Intelligent detergent putting control method and system for automatic car washing equipment

Abstract

The invention relates to the technical field of automatic control, and provides an intelligent detergent putting control method and system for automatic car washing equipment. The method comprises the following steps: acquiring a vehicle surface stain image through a high-precision camera, and identifying stain features to generate a multi-dimensional feature set; performing clustering division on the feature set to obtain a stain area feature set, and calling a car washing scheme parameter space; analyzing the region feature set based on the parameter space, and determining cleaning scheme parameters of each region; a cleaning agent putting part control logic is associated, and area putting control parameters are generated; the environment temperature and humidity are monitored in real time, and cleaning agent putting and feedback regulation and control optimization are carried out in combination with control parameters, so that the technical problems that existing automatic car washing equipment lacks pertinence during cleaning agent putting, the cleaning agent putting amount cannot be accurately controlled according to the actual stain condition of a car and environment changes, the car washing effect is poor, and cleaning agents are wasted are solved. The technical effects of improving the car washing quality, reducing the waste of the cleaning agent and achieving intelligent and accurate control are achieved.

Description

Technical Field

[0001] This application relates to the field of automation control technology, and particularly to an intelligent cleaner dosing control method and system for automated car washing equipment. Background Art

[0002] Although the current automated car washing equipment on the market can achieve basic cleaning processes, it generally uses fixed programs to control the dosing of cleaners, and can only make rough ratios based on preset time or water volume. It cannot accurately identify the differences in stain types, distribution density, and adhesion degree in different areas of the vehicle body, resulting in incomplete cleaning of stubborn stains or excessive use of cleaners in low-pollution areas. In addition, changes in environmental temperature and humidity directly affect the chemical activity and fluidity of the cleaner, and traditional equipment lacks the ability to respond to such dynamic parameters in real time, easily causing problems such as cleaner residue in low-temperature environments or too rapid dissipation of foam in high-temperature environments. Especially in the scenario of dealing with mixed stains such as mud, oil, and tree gum, static dosing strategies are difficult to meet the differentiated cleaning needs, not only increasing the comprehensive consumption of water, electricity, and chemical agents, but also possibly accelerating equipment corrosion or damaging the car paint due to the abuse of strong cleaners. Against this background, there is an urgent need for a cleaner dosing method and system that integrates intelligent perception, dynamic zoning control, and environmental adaptability to improve car washing efficiency, reduce resource consumption, and ensure the consistency of cleaning quality. Summary of the Invention

[0003] This application provides an intelligent cleaner dosing control method and system for automated car washing equipment, aiming to solve the technical problem that existing automated car washing equipment lacks pertinence in cleaner dosing and cannot accurately control the dosing amount of cleaners according to the actual stain conditions of the vehicle and environmental changes, resulting in poor car washing effects and waste of cleaners. The technical effect is achieved of accurately dividing the stain areas according to the stain characteristics of the vehicle body, formulating a personalized cleaning plan in combination with the parameter space of the car washing plan, and at the same time adjusting the cleaner dosing in real time according to the environmental temperature and humidity, improving the car washing quality, reducing the waste of cleaners, and achieving intelligent and accurate control.

[0004] In the first aspect disclosed in this application, an intelligent cleaner dispensing control method for an automated car wash equipment is provided. The method includes: acquiring an image of the stains on the body surface of a target vehicle through a high-precision camera, performing stain feature recognition on the image of the stains on the body surface to obtain a multi-dimensional feature set of the body stains; performing clustering division on the multi-dimensional feature set of the body stains to obtain N feature sets of body stain regions, and at the same time calling a car wash plan parameter space according to the target automated car wash equipment; parsing a cleaning plan for the N feature sets of body stain regions based on the car wash plan parameter space to determine cleaning plan parameters for N stain regions; performing control logic association on the cleaner dispensing components of the target automated car wash equipment based on the cleaning plan parameters for the N stain regions to obtain N regional cleaner dispensing control parameters; monitoring and acquiring an environmental temperature and humidity data stream in real time, and through the target automated car wash equipment, performing cleaner dispensing control and feedback regulation optimization on the target vehicle based on the N regional cleaner dispensing control parameters and the environmental temperature and humidity data stream.

