vehicle to be repaired enters the auto repair shop, obtain the vehicle image collected by the camera.

[0032] In step S101, when the vehicle to be repaired enters the auto repair shop, the server establishes a communication connection with the camera in the auto repair shop and obtains the vehicle image collected by the camera in real time. The camera is connected to the server through a wired or wireless network to transmit the collected image data to the server. When the vehicle to be repaired drives into the auto repair shop, the camera at the entrance will automatically detect the entry of the vehicle and trigger the image acquisition process. The camera continuously captures the vehicle at a certain frame rate (such as 30 frames per second) to obtain vehicle images at different angles and positions.

[0033] Step S102: Identify the vehicle image to obtain the basic vehicle information and appearance damage information. The basic vehicle information includes the license plate number and vehicle model, and the appearance damage information includes the fault location.

[0034] In step S102, the server identifies and analyzes the obtained vehicle image to extract the basic vehicle information and appearance damage information. First, the server inputs the obtained vehicle image into a pre-trained deep learning model, which consists of a convolutional neural network (CNN) and a recurrent neural network (RNN), and can effectively extract the feature information in the image. The CNN is responsible for extracting the local features of the image, such as license plates, logos, headlights, etc.; the RNN is responsible for analyzing the overall structure and context information of the image, such as the body contour, damage location, etc. For the identification of the basic vehicle information, the server identifies two aspects: the license plate and the vehicle model. When identifying the license plate, the server uses an image segmentation algorithm, such as the YOLO algorithm, to locate the position of the license plate in the vehicle image. Then, the license plate image is input into a character recognition model, which segments and recognizes the characters in the license plate image to obtain the text information of the license plate number. When identifying the vehicle model, the server uses a classification model based on a cascaded convolutional neural network to compare the vehicle image with a preset vehicle model image library, find the most similar vehicle model, and return the corresponding vehicle model name and model.

[0035] For the recognition of appearance damage information, the server mainly detects defects such as scratches, dents, and cracks on the vehicle body. The server uses a deep learning model based on semantic segmentation, such as Mask R-CNN, to divide the vehicle body image into different regions and perform defect detection on each region. The deep learning model based on semantic segmentation analyzes features such as texture, color, and shape of the regions to determine whether there is damage and marks the location and type of the damage. During the recognition process, the server comprehensively analyzes vehicle images from multiple angles and positions to improve the accuracy and reliability of the recognition. For example, when detecting scratches on the vehicle body, the server compares images of different parts of the vehicle body. Finally, the server stores the recognized license plate number, vehicle model, fault location and other information in a structured manner to generate a complete vehicle information report. This report includes basic vehicle information and appearance damage information, providing a basis for subsequent maintenance plan formulation and cost estimation.

[0036] Step S103: Based on the basic vehicle information and appearance damage information, determine the target maintenance case corresponding to the vehicle to be inspected from the preset historical maintenance case library.

[0037] In step S103, the preset historical maintenance case library includes multiple maintenance cases. Based on the basic vehicle information and appearance damage information, determining the target maintenance case corresponding to the vehicle to be inspected from the preset historical maintenance case library specifically includes: constructing a vehicle fault feature vector based on the basic vehicle information and appearance damage information; extracting keywords from each maintenance case to obtain multiple keywords, and constructing corresponding maintenance case feature vectors based on the multiple keywords; calculating the similarity between the vehicle fault feature vector and each maintenance case feature vector, and determining the target maintenance case according to the similarity. The target maintenance case is the maintenance case corresponding to the maximum similarity among the multiple similarities.

[0038] Specifically, the server extracts features and vectorizes the basic information of the vehicle (such as license plate number, vehicle model, etc.) and appearance damage information (such as fault location, damage degree) to construct a multi-dimensional vehicle fault feature vector. This vector synthesizes various key attributes and damage characteristics of the vehicle and can comprehensively reflect the fault condition of the vehicle. Then, the server performs text mining and keyword extraction on each maintenance case in the preset historical maintenance case library. Through natural language processing techniques, such as the TF-IDF algorithm or the Word2Vec model, the server can automatically identify the keywords that best represent the characteristics of the case from the text description of the maintenance case, such as the name of the faulty component, the fault phenomenon, and the maintenance process. The server vectorizes the extracted keywords to generate a maintenance case feature vector, which reflects the core content and characteristics of the maintenance case.

