Intelligent car washing guiding method and system of self-service car washing machine

Abstract

The invention relates to an intelligent car washing guiding method and system of a self-service car washer, and the method comprises the steps: obtaining image information and inputting the image information to an image processing model when a starting instruction is received, enabling the image processing model to output vehicle model information and vehicle state information, and inputting the vehicle model information and the vehicle state information to a cleaning guiding model; when a car washing instruction is received, acquiring stain image information and stain property information in real time; and dynamically adjusting the cleaning guidance information, and performing vehicle cleaning guidance on the user. The vehicle type, the damaged state and the refitting state of the vehicle are recognized, the cleaning guide information is generated based on the recognition result, the stain information of the vehicle is recognized, and the vehicle washing parameter information in the cleaning guide information is dynamically adjusted based on the recognition result, so that the vehicle washing difficulty is reduced, the user experience is improved, and the user experience is improved. And the car washing efficiency and the car washing instruction are improved, and meanwhile, the damage to the car body caused by car washing is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent analysis, and in particular to an intelligent car wash guidance method and system for a self-service car wash machine. Background Art

[0002] At present, self-service car wash machines have been found in major gas stations, service areas, parking lots, towns, villages, communities and other places, which means that self-service car wash machines have been widely used in the domestic market.

[0003] However, when using existing self-service car wash machines, users are faced with an important technical problem: how to accurately provide washing instructions for vehicles that have been damaged or modified. Due to changes in the appearance and structure of such vehicles, traditional washing instructions may not be able to accurately identify the vehicle model and vehicle status, and thus cannot provide appropriate car washing parameters and guidance information, which not only affects the car washing effect, but may also cause further damage to the vehicle. Summary of the invention

[0004] In order to provide intelligent and appropriate self-service car wash instructions to owners of damaged and modified vehicles, improve car wash efficiency and quality, reduce the difficulty of car wash, and reduce damage to the vehicle body caused by car wash, the present application provides an intelligent car wash instruction method and system for a self-service car wash machine.

[0005] The above-mentioned invention objective of the present application is achieved through the following technical solutions:

[0006] An intelligent car washing guidance method for a self-service car washing machine comprises the following steps:

[0007] When receiving a start command from the car washing machine terminal, a vehicle image acquisition command is sent to the image acquisition terminal to acquire image information of the pre-washed vehicle;

[0008] Inputting the acquired image information of the pre-washed vehicle into the image processing model, so that the image processing model outputs vehicle model information and vehicle status information including vehicle damage status and modification status;

[0009] Input the received vehicle model information and vehicle status information into the cleaning guidance model, so that the cleaning guidance model outputs cleaning guidance information;

[0010] When receiving a car washing instruction from the car washing machine terminal, a stain image information acquisition instruction is sent to the image acquisition terminal to acquire the stain image information and stain property information in real time;

[0011] Based on the cleaning guidance information and the real-time acquired stain image information and stain property information, the car washing parameter information in the cleaning guidance information is dynamically adjusted, and the car washing guidance is provided to the user in real time on the display terminal.

[0012] By adopting the above technical solution, when the user starts the self-service car wash machine and issues a start command, a vehicle image acquisition command is sent to the image acquisition terminal, thereby obtaining image information of the user's current pre-washed vehicle, and the obtained vehicle image information is input into a pre-trained image processing model, so that the image processing model recognizes the vehicle image information and outputs vehicle model information and vehicle status information including whether the vehicle is damaged and modified, and inputs the received information into the washing guidance model, so that the washing guidance model outputs preliminary washing guidance information. When the user officially starts washing the car, a stain image information acquisition command is sent to the image acquisition terminal, thereby obtaining stain image information in real time, and obtaining stain information based on the stain image information. The cleaning instruction information and the stain image information and the stain property information acquired in real time are used to dynamically adjust the car washing parameter information in the cleaning instruction information, and provide guidance on the car washing operation to the car washing user in real time on the display terminal; the present application identifies the vehicle model, damage status and modification status, generates specific cleaning instruction information based on the identification result, identifies the stain information on the surface of the vehicle, dynamically adjusts the car washing parameter information in the cleaning instruction information based on the identification result, and provides cleaning instructions to the user in real time through the display terminal, which has the effect of reducing the difficulty of car washing, improving user experience, and improving car washing efficiency and car washing instructions, while reducing the damage to the vehicle body caused by car washing.