[0005] In another aspect disclosed in this application, an intelligent cleaner dispensing control system for an automated car wash equipment is provided. The system includes: a stain feature recognition module: acquiring an image of the stains on the body surface of a target vehicle through a high-precision camera, performing stain feature recognition on the image of the stains on the body surface to obtain a multi-dimensional feature set of the body stains; a clustering division module: performing clustering division on the multi-dimensional feature set of the body stains to obtain N feature sets of body stain regions, and at the same time calling a car wash plan parameter space according to the target automated car wash equipment; a cleaning plan parsing module: parsing a cleaning plan for the N feature sets of body stain regions based on the car wash plan parameter space to determine cleaning plan parameters for N stain regions; a control logic association module: performing control logic association on the cleaner dispensing components of the target automated car wash equipment based on the cleaning plan parameters for the N stain regions to obtain N regional cleaner dispensing control parameters; a dispensing control module: monitoring and acquiring an environmental temperature and humidity data stream in real time, and through the target automated car wash equipment, performing cleaner dispensing control and feedback regulation optimization on the target vehicle based on the N regional cleaner dispensing control parameters and the environmental temperature and humidity data stream.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The above intelligent cleaner dosing control method for automated car washing equipment first uses a high-precision camera to collect images of the vehicle surface, and extracts multi-dimensional features such as the color, shape, and texture of the stains through image recognition technology. Subsequently, clustering analysis is performed on these features to divide multiple stain areas, and specific cleaning strategies are formulated for each area in combination with the preset scheme parameters of the car washing equipment, including the type, concentration, and dosing amount of the cleaner to be used. After that, these cleaning strategies are converted into executable control parameters to drive the cleaner dosing components of the car washing equipment to accurately dose by area. At the same time, the temperature and humidity data in the car washing environment are also collected in real time to dynamically adjust the dosing logic of the cleaner to adapt to the impact of environmental changes on the cleaning effect, realizing feedback control and self-optimization of the cleaning process. The overall process constructs a closed-loop control system from image recognition, feature analysis, scheme formulation to environmental perception and adjustment, improving the car washing effect and resource utilization efficiency.

[0007] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically gives the specific implementation manners of this application. Brief Description of the Drawings

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0009] Figure 1 It is a schematic flowchart of the intelligent cleaner dosing control method for automated car washing equipment in one embodiment.

[0010] Figure 2 It is a schematic diagram of the architecture of the intelligent cleaner dosing control system for automated car washing equipment in one embodiment.

[0011] Description of the reference numerals: stain feature recognition module 11, clustering and division module 12, cleaning scheme analysis module 13, control logic association module 14, dosing control module 15. Detailed Description of the Invention

[0012] By providing an intelligent detergent dosing control method and system for an automated car wash device, the embodiments of the present application solve the technical problems that existing automated car wash devices lack pertinence in detergent dosing, cannot accurately control the detergent dosing amount according to the actual stain conditions of the vehicle and environmental changes, resulting in poor car wash effects and detergent waste, and achieve the technical effects of accurately dividing the stain area according to the stain characteristics of the vehicle body, formulating a personalized cleaning plan in combination with the parameter space of the car wash plan, and at the same time adjusting the detergent dosing in real time according to the environmental temperature and humidity, improving the car wash quality, reducing detergent waste, and realizing intelligent and accurate control.

[0013] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0014] It should be noted that the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0015] Embodiment 1, as Figure 1 shown, the present application provides an intelligent detergent dosing control method for an automated car wash device, and the method includes: Acquire the body surface stain image of the target vehicle through a high-precision camera, and perform stain feature recognition on the body surface stain image to obtain a multi-dimensional body stain feature set.

[0016] In the embodiments of the present application, first, a high-precision camera is used to collect images of the body of the target vehicle, and high-resolution stain images covering the entire vehicle surface are obtained. These images can clearly record the stain distribution on different parts of the vehicle body. Subsequently, the collected stain images are denoised through an image filter, and through stain recognition and feature extraction, a multi-dimensional body stain feature set is obtained, such as color features (such as color differences of different stains such as oil stains, mud, or bird droppings), shape features (the edge contour and area size of the stain), and texture features (texture consistency, uniformity, etc.). These features will serve as the basic data for subsequent clustering division and cleaning strategy formulation to help the system comprehensively identify the stains.

[0017] Furthermore, the present application provides that the obtained multi-dimensional body stain feature set includes: Identify the noise characteristics of the vehicle body surface stain image to obtain the vehicle body image noise characteristic information. Initialize the image filter according to the vehicle body image noise characteristic information. Use the image filter to perform filtering and denoising processing on the vehicle body surface stain image to obtain a denoised vehicle body surface stain image. Obtain a standard vehicle body surface image according to the specification and model information of the target vehicle. Based on the standard vehicle body surface image, perform stain identification and feature extraction on the denoised vehicle body surface stain image to obtain a multi-dimensional feature set of vehicle body stains.