[0039] Next, the server calculates the similarity between the vehicle fault feature vector and each maintenance case feature vector to find the target maintenance case that best matches the vehicle to be repaired. The calculation of similarity can use measurement methods such as cosine similarity and Euclidean distance. The higher the value, the more similar the two feature vectors are, and the closer the corresponding vehicle fault conditions and maintenance requirements are.

[0040] For example, if the vehicle to be repaired is an old BMW X3 that is 2 years old and has problems such as the engine fault light being on and power loss, its vehicle fault feature vector may contain elements such as "BMW", "X3", "2 years", "engine fault light on", and "power loss". The server matches this vehicle fault feature vector with the maintenance case feature vectors in the historical maintenance case database and finds that the feature vector of a case contains elements such as "BMW", "X3", "3 years", "engine fault light on", and "throttle fault". The similarity between the two is very high, so this case is determined as the target maintenance case.

[0041] Finally, the server selects the maintenance case with the highest similarity from the matching results as the final target maintenance case. This case usually contains the fault conditions and maintenance solutions that are most similar to the vehicle to be repaired and can provide the most direct and reliable reference for subsequent maintenance work.

[0042] Step S104: Send the target maintenance case to the maintenance personnel so that the maintenance personnel can repair the vehicle to be repaired.

[0043] In step S104, the server sends the selected target maintenance case to the maintenance personnel to provide them with maintenance reference and guidance. First, the server establishes a communication connection with the maintenance personnel's work terminals (such as tablets, mobile phones, etc.) through the local area network or the Internet within the store. The server packs the detailed information of the target maintenance case into a structured data file, such as JSON or XML format, and transmits the file to the terminal device of the maintenance personnel through a network protocol (such as HTTP, FTP, etc.).

[0044] For example, if the target maintenance case is a BMW X3 with an engine fault, the case information may include the following: Vehicle information: BMW X3, 2018 model, 2.0T, 50,000 kilometers; Fault phenomenon: Engine fault light on, weak acceleration, idle shake; Diagnostic result: Throttle blockage, need to clean the throttle and intake duct; Maintenance plan: 1. Disassemble the intake pipe; 2. Disassemble the throttle assembly; 3. Clean the throttle and intake duct; 4. Replace the air filter element; 5. Reinstall the throttle and intake pipe; 6. Clear the fault code and verify the maintenance effect; Required tools: Star screwdriver, 10mm socket wrench, throttle cleaner, compressed air gun, etc.; Estimated working hours: 2 hours.

[0045] Maintenance personnel can refer to the target maintenance case and formulate a specific maintenance plan based on the fault manifestations of the actual vehicle. During the maintenance process, maintenance personnel can also consult the case information at any time to obtain guidance and tips.

[0046] Step S105: During the maintenance process, obtain images inside and outside the vehicle. The images inside and outside the vehicle include images inside the vehicle and images outside the vehicle.

[0047] In step S105, the server obtains the image data during the vehicle maintenance process through the cameras inside and outside the vehicle. First, the auto repair shop installs multiple high-definition cameras in the vehicle maintenance area, aiming at the inside and outside of the vehicle respectively. These cameras are connected to the server through wired or wireless networks and can transmit the collected image data (images inside the vehicle and images outside the vehicle) to the server in real time.

[0048] Step S106: Generate a VR maintenance animation based on the images inside and outside the vehicle.

[0049] In step S106, generating a VR maintenance animation based on the images inside and outside the vehicle specifically includes: extracting key frames from the images inside and outside the vehicle respectively to obtain key step images from different perspectives; segmenting the key step images to extract each component of the vehicle to be repaired in the key step images; generating a three-dimensional scene model corresponding to the vehicle to be repaired based on each component; and inputting the three-dimensional scene model into the VR engine for rendering to obtain the VR maintenance animation.