[0013] In a preferred example, the present application can be further configured as follows: the image processing model includes a processing layer, a matching layer, a first recognition layer and a second recognition layer, the vehicle status information includes first status information and second status information, and the step of inputting the acquired image information of the pre-washed vehicle into the image processing model so that the image processing model outputs vehicle model information and vehicle status information including vehicle damage status and modification status includes the steps of:

[0014] The processing layer pre-processes the received image information of the pre-washed vehicle, thereby outputting processed image information;

[0015] The matching layer matches the corresponding vehicle model information based on the processed image information;

[0016] The first recognition layer identifies whether there is a damaged area of ​​the vehicle in the processed image information and its corresponding damage type based on a preset image segmentation strategy, and generates first state information based on the recognition result;

[0017] The second recognition layer identifies whether there is a modification area and its corresponding modification type for the vehicle in the processed image information based on a preset vehicle information library, and generates second state information based on the recognition result.

[0018] By adopting the above technical solution, the processing layer is the first layer of the image processing model, which is used to preprocess the original image information to improve the image quality; the matching layer identifies and matches the corresponding vehicle model information based on the preprocessed image information; the first recognition layer analyzes the preprocessed image based on a preset image segmentation strategy to identify whether there is a damaged area on the vehicle, and determines the type of damage, and generates first state information; the second recognition layer identifies the preprocessed image based on a preset vehicle information library for further analysis, thereby identifying whether there is a modified area on the vehicle in the image, and determines the modification type, and generates second state information.

[0019] In a preferred example, the present application can be further configured as follows: the first recognition layer identifies whether there is a damaged area and the corresponding damage type of the vehicle in the image information based on a preset image segmentation strategy, and generates the first state information based on the recognition result, including the steps of:

[0020] The first recognition layer divides the processed image information into several regional images based on a preset image segmentation strategy;

[0021] The first recognition layer recognizes and extracts the body area features that indicate the damage to the vehicle based on the divided regional images;

[0022] The first recognition layer determines the damage type of the vehicle body based on the extracted features of the vehicle body area.

[0023] By adopting the above technical solution, the first recognition layer divides the processed image information into several regional images based on a preset image segmentation strategy, and performs recognition based on the several divided regional images respectively to extract the vehicle body regional features that symbolize the damage to the vehicle body. After extracting the vehicle body regional features, the first recognition layer determines the damage type of the vehicle through the extracted vehicle body regional features; by adopting the image segmentation strategy to perform fine division of the processed image information, and performing feature extraction and damage type judgment based on the divided regional images, the accuracy and efficiency of vehicle damage recognition are improved.

[0024] In a preferred example, the present application can be further configured as follows: the second recognition layer identifies whether there is a modification area and its corresponding modification type in the vehicle in the image information based on a preset vehicle information database, and generates the second state information based on the recognition result, including the steps of:

[0025] The second recognition layer matches the corresponding vehicle model image from the preset vehicle information library based on the vehicle model information;

[0026] The second recognition layer compares and analyzes the matched vehicle model image with the processed image information, and extracts the body modification features that symbolize vehicle modification;

[0027] The second recognition layer determines the modification type of the vehicle based on the extracted vehicle body modification features.

[0028] By adopting the above technical solution, the second recognition layer matches the vehicle model image corresponding to the vehicle model information from the preset vehicle information library based on the vehicle model information, and compares and analyzes the matched vehicle model image with the processed image information, so as to extract the body modification features that symbolize vehicle modification and judge its modification type; by matching the original vehicle model image based on the vehicle model information and comparing and analyzing it with the current image processing information, it is possible to accurately identify whether the vehicle has a modification area and its corresponding modification type, which has the effect of improving the accuracy and reliability of vehicle modification recognition.

[0029] In a preferred example, the present application can be further configured as follows: when receiving a car washing instruction from a car washing machine terminal, sending a stain image information acquisition instruction to an image acquisition terminal to acquire stain image information and stain property information in real time, the step includes the following steps:

[0030] Identify and analyze the image information collected by the image acquisition terminal, and adjust the acquisition area of ​​the image acquisition terminal based on the identification result, so as to obtain the stain image information in real time;

[0031] The stain image information acquired in real time is identified, classified and analyzed to obtain the stain property information, which includes the stain type and the stain degree.

[0032] By adopting the above technical solution, the preliminary image information collected by the image acquisition terminal is identified and analyzed, so as to identify the key stain areas on the vehicle surface, and the acquisition area of ​​the image acquisition terminal is adjusted based on the recognition result so that it focuses on collecting images of the key stain areas. The stain image information obtained in real time is identified, classified and analyzed, so as to determine the type of stain and analyze its degree of stain, and output it as stain property information.

[0033] In a preferred example, the present application can be further configured as follows: the step of dynamically adjusting the car washing parameter information in the cleaning guidance information based on the cleaning guidance information and the stain image information and stain property information obtained in real time, and providing the user with car washing guidance in real time on the display terminal includes the steps of:

[0034] Inputting the stain image information and the stain property information into a pre-trained detergent matching model, so that the detergent matching model identifies the corresponding stain type and its stain degree and matches several detergent types and their corresponding detergent concentrations;

[0035] When the detergent information output by the detergent matching model is received, based on a preset screening strategy, at least one type of detergent information is screened in combination with the cleaning instruction information and the detergent information and the corresponding detergent concentration information is obtained;

[0036] The car wash parameter information is dynamically adjusted based on the detergent type information and the detergent concentration information.