[0018] Preferably, first, the collected body stain image is preliminarily analyzed to identify the types and distribution characteristics of noise therein. In this process, the brightness histogram and pixel value distribution of the image are analyzed to determine whether there is obvious random noise (such as salt-and-pepper noise) or uniform noise (such as Gaussian noise). If there is no obvious random noise or uniform noise, local region texture analysis is also performed (such as statistically calculating the standard deviation of edge directions in multiple local windows. If the direction distribution in a certain region is dispersed and random, there may be blurring). The edge area and flat area of the image are compared. If the edge is blurred or the texture mutation between regions is abnormal, it is regarded as having blurring-type noise. After the preliminary analysis is completed, body image noise characteristic information, including noise type, noise intensity, etc., is generated according to the identified noise type. Subsequently, an appropriate image filter type and its parameter configuration are automatically selected according to the body image noise characteristic information. For example, if Gaussian noise is identified, a Gaussian filter is initialized, and the standard deviation σ and convolution kernel size are set according to the noise intensity; if salt-and-pepper noise is identified, a median filter is used, and the optimal sliding window size is determined; if blurring-type noise is identified, a bilateral filter or an adaptive sharpening filter is used to enhance edge details. After the image filter is initialized, the filter is loaded as a dynamic parameter into the image processing module for subsequent image cleaning. Then, using the initialized image filter, the original stain image is filtered to remove background interference and redundant texture, retain the main information of the stain, and obtain a denoised body surface stain image. This denoised body surface stain image has higher clarity and edge resolution ability, providing a clean data basis for subsequent stain identification. Then, according to the model, specification information of the current vehicle (such as vehicle model, color, body size, etc.), a standard body surface image matching the vehicle is called from the body surface database. This image represents the standard appearance of the body in an ideal state without stain coverage and is an important reference template for subsequent identification of stain areas. After obtaining the standard body surface image, the standard body surface image is compared with the denoised body surface stain image. By analyzing the body background area, the area deviating from the standard appearance, that is, the possible position of the stain, is identified, and information in multiple dimensions, including color (shape feature), shape (edge contour, area), texture (contrast), etc., is extracted to construct a complete multi-dimensional feature set of body stains. In summary, through the above processes of denoising, standard comparison, feature extraction, etc., accurate identification of stains is achieved, laying a stable and high-quality data basis for subsequent clustering division and cleaning plan matching.

[0019] Furthermore, the present application provides the obtaining of the multi-dimensional feature set of body stains, including: Taking the standard body surface image as the background area, identifying the gray-scale distribution of the background area to obtain the gray-scale threshold of the body background area; based on the gray-scale threshold of the body background area, performing stain comparison and identification on the denoised body surface stain image to obtain the set of body surface stain areas; successively performing edge detection and region segmentation on the set of body surface stain areas, and marking to obtain the set of body surface stain target frames; respectively performing multi-dimensional feature extraction on the set of body surface stain target frames to obtain the multi-dimensional body stain feature set, and the multi-dimensional body stain feature set includes stain area color features, shape features and texture features.

[0020] Optionally, first use the standard body image as the background area, and convert the RGB value of each pixel point in the background area of the standard body image into a gray value through the formula: gray value = 0.2989×R + 0.5870×G + 0.1140×B, where R is the value of the red channel, G is the value of the green channel, and B is the value of the blue channel. After obtaining the gray value of each pixel point, a gray histogram will be generated, which represents the frequency of occurrence of each gray value (usually gray levels in the range of 0-255) in the image, and according to the concentrated area in the gray histogram, an upper and lower threshold will be set, and the range composed of these two thresholds will be used as the gray-scale threshold of the body background area to distinguish the pixels in the body surface stain image from the background area, so as to help the system accurately identify the stain area. Subsequently, the denoised body surface stain image is converted into the corresponding gray value for each pixel point in the image in the same way as above, and then based on the determined gray-scale threshold of the body background area, a gray-scale comparison is performed with the denoised body surface stain image converted into a gray value to identify the parts in the image that deviate from the standard background area. These parts are recorded as stain areas and organized into a set of body surface stain areas. Then, for each stain area in the set of body surface stain areas, the Canny edge detection algorithm can be used to perform edge detection to accurately identify the contour and boundary of the stain. After edge detection, each stain area will be subjected to region segmentation, and each segmented region will be marked to obtain a set of body surface stain target frames. Each body surface stain target frame in this set of body surface stain target frames contains the position information of the stain area, which is convenient for subsequent feature extraction and analysis. Then, multi-dimensional feature extraction is performed on each body surface stain target frame respectively. Specifically, based on the RGB value of each body surface stain target frame, through the conversion of the HSV color space, the color information of the stain area is extracted, and the hue, saturation and brightness of the stain are identified, etc., to form the stain area color feature; based on the number of pixel points of each body surface stain target frame, the area of each body surface stain target frame is calculated, based on the formula Calculate the edge smoothness of each stain target box on the vehicle body surface. Here, Smoothness is the edge smoothness of the stain target box on the vehicle body surface, Perimeter is the total length of the edge contour in the stain target box on the vehicle body surface, Area is the area of the stain target box on the vehicle body