[0050] Specifically, the server extracts key frames from the internal vehicle images and external vehicle images. Through motion detection algorithms, the server can automatically identify the key actions and nodes in the internal vehicle images and external vehicle images, and extract the key step images that best represent each maintenance step. Next, the server performs semantic segmentation on the extracted key step images. Based on deep learning algorithms such as Mask R-CNN or DeepLab, etc., the server can accurately identify and segment each component of the vehicle to be repaired in the key step images, such as the body, tires, etc. After obtaining the segmentation information of the vehicle components, the server uses 3D reconstruction technology, such as SFM (Structure from Motion), to convert the two-dimensional image information into a three-dimensional scene model. The server first estimates the three-dimensional shape and spatial coordinates of each component according to the feature points and camera poses in the image; then, combines the three-dimensional models of each component according to the actual assembly relationship to construct a complete three-dimensional scene model. Finally, the server imports the generated three-dimensional vehicle scene model into the VR engine for rendering and interactive design. Through mainstream VR development platforms such as Unity, Unreal, etc., the server can add realistic materials, lighting, shadows and other visual effects to the three-dimensional model to create an immersive maintenance scene.

[0051] In a possible implementation manner, the key step images are segmented to extract each component of the vehicle to be repaired in the key step images, specifically including: performing edge detection on the key step images to obtain an edge contour map of the key step images; performing region segmentation on the key step images based on the edge contour map to obtain a segmentation result map of the key step images, where different regions in the segmentation result map correspond to different components of the vehicle to be repaired; identifying each region in the segmentation result map according to a preset component model to obtain each component.

[0052] Specifically, the server performs edge detection on the key step images. Edge detection can be used to identify the contours and boundaries of objects in the image. The server can use various classic edge detection algorithms, such as the Canny algorithm, Sobel algorithm, etc., or can also adopt edge detection methods based on deep learning, such as DeepEdge, RCF, etc. Through edge detection, the server can extract the edge contours of the vehicle components from the image to generate an edge contour map. The white pixels in this image represent the edges of the objects, and the black pixels represent the background or internal regions. Then, the server performs region segmentation on the original key step images based on the edge contour map. Region segmentation is to divide the image into several non-overlapping regions, so that the pixel features within each region are similar, while the pixel features between different regions are quite different. The server can use region segmentation algorithms, such as region growing method, watershed algorithm, etc. for segmentation, and the result of the segmentation is a segmentation result map.

[0053] Finally, the server identifies and labels each region in the segmentation result map according to a preset component model. The preset component model is a database containing prior knowledge such as the shape, size, and position of each vehicle component, which can be constructed by manual annotation or machine learning. The server matches each region in the segmentation result map with each component in the preset component model, finds the most similar component category, and labels it on the corresponding region. For example, the server may identify a rectangular region as a "door" and a circular region as a "tire", etc. Through component identification, the server can transform the segmentation result map into a structured vehicle component information, including attributes such as the category, position, and size of each component.

[0054] In a possible implementation manner, a three-dimensional scene model corresponding to the vehicle to be repaired is generated based on each component, specifically including: obtaining the three-dimensional size information of each component; constructing a three-dimensional skeleton model of the vehicle to be repaired based on the three-dimensional size information of each component; performing texture mapping on the surface of each component to generate a three-dimensional solid model of each component; attaching the three-dimensional solid models of each component to the corresponding positions on the three-dimensional skeleton model to obtain a three-dimensional scene model.

[0055] Specifically, the server obtains the three-dimensional size information of each vehicle component. These size information can come from multiple channels, such as CAD design drawings provided by vehicle manufacturers and point cloud data collected by three-dimensional scanners. The server fuses and optimizes the size data to generate a three-dimensional size model of each component. The three-dimensional size model parametrically describes the geometric shape and size of each component, such as length, width, and height, providing an accurate reference basis for subsequent three-dimensional modeling.

[0056] Next, the server constructs a three-dimensional skeleton model of the vehicle to be repaired based on the three-dimensional size information of the components. The skeleton model is a vehicle structure framework composed of simple geometric elements (such as line segments, polygons, curved surfaces, etc.), which describes the positions, orientations, connection relationships, etc. of the main components such as the body, chassis, and suspension. Then, the server performs texture mapping on the surfaces of each component to generate a three-dimensional solid model. Texture mapping is a technology that maps a two-dimensional image onto the surface of a three-dimensional model, which can add appearance attributes such as real materials, colors, and details to the model. The server first extracts the texture images of each component from the vehicle images, such as the painted color of the body and the transparent glass of the headlights. Then, the server uses methods such as UV mapping to map these texture images onto the surfaces of the three-dimensional models of the corresponding components, and performs rendering processes such as sampling, filtering, and lighting calculation. Finally, the server attaches the three-dimensional solid models of each component to the vehicle skeleton model to form a complete three-dimensional scene model of the whole vehicle. The server performs geometric transformations such as translation, rotation, and scaling according to the spatial coordinates and orientation information of each component in the skeleton model, and attaches the three-dimensional solid models of each component to the corresponding positions on the three-dimensional skeleton model.