[0037] By adopting the above technical solution, the stain image information and the stain property information are input into the pre-trained detergent matching model, so that the detergent matching model can identify the corresponding stain type and its stain degree, and match several detergent types and the detergent concentrations corresponding to the detergent types. When the detergent information output by the detergent matching model is received, based on the preset screening strategy, combined with the cleaning guidance information and the detergent information, at least one most suitable detergent type is screened out and its corresponding detergent concentration is obtained. Based on the screened detergent type information and the obtained detergent concentration information, the car washing parameter information is dynamically adjusted to improve the cleaning effect. By real-time analysis of the stain image information and the stain property information, and dynamic adjustment of the car washing parameter information, the car washing efficiency and cleaning quality are improved.

[0038] The second object of the invention is achieved by the following technical solutions:

[0039] An intelligent car washing guidance system for a self-service car washing machine, comprising:

[0040] An image information receiving module is used to send a vehicle image acquisition instruction to the image acquisition terminal to acquire image information of the pre-washed vehicle when receiving a start instruction from the vehicle washing machine terminal;

[0041] An image information input module, used for inputting the acquired image information of the pre-washed vehicle into the image processing model, so that the image processing model outputs vehicle model information and vehicle status information including vehicle damage status and modification status;

[0042] A status information input module, used to input the received vehicle model information and vehicle status information into the cleaning guidance model, so that the cleaning guidance model outputs cleaning guidance information;

[0043] The stain information acquisition module is used to send a stain image information acquisition instruction to the image acquisition terminal when receiving a car washing instruction from the car washing machine terminal, so as to obtain the stain image information and stain property information in real time;

[0044] The adjustment module is used to dynamically adjust the car washing parameter information in the cleaning guidance information based on the cleaning guidance information and the stain image information and stain property information obtained in real time, and provide car washing guidance to the user in real time on the display terminal.

[0045] By adopting the above technical solution, the image information receiving module is used to send a vehicle image acquisition instruction to the image acquisition terminal when receiving a start instruction from the car washing machine terminal, so as to obtain image information of the pre-washed vehicle; the image information input module is used to input the acquired image information of the pre-washed vehicle into the image processing model, so that the image processing model outputs vehicle model information and vehicle status information including the vehicle damage status and modification status; the status information input module is used to input the received vehicle model information and vehicle status information into the cleaning guidance model, so that the cleaning guidance model outputs cleaning guidance information; the stain information acquisition module is used to send a stain image information acquisition instruction to the image acquisition terminal when receiving a car washing instruction from the car washing machine terminal, so as to obtain stain image information and stain property information in real time; the adjustment module is used to dynamically adjust the car washing parameter information in the cleaning guidance information based on the cleaning guidance information and the stain image information and stain property information acquired in real time, and provide car washing guidance to the user in real time on the display terminal.

[0046] In a preferred example, the present application can be further configured as follows:

[0047] A processing layer module, used for preprocessing the received image information of the pre-washed vehicle, thereby outputting processed image information;

[0048] A matching layer module is used to match the corresponding vehicle model information based on the processed image information;

[0049] A first recognition layer module is used to identify whether there is a damaged area of ​​the vehicle in the processed image information and the corresponding damage type based on a preset image segmentation strategy, and generate first status information based on the recognition result;

[0050] The second recognition layer module is used to identify whether there is a modification area and its corresponding modification type on the vehicle in the processed image information based on a preset vehicle information database, and generate second state information based on the recognition result.

[0051] By adopting the above technical solution, the processing layer module is used to preprocess the received image information of the pre-washed vehicle, so as to output the processed image information; the matching layer module is used to match the corresponding vehicle model information based on the processed image information; the first recognition layer module is used to identify whether the vehicle in the processed image information has a damaged area and its corresponding damage type based on a preset image segmentation strategy, and generate first status information based on the recognition result; the second recognition layer module is used to identify whether the vehicle in the processed image information has a modified area and its corresponding modification type based on a preset vehicle information library, and generate second status information based on the recognition result.

[0052] In summary, the present application includes at least one of the following beneficial technical effects:

[0053] 1. This application identifies the vehicle model, damage status and modification status, generates specific cleaning guidance information based on the identification results, and identifies the stain information on the vehicle surface, dynamically adjusts the car washing parameter information in the cleaning guidance information based on the identification results, and provides cleaning guidance to users in real time through a display terminal, which has the effect of reducing the difficulty of car washing, improving user experience, and improving car washing efficiency and car washing instructions, while reducing the damage to the vehicle body caused by car washing;

[0054] 2. By adopting image segmentation strategy to finely divide the processed image information, and extracting features and judging the damage type based on the divided regional images, it has the effect of improving the accuracy and efficiency of vehicle damage identification;

[0055] 3. By matching the original vehicle model image based on the vehicle model information and comparing and analyzing it with the current image processing information, it is possible to accurately identify whether the vehicle has a modified area and its corresponding modification type, which has the effect of improving the accuracy and reliability of vehicle modification identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flow chart of an embodiment of an intelligent car washing guidance method for a self-service car washing machine of the present application;

[0057] Figure 2 This is a flowchart of an implementation of step S20 in an embodiment of an intelligent car wash guidance method for a self-service car wash machine of the present application;

[0058] Figure 3 This is a flowchart of implementing step S23 in an embodiment of an intelligent car wash guidance method for a self-service car wash machine of the present application;

[0059] Figure 4 This is a flowchart for implementing step S24 in an embodiment of an intelligent car washing guidance method for a self-service car washing machine of the present application;

[0060] Figure 5 This is a flowchart for implementing step S50 in an embodiment of an intelligent car wash guidance method for a self-service car wash machine of the present application. DETAILED DESCRIPTION

[0061] The following is combined with Figure 1-5 This application is described in further detail.

[0062] In one embodiment, if Figure 1 As shown, the present application discloses an intelligent car washing guidance method for a self-service car washing machine, which specifically includes the following steps:

[0063] S10: When receiving a start instruction from the car washing machine terminal, sending a vehicle image acquisition instruction to the image acquisition terminal to acquire image information of the pre-washed vehicle;

[0064] In this embodiment, the start instruction is an electrical signal instruction issued when the user starts the self-service car washing machine to indicate that the car washing process is about to begin; the image acquisition terminal is a terminal device for capturing and recording image information; the vehicle image acquisition instruction is an electrical signal instruction issued for acquiring image information of the pre-washed vehicle; the image information of the pre-washed vehicle is the image information of the appearance of the vehicle body to be washed;

[0065] Specifically, when the user starts the self-service car wash machine and issues a start command, a vehicle image acquisition command is sent to the image acquisition terminal, thereby acquiring image information of the user's current pre-washed vehicle.

[0066] S20: inputting the acquired image information of the pre-washed vehicle into an image processing model, so that the image processing model outputs vehicle model information and vehicle status information including a damaged state and a modified state of the vehicle;

[0067] In this embodiment, the image processing model is a pre-trained machine learning model for analyzing and processing vehicle images, thereby outputting vehicle model identification information and vehicle status information (body damage status, body modification status); vehicle model information is a general term for information about vehicle model, size, etc.; vehicle status information is information reflecting the appearance status of the vehicle, including the body damage status and body modification status;

[0068] Specifically, the acquired vehicle image information is input into a pre-trained image processing model, so that the image processing model recognizes the vehicle image information and outputs vehicle model information and vehicle status information including whether the vehicle is damaged and the modification status.

[0069] S30: inputting the received vehicle model information and vehicle status information into the cleaning guidance model, so that the cleaning guidance model outputs cleaning guidance information;

[0070] In this embodiment, the washing guidance model is a pre-trained model for outputting washing guidance information for different vehicles based on vehicle model information and vehicle status information. The washing guidance information specifically includes outputting vehicle washing parameter information and vehicle washing step information.

[0071] Specifically, the received information is input into the cleaning guidance model, so that the cleaning guidance model outputs preliminary cleaning guidance information.

[0072] S40: When a car washing instruction is received from the car washing machine terminal, a stain image information acquisition instruction is sent to the image acquisition terminal to acquire the stain image information and stain property information in real time;

[0073] In this embodiment, the stain image information acquisition instruction is an electrical signal instruction issued for acquiring stain image information; the stain image information is image information containing a stain image; the stain property information is information including the stain type and the stain degree, wherein the stain type includes oil stains, mud stains, water stains and others, and the stain degree includes the area, color and depth of the stain;

[0074] Specifically, when the user officially starts washing the car, he sends a stain image information acquisition instruction to the image acquisition terminal, so as to acquire the stain image information in real time, and acquire the stain property information according to the stain image information.

[0075] S50: dynamically adjusting the car washing parameter information in the cleaning guidance information based on the cleaning guidance information and the stain image information and stain property information obtained in real time, and providing the user with car washing guidance in real time on the display terminal;

[0076] In this embodiment, the car wash parameter information is the specific parameter setting information of the car wash process steps in the cleaning guide information, including the car wash time, water pressure intensity, detergent type and detergent concentration; the display terminal is a display area device on the self-service car wash machine for interacting with the user;

[0077] Specifically, based on the cleaning guidance information and the stain image information and stain property information obtained in real time, the car washing parameter information in the cleaning guidance information is dynamically adjusted, and the car washing operation is guided to the car washing user in real time on the display terminal.