[0057] S107: Generate a repair operation instruction text and a voice commentary according to the target repair case, and synchronously present the repair operation instruction text and the voice commentary in the VR repair animation to obtain the target VR repair animation, so that the user can view the vehicle status through the target VR repair animation.

[0058] In step S107, generating a repair operation instruction text and a voice commentary according to the target repair case specifically includes: extracting key information from the target repair case, where the key information includes repair steps and the repair components and repair tools corresponding to each repair step; filling the key information into a preset repair template to obtain the repair operation instruction text; and using a preset text-to-speech model to convert the repair operation instruction text into a voice commentary.

[0059] Specifically, the server extracts key information from the target maintenance case, including the specific operations of each maintenance step. The server can use natural language processing techniques to automatically identify and extract this key information from the text description of the maintenance case. Next, the server fills the extracted key information into a preset maintenance template to generate a structured and standardized maintenance operation instruction text. The maintenance template is a text framework with a fixed format that contains the standard description method for maintenance steps. The server fills and replaces the operations of each maintenance step according to the syntax rules and placeholders in the preset maintenance template to form the maintenance operation instruction text. For example, "Remove the xxx component from the xxx position, loosen the xxx bolt with the xxx tool, remove the old xxx, install the new xxx, pay attention to xxx". Then, the server uses a preset text-to-speech model to automatically convert the generated maintenance operation instruction text into a voice commentary.

[0060] In step S107, the maintenance operation instruction text and the voice commentary are synchronously presented in the VR maintenance animation, specifically including: identifying and labeling each animation scene in the VR maintenance animation to obtain animation scene identification information, where the animation scene identification information includes the name and time period of each animation scene; dividing the maintenance operation instruction text into multiple text paragraphs, with each text paragraph corresponding to a maintenance operation step; matching the names of each animation scene with the maintenance operation steps to determine the display time points of each text paragraph in the VR maintenance animation; dividing the voice commentary into multiple audio segments, with each audio segment corresponding to a maintenance operation step; and based on the display time points, overlaying each text paragraph and each audio segment in the VR maintenance animation to obtain the target VR maintenance animation.

[0061] Specifically, the server precisely synchronizes the generated maintenance operation instruction text and the voice commentary with the VR maintenance animation to obtain the target VR maintenance animation. The server first identifies and labels each scene in the VR maintenance animation, such as "Open the engine hood", "Remove the oil filter", etc., and records the start and end time points of each scene. Then, the server divides the instruction text into multiple text paragraphs according to the maintenance steps, divides the voice commentary into corresponding audio segments, and matches them one by one with the scenes in the VR animation. The server calculates the precise display time of each text paragraph and audio segment in the animation based on the start and end time points of each scene, and inserts them as subtitles and voiceovers into the corresponding positions in the VR animation. Through timeline synchronization, when the user watches the VR maintenance animation, they can also timely see the corresponding text instructions and hear the oral commentary, greatly improving the information transmission efficiency and learning experience of the VR animation.

[0062] Refer to Figure 2, this application also provides a vehicle repair shop pick-up management device based on AI and VR. The device is a server, and the server includes an acquisition module 201 and a processing module 202. The acquisition module 201 and the processing module 202 are as follows: The acquisition module 201 is used to obtain the vehicle image collected by the camera when the vehicle to be repaired enters the vehicle repair shop; The processing module 202 is used to identify the vehicle image to obtain the vehicle basic information and the appearance damage information. The vehicle basic information includes the license plate number and the vehicle model, and the appearance damage information includes the fault location; The processing module 202 is also used to determine the target repair case corresponding to the vehicle to be repaired from the preset historical repair case library based on the vehicle basic information and the appearance damage information; The processing module 202 is also used to send the target repair case to the repair personnel so that the repair personnel can repair the vehicle to be repaired; The processing module 202 is also used to obtain the in-vehicle and out-of-vehicle images during the repair process. The in-vehicle and out-of-vehicle images include the vehicle internal image and the vehicle external image; The processing module 202 is also used to generate a VR repair animation based on the vehicle internal image and the vehicle external image; The processing module 202 is also used to generate a repair operation instruction text and a voice commentary according to the target repair case, and synchronously present the repair operation instruction text and the voice commentary to the VR repair animation to obtain the target VR repair animation, so that the user can view the vehicle status through the target VR repair animation.