[0078] In one embodiment, the image processing model includes a processing layer, a matching layer, a first recognition layer and a second recognition layer, and the vehicle state information includes first state information and second state information, such as Figure 2 As shown, step S20 includes the steps of:

[0079] S21: the processing layer pre-processes the received image information of the pre-washed vehicle, thereby outputting processed image information;

[0080] S22: The matching layer matches the corresponding vehicle model information based on the processed image information;

[0081] S23: The first recognition layer identifies whether there is a damaged area of ​​the vehicle in the processed image information and its corresponding damage type based on a preset image segmentation strategy, and generates first state information based on the recognition result;

[0082] S24: The second recognition layer identifies whether there is a modification area and its corresponding modification type for the vehicle in the processed image information based on a preset vehicle information database, and generates second state information based on the recognition result.

[0083] In this embodiment, the processing layer is the first layer of the image processing model, which is used to pre-process the received image information of the pre-washed vehicle, wherein the pre-processing includes image denoising, brightness / contrast adjustment, image cropping, etc., to improve the image quality and facilitate subsequent analysis and recognition; the matching layer is the second layer of the image processing model, which is used to match the corresponding vehicle model information based on the processed image information, specifically, to compare the image features with the pre-stored vehicle model database to determine the vehicle model; the first recognition layer is used to identify whether the vehicle has a damaged area and its corresponding damage type; the second recognition layer is used to identify whether the vehicle has a modified area and its corresponding modification type;

[0084] Specifically, the processing layer is the first layer of the image processing model, which is used to preprocess the original image information to improve the image quality; the matching layer identifies and matches the corresponding vehicle model information based on the preprocessed image information; the first recognition layer analyzes the preprocessed image based on a preset image segmentation strategy to identify whether there is a damaged area on the vehicle, and determines the type of damage to generate first state information; the second recognition layer identifies the preprocessed image based on a preset vehicle information library for further analysis, thereby identifying whether there is a modified area on the vehicle in the image, and determines the modification type, and generates second state information.

[0085] In one embodiment, if Figure 3 As shown, step S23 includes the steps of:

[0086] S231: The first recognition layer divides the processed image information into a plurality of regional images based on a preset image segmentation strategy;

[0087] S232: the first recognition layer recognizes and extracts vehicle body area features that represent vehicle damage based on the divided region images;

[0088] S233: The first recognition layer determines the damage type of the vehicle body based on the extracted vehicle body region features.

[0089] In this embodiment, the image segmentation strategy is a technology for dividing an image into multiple sub-regions; the regional image is an image sub-region divided by the image segmentation strategy, and each regional image contains a part of the vehicle body; the vehicle body regional feature is a specific image feature in each regional image that can reflect the damage condition of the vehicle body, including color change, texture abnormality and shape deformation; the damage type is the specific type or nature of the damage to the vehicle, including scratches, dents, impact marks and cracks;

[0090] Furthermore, the extraction of vehicle body area features includes color analysis, texture detection and shape analysis; color change features are identified by comparing the color differences of different vehicle body areas in the regional image, and further color difference identification can be performed by color transformation, contrast change and exposure change of the regional image; texture anomaly features are identified by detecting irregular textures or patterns in the image; shape deformation features are identified by comparing the pre-stored corresponding vehicle model image and the shape difference or edge detection in the regional image;

[0091] Furthermore, judging the damage type based on the extracted features of the vehicle body region usually involves a machine learning algorithm, such as a classifier or a neural network; judging the damage type based on the extracted features (such as color change features, texture abnormality features, shape deformation features, etc.), for example, scratches appear as elongated color change features, dents appear as local shape deformation features, impact marks may contain color change features, texture abnormality features and shape deformation features at the same time, and cracks appear as obvious texture abnormality features (texture fractures) and shape deformation features;

[0092] Specifically, the first recognition layer divides the processed image information into several regional images based on a preset image segmentation strategy, and performs recognition based on the divided regional images respectively to extract vehicle body regional features that symbolize vehicle body damage. After extracting the vehicle body regional features, the first recognition layer determines the type of damage to the vehicle through the extracted vehicle body regional features. By adopting an image segmentation strategy to perform fine division of the processed image information, and performing feature extraction and damage type judgment based on the divided regional images, the accuracy and efficiency of vehicle damage recognition are improved.

[0093] In one embodiment, if Figure 4 As shown, step S24 includes the steps of:

[0094] S241: The second recognition layer matches the corresponding vehicle model image from a preset vehicle information library based on the vehicle model information;

[0095] S242: the second recognition layer compares and analyzes the matched vehicle model image and the processed image information, and extracts vehicle body modification features that represent vehicle modification;

[0096] S243: The second recognition layer determines the modification type of the vehicle body based on the extracted vehicle body modification features.