[0063] In a possible implementation manner, the preset historical repair case library includes multiple repair cases. The processing module 202 determines the target repair case corresponding to the vehicle to be repaired from the preset historical repair case library based on the vehicle basic information and the appearance damage information, specifically including: The processing module 202 constructs a vehicle fault feature vector based on the vehicle basic information and the appearance damage information; The processing module 202 extracts keywords from each repair case to obtain multiple keywords, and constructs a corresponding repair case feature vector based on the multiple keywords; The processing module 202 calculates the similarity between the vehicle fault feature vector and each repair case feature vector, and determines the target repair case according to the similarity. The target repair case is the repair case corresponding to the maximum similarity among the multiple similarities.

[0064] In a possible implementation manner, the processing module 202 generates a VR repair animation based on the vehicle internal image and the vehicle external image, specifically including: The processing module 202 extracts key frames from the vehicle internal image and the vehicle external image respectively to obtain key step images from different perspectives; The processing module 202 segments the key step images and extracts each component of the vehicle to be repaired in the key step images; The processing module 202 generates a three-dimensional scene model corresponding to the vehicle to be repaired based on each component; The processing module 202 inputs the three-dimensional scene model into the VR engine for rendering to obtain the VR repair animation.

[0065] In a possible implementation, the processing module 202 segments the key-step image and extracts each component of the vehicle to be repaired in the key-step image. Specifically, the processing module 202 performs edge detection on the key-step image to obtain an edge contour map of the key-step image; the processing module 202 performs region segmentation on the key-step image based on the edge contour map to obtain a segmentation result map of the key-step image, where different regions in the segmentation result map correspond to different components of the vehicle to be repaired; the processing module 202 identifies each region in the segmentation result map according to a preset component model to obtain each component.

[0066] In a possible implementation, the processing module 202 generates a three-dimensional scene model corresponding to the vehicle to be repaired based on each component. Specifically, the acquisition module 201 acquires the three-dimensional dimension information of each component; the processing module 202 constructs a three-dimensional skeleton model of the vehicle to be repaired based on the three-dimensional dimension information of each component; the processing module 202 performs texture mapping on the surface of each component to generate a three-dimensional solid model of each component; the processing module 202 attaches the three-dimensional solid models of each component to the corresponding positions on the three-dimensional skeleton model to obtain a complete three-dimensional scene model.

[0067] In a possible implementation, the processing module 202 generates a maintenance operation instruction text and a voice commentary according to a target maintenance case. Specifically, the processing module 202 extracts key information from the target maintenance case, where the key information includes maintenance steps and the corresponding maintenance components and maintenance tools for each maintenance step; the processing module 202 fills the key information into a preset maintenance template to obtain a maintenance operation instruction text; the processing module 202 uses a preset text-to-speech model to convert the maintenance operation instruction text into a voice commentary.

[0068] In a possible implementation, the processing module 202 synchronously presents the maintenance operation instruction text and the voice commentary in the VR maintenance animation. Specifically, the processing module 202 identifies and labels each animation scene in the VR maintenance animation to obtain animation scene identification information, where the animation scene identification information includes the name and time period of each animation scene; the processing module 202 divides the maintenance operation instruction text into multiple text paragraphs, and each text paragraph corresponds to a maintenance operation step; the processing module 202 matches the name of each animation scene with the maintenance operation step to determine the display time points of each text paragraph in the VR maintenance animation; the processing module 202 divides the voice commentary into multiple audio segments, and each audio segment corresponds to a maintenance operation step; the processing module 202 superimposes each text paragraph and each audio segment in the VR maintenance animation based on the display time points to obtain a target VR maintenance animation.