[0097] In this embodiment, the preset vehicle information database is a database containing standard images and configuration information of various vehicle models; the vehicle body modification features are specific image features reflecting the vehicle body modification conditions, including color change features, shape deformation features, and additional component features; the modification type is the specific type or nature of vehicle modification, including adding components, replacing wheels, and film and painting;

[0098] Furthermore, the features of vehicle body modification are extracted based on image processing technology, including edge detection, color analysis, and shape matching;

[0099] Specifically, the second recognition layer matches the vehicle model image corresponding to the vehicle model information from a preset vehicle information library based on the vehicle model information, and compares and analyzes the matched vehicle model image with the processed image information, so as to extract the body modification features that symbolize vehicle modification and determine its modification type; by matching the original vehicle model image based on the vehicle model information and comparing and analyzing it with the current image processing information, it can accurately identify whether the vehicle has a modification area and its corresponding modification type, which has the effect of improving the accuracy and reliability of vehicle modification recognition.

[0100] In one embodiment, step S40 includes the steps of:

[0101] S41: Identify and analyze the image information collected by the image acquisition terminal, and adjust the acquisition area of ​​the image acquisition terminal based on the identification result, so as to obtain the stain image information in real time;

[0102] S42: Identify, classify and analyze the stain image information obtained in real time, so as to obtain the stain property information, wherein the stain property information includes the stain type and the stain degree.

[0103] In this embodiment, the recognition result includes the position of the key stain area, and adjusting the acquisition area of ​​the image acquisition terminal is to adjust the settings of the image acquisition terminal so that it focuses on acquiring the key stain area;

[0104] Furthermore, the type of stain is identified by analyzing the color, shape, texture and other characteristics of the stain. For example, different types of stains such as oil stains, water stains, and dust will have different manifestations on the image. The degree of the stain is analyzed based on image processing techniques such as image segmentation and area calculation, that is, the coverage area and density of the stain.

[0105] Specifically, the preliminary image information collected by the image acquisition terminal is identified and analyzed to identify the key stain areas on the vehicle surface, and the acquisition area of ​​the image acquisition terminal is adjusted based on the identification result to focus on collecting images of the key stain areas. The stain image information obtained in real time is identified, classified and analyzed to determine the type of stain and analyze its degree of stain, and output it as stain property information.

[0106] In one embodiment, if Figure 5 As shown, step S50 includes the steps of:

[0107] S51: inputting the stain image information and the stain property information into a pre-trained detergent matching model, so that the detergent matching model identifies the corresponding stain type and its stain degree and matches several detergent types and their corresponding detergent concentrations;

[0108] S52: when receiving the detergent information output by the detergent matching model, based on a preset screening strategy, screening at least one type of detergent information in combination with the cleaning instruction information and the detergent information and obtaining the corresponding detergent concentration information;

[0109] S53: Dynamically adjust the car wash parameter information based on the detergent type information and the detergent concentration information.

[0110] In this embodiment, the detergent matching model is a pre-trained deep learning model for matching the appropriate detergent type and detergent concentration according to the stain type and stain degree; the detergent information is information output by the detergent matching model including the detergent type and detergent concentration; the preset screening strategy is a method strategy for screening at least one most suitable detergent type and its concentration by comprehensively considering all detergent types and detergent concentrations in the cleaning guide information and the detergent information; the detergent type information is the specific information of the screened detergent type; the detergent concentration information is the concentration information corresponding to the screened detergent type; the dynamic adjustment of the car wash parameter information is based on the screened detergent type and concentration information, dynamically adjusting the detergent dosage, washing time, water temperature and other parameters to ensure the best cleaning effect;

[0111] Specifically, the stain image information and the stain property information are input into a pre-trained detergent matching model, so that the detergent matching model identifies the corresponding stain type and its degree of stain, and matches several detergent types and the detergent concentrations corresponding to the detergent types. When the detergent information output by the detergent matching model is received, based on a preset screening strategy, in combination with the cleaning guidance information and the detergent information, at least one most suitable detergent type is screened out and its corresponding detergent concentration is obtained. Based on the screened detergent type information and the obtained detergent concentration information, the car wash parameter information is dynamically adjusted to improve the cleaning effect. By real-time analysis of the stain image information and the stain property information, and dynamic adjustment of the car wash parameter information, the car wash efficiency and cleaning quality are improved.

[0112] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0113] In one embodiment, a smart car wash guidance system for a self-service car wash machine is provided, and the smart car wash guidance system for a self-service car wash machine corresponds one-to-one to a smart car wash guidance method for a self-service car wash machine in the above embodiment. The smart car wash guidance system for a self-service car wash machine includes:

[0114] An image information receiving module is used to send a vehicle image acquisition instruction to the image acquisition terminal to acquire image information of the pre-washed vehicle when receiving a start instruction from the vehicle washing machine terminal;

[0115] An image information input module, used for inputting the acquired image information of the pre-washed vehicle into the image processing model, so that the image processing model outputs vehicle model information and vehicle status information including vehicle damage status and modification status;

[0116] A status information input module, used to input the received vehicle model information and vehicle status information into the cleaning guidance model, so that the cleaning guidance model outputs cleaning guidance information;

[0117] The stain information acquisition module is used to send a stain image information acquisition instruction to the image acquisition terminal when receiving a car washing instruction from the car washing machine terminal, so as to obtain the stain image information and stain property information in real time;

[0118] An adjustment module, for dynamically adjusting the car washing parameter information in the cleaning guidance information based on the cleaning guidance information and the stain image information and stain property information obtained in real time, and providing car washing guidance to the user in real time on the display terminal;

[0119] Optionally, also include:

[0120] A processing layer module, used for preprocessing the received image information of the pre-washed vehicle, thereby outputting processed image information;

[0121] A matching layer module is used to match the corresponding vehicle model information based on the processed image information;

[0122] A first recognition layer module is used to identify whether there is a damaged area of ​​the vehicle in the processed image information and the corresponding damage type based on a preset image segmentation strategy, and generate first status information based on the recognition result;

[0123] The second recognition layer module is used to identify whether there is a modification area and the corresponding modification type of the vehicle in the image information based on a preset vehicle information database, and generate second state information based on the recognition result;

[0124] Optionally, the first identification layer module includes:

[0125] An image segmentation submodule is used to divide the processed image information into a number of regional images based on a preset image segmentation strategy;

[0126] A vehicle body region feature extraction submodule is used to identify and extract vehicle body region features that indicate vehicle damage based on the divided region images;

[0127] A damage type judgment submodule is used to judge the damage type of the vehicle body based on the extracted vehicle body area features;

[0128] Optionally, the second identification layer module includes:

[0129] A matching submodule, used to match the corresponding vehicle model image from a preset vehicle information library based on the vehicle model information;

[0130] The vehicle body modification feature extraction submodule is used to compare and analyze the matched vehicle model image with the processed image information, and extract the vehicle body modification features that indicate vehicle modification;

[0131] A modification type judgment submodule is used to judge the modification type of the vehicle based on the extracted vehicle body modification features;

[0132] Optionally, also include:

[0133] An image acquisition area adjustment module is used to identify and analyze the image information collected by the image acquisition terminal, and adjust the acquisition area of ​​the image acquisition terminal based on the identification result, so as to obtain the stain image information in real time;

[0134] A stain property information acquisition module, used to identify, classify and analyze the stain image information acquired in real time, so as to acquire the stain property information, wherein the stain property information includes the stain type and the stain degree;

[0135] Optionally, also include:

[0136] A stain information input module is used to input stain image information and stain property information into a pre-trained detergent matching model, so that the detergent matching model identifies the corresponding stain type and its stain degree and matches several detergent types and their corresponding detergent concentrations;

[0137] A detergent information screening module, for, upon receiving detergent information output by a detergent matching model, screening at least one type of detergent information based on a preset screening strategy, in combination with the cleaning instruction information and the detergent information, and obtaining the corresponding detergent concentration information;

[0138] The adjustment module is specifically used to dynamically adjust the car wash parameter information based on the detergent type information and the detergent concentration information.

[0139] For the specific definition of an intelligent car wash guidance system for a self-service car wash machine, please refer to the definition of an intelligent car wash guidance method for a self-service car wash machine mentioned above, which will not be repeated here. Each module in the above-mentioned intelligent car wash guidance system for a self-service car wash machine can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0140] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An intelligent car washing guidance method for a self-service car washing machine, characterized in that: Includes steps: When receiving a start command from the car washing machine terminal, a vehicle image acquisition command is sent to the image acquisition terminal to acquire image information of the pre-washed vehicle; Inputting the acquired image information of the pre-washed vehicle into the image processing model, so that the image processing model outputs vehicle model information and vehicle status information including vehicle damage status and modification status; Input the received vehicle model information and vehicle status information into the cleaning guidance model, so that the cleaning guidance model outputs cleaning guidance information; When receiving a car washing instruction from the car washing machine terminal, a stain image information acquisition instruction is sent to the image acquisition terminal to acquire the stain image information and stain property information in real time; Based on the cleaning guidance information and the real-time acquired stain image information and stain property information, the car washing parameter information in the cleaning guidance information is dynamically adjusted, and the car washing guidance is provided to the user in real time on the display terminal.