[0069] It should be noted that: when the device provided in the above embodiment realizes its functions, only the division of the above functional modules is used for illustration. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiments belong to the same concept. For the specific implementation process, please refer to the method embodiment, which will not be elaborated here.

[0070] The present application also provides an electronic device. Referring to Figure 3 , Figure 3 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 300 may include: at least one processor 301, at least one network interface 304, a user interface 303, a memory 305, and at least one communication bus 302.

[0071] Among them, the communication bus 302 is used to realize the connection and communication between these components.

[0072] Among them, the user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may further include a standard wired interface and a wireless interface.

[0073] Among them, the network interface 304 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0074] Among them, the processor 301 may include one or more processing cores. The processor 301 connects various parts within the entire server through various interfaces and lines, and executes various functions of the server and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by invoking the data stored in the memory 305. Optionally, the processor 301 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 301 may integrate a combination of one or more of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301 and may be implemented separately by a single chip.

[0075] Among them, the memory 305 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store the data involved in the above-mentioned various method embodiments. Optionally, the memory 305 may also be at least one storage device located far from the aforementioned processor 301. Refer to Figure 3 , as a computer storage medium, the memory 305 may include an operating system, a network communication module, a user interface module, and an application program for the AI and VR-based vehicle reception management method for auto repair shops.

[0076] In Figure 3In the electronic device 300 shown, the user interface 303 is mainly used to provide an interface for the user to input and obtain the data input by the user; while the processor 301 can be used to call the application program stored in the memory 305 and based on the AI and VR vehicle reception management method for auto repair shops. When executed by one or more processors 301, the electronic device 300 executes one or more of the methods as described in the above embodiments. It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0077] The present application also provides a computer-readable storage medium storing instructions. When executed by one or more processors 301, the electronic device 300 executes one or more of the methods as described in the above embodiments.

[0078] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0079] In several implementation manners provided by the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some service interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical or other form.

[0080] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0081] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0082] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned memory includes: various media such as USB flash drives, mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0083] The above are only exemplary embodiments of the present disclosure, and the scope of the present disclosure cannot be limited thereby. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. After considering the specification and the disclosure of the practical truth, those skilled in the art will easily think of other implementation manners of the present disclosure.

[0084] The present application aims to cover any variations, uses, or adaptive changes of the present disclosure. These variations, uses, or adaptive changes follow the general principles of the present disclosure and include the common general knowledge or conventional technical means in the technical field not recorded in the present disclosure. The specification and the embodiments are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A vehicle reception management method for auto repair shops based on AI and VR, characterized in that, The method includes: When the vehicle to be repaired enters the auto repair shop, obtain the vehicle image collected by the camera; Identify the vehicle image to obtain the basic vehicle information and appearance damage information. The basic vehicle information includes the license plate number and vehicle model, and the appearance damage information includes the fault location; Based on the basic vehicle information and the appearance damage information, determine the target repair case corresponding to the vehicle to be repaired from the preset historical repair case library; Send the target repair case to the repair personnel so that the repair personnel can repair the vehicle to be repaired; During the repair process, obtain the internal and external images of the vehicle. The internal and external images of the vehicle include the internal image of the vehicle and the external image of the vehicle; Generate a VR repair animation based on the internal image of the vehicle and the external image of the vehicle; According to the target repair case, generate a repair operation instruction text and a voice commentary, and synchronously present the repair operation instruction text and the voice commentary to the VR repair animation to obtain a target VR repair animation, so that the user can view the vehicle status through the target VR repair animation.

2. The method according to claim 1, characterized in that, The preset historical repair case library includes multiple repair cases. The step of determining the target repair case corresponding to the vehicle to be repaired from the preset historical repair case library based on the basic vehicle information and the appearance damage information specifically includes: Based on the basic vehicle information and the appearance damage information, construct a vehicle fault feature vector; Extract keywords from each of the repair cases to obtain multiple keywords, and construct corresponding repair case feature vectors based on the multiple keywords; Calculate the similarity between the vehicle fault feature vector and each of the repair case feature vectors, and determine the target repair case according to the similarity. The target repair case is the repair case corresponding to the maximum similarity among the multiple similarities.