2. The intelligent car washing guidance method of a self-service car washing machine according to claim 1, characterized in that: The image processing model includes a processing layer, a matching layer, a first recognition layer and a second recognition layer, the vehicle status information includes first status information and second status information, and the step of inputting the acquired image information of the pre-washed vehicle into the image processing model so that the image processing model outputs vehicle model information and vehicle status information including vehicle damage status and modification status includes the steps of: The processing layer pre-processes the received image information of the pre-washed vehicle, thereby outputting processed image information; The matching layer matches the corresponding vehicle model information based on the processed image information; The first recognition layer identifies whether there is a damaged area of ​​the vehicle in the processed image information and its corresponding damage type based on a preset image segmentation strategy, and generates first state information based on the recognition result; The second recognition layer identifies whether there is a modification area and its corresponding modification type for the vehicle in the processed image information based on a preset vehicle information library, and generates second state information based on the recognition result.

3. The intelligent car washing guidance method of a self-service car washing machine according to claim 2, characterized in that: The first recognition layer identifies whether there is a damaged area of ​​the vehicle in the image information and its corresponding damage type based on a preset image segmentation strategy, and generates first state information based on the recognition result, including the steps of: The first recognition layer divides the processed image information into several regional images based on a preset image segmentation strategy; The first recognition layer recognizes and extracts the body area features that indicate the damage to the vehicle based on the divided regional images; The first recognition layer determines the damage type of the vehicle body based on the extracted features of the vehicle body area.

4. The intelligent car washing guidance method of a self-service car washing machine according to claim 2, characterized in that: The second recognition layer identifies whether there is a modification area and its corresponding modification type in the vehicle in the image information based on a preset vehicle information library, and generates the second state information based on the recognition result, including the steps of: The second recognition layer matches the corresponding vehicle model image from the preset vehicle information library based on the vehicle model information; The second recognition layer compares and analyzes the matched vehicle model image with the processed image information, and extracts the body modification features that symbolize vehicle modification; The second recognition layer determines the modification type of the vehicle based on the extracted vehicle body modification features.

5. The intelligent car washing guidance method of a self-service car washing machine according to claim 1, characterized in that: The step of sending a stain image information acquisition instruction to the image acquisition terminal when receiving a car washing instruction from the car washing machine terminal to acquire the stain image information and stain property information in real time comprises the following steps: Identify and analyze the image information collected by the image acquisition terminal, and adjust the acquisition area of ​​the image acquisition terminal based on the identification result, so as to obtain the stain image information in real time; The stain image information acquired in real time is identified, classified and analyzed to obtain the stain property information, which includes the stain type and the stain degree.

6. The intelligent car washing guidance method of a self-service car washing machine according to claim 1, characterized in that: The step of dynamically adjusting the car washing parameter information in the cleaning guidance information based on the cleaning guidance information and the stain image information and stain property information obtained in real time, and providing the user with car washing guidance in real time on the display terminal, comprises the following steps: Inputting the stain image information and the stain property information into a pre-trained detergent matching model, so that the detergent matching model identifies the corresponding stain type and its stain degree and matches several detergent types and their corresponding detergent concentrations; When the detergent information output by the detergent matching model is received, based on a preset screening strategy, at least one type of detergent information is screened in combination with the cleaning instruction information and the detergent information and the corresponding detergent concentration information is obtained; The car wash parameter information is dynamically adjusted based on the detergent type information and the detergent concentration information.

7. An intelligent car washing guidance system for a self-service car washing machine, characterized in that: include: An image information receiving module is used to send a vehicle image acquisition instruction to the image acquisition terminal to acquire image information of the pre-washed vehicle when receiving a start instruction from the vehicle washing machine terminal; An image information input module, used for inputting the acquired image information of the pre-washed vehicle into the image processing model, so that the image processing model outputs vehicle model information and vehicle status information including vehicle damage status and modification status; A status information input module, used to input the received vehicle model information and vehicle status information into the cleaning guidance model, so that the cleaning guidance model outputs cleaning guidance information; The stain information acquisition module is used to send a stain image information acquisition instruction to the image acquisition terminal when receiving a car washing instruction from the car washing machine terminal, so as to obtain the stain image information and stain property information in real time; The adjustment module is used to dynamically adjust the car washing parameter information in the cleaning guidance information based on the cleaning guidance information and the stain image information and stain property information obtained in real time, and provide car washing guidance to the user in real time on the display terminal.

8. The intelligent car washing guidance system for a self-service car washing machine according to claim 7, characterized in that: Also includes: A processing layer module, used for preprocessing the received image information of the pre-washed vehicle, thereby outputting processed image information; A matching layer module is used to match the corresponding vehicle model information based on the processed image information; A first recognition layer module is used to identify whether there is a damaged area of ​​the vehicle in the processed image information and the corresponding damage type based on a preset image segmentation strategy, and generate first status information based on the recognition result; The second recognition layer module is used to identify whether there is a modification area and its corresponding modification type on the vehicle in the processed image information based on a preset vehicle information database, and generate second state information based on the recognition result.

Images