3. The method according to claim 1, characterized in that The step of generating a VR repair animation based on the internal image of the vehicle and the external image of the vehicle specifically includes: Extract key frames from the internal image of the vehicle and the external image of the vehicle respectively to obtain key step images from different perspectives; Segment the key step images to extract each component of the vehicle to be repaired in the key step images; Generate a three-dimensional scene model corresponding to the vehicle to be repaired based on each of the components; Input the three-dimensional scene model into a VR engine for rendering to obtain the VR repair animation.

4. The method according to claim 3, wherein The step of segmenting the key step images to extract each component of the vehicle to be repaired in the key step images specifically includes: Perform edge detection on the key step images to obtain an edge contour map of the key step images; Perform region segmentation on the key step images based on the edge contour map to obtain a segmentation result map of the key step images. Different regions in the segmentation result map correspond to different components of the vehicle to be repaired; Identify each of the regions in the segmentation result map according to a preset component model to obtain each of the components.

5. The method according to claim 3, characterized in that, The step of generating a three-dimensional scene model corresponding to the vehicle to be repaired based on each of the components specifically includes: Obtain the three-dimensional dimension information of each of the components; Construct a three-dimensional skeleton model of the vehicle to be repaired based on the three-dimensional dimension information of each of the components; Perform texture mapping on the surfaces of each of the components to generate three-dimensional solid models of each of the components; Attach the three-dimensional solid models of each of the components to the corresponding positions on the three-dimensional skeleton model to obtain the complete three-dimensional scene model.

6. The method according to claim 1, wherein The generation of the maintenance operation instruction text and voice commentary according to the target maintenance case specifically includes: Extract key information from the target maintenance case, where the key information includes maintenance steps and the corresponding maintenance components and maintenance tools for each of the maintenance steps; Fill the key information into a preset maintenance template to obtain the maintenance operation instruction text; Use a preset text-to-speech model to convert the maintenance operation instruction text into the voice commentary.

7. The method according to claim 1, wherein The synchronous presentation of the maintenance operation instruction text and voice commentary to the VR maintenance animation specifically includes: Identify and label each animation scene in the VR maintenance animation to obtain animation scene identification information, where the animation scene identification information includes the name and time period of each animation scene; Divide the maintenance operation instruction text into multiple text paragraphs, with each text paragraph corresponding to a maintenance operation step; Match the names of each of the animation scenes with the maintenance operation steps to determine the display time points of each text paragraph in the VR maintenance animation; Divide the voice commentary into multiple audio segments, with each audio segment corresponding to a maintenance operation step; Based on the display time points, superimpose each of the text paragraphs and each of the audio segments in the VR maintenance animation to obtain the target VR maintenance animation.

8. An AI and VR-based vehicle reception management device for auto repair shops, characterized in that, The device includes an acquisition module (201) and a processing module (202), where: The acquisition module (201) is configured to acquire a vehicle image collected by a camera when the vehicle to be repaired enters an auto repair shop; The processing module (202) is configured to identify the vehicle image to obtain vehicle basic information and appearance damage information, where the vehicle basic information includes the license plate number and vehicle model, and the appearance damage information includes the fault location; The processing module (202) is further configured to determine a target maintenance case corresponding to the vehicle to be repaired from a preset historical maintenance case library based on the vehicle basic information and the appearance damage information; The processing module (202) is further configured to send the target maintenance case to a maintenance personnel for the maintenance personnel to repair the vehicle to be repaired; The processing module (202) is further configured to acquire vehicle interior and exterior images during the maintenance process, where the vehicle interior and exterior images include vehicle interior images and vehicle exterior images; The processing module (202) is further configured to generate a VR maintenance animation based on the vehicle interior images and the vehicle exterior images; The processing module (202) is further configured to generate a maintenance operation instruction text and a voice commentary according to the target maintenance case, and synchronously present the maintenance operation instruction text and the voice commentary in the VR maintenance animation to obtain a target VR maintenance animation, so that a user can view the vehicle state through the target VR maintenance animation.

9. An electronic device, characterized in that, It includes a processor (301), a memory (305), a user interface (303) and a network interface (304). The memory (305) is used to store instructions. The user interface (303) and the network interface (304) are used to communicate with other devices. The processor (301) is used to execute the instructions stored in the memory (305) so that the electronic device (300) executes the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, execute the method according to any one of claims 1-7.

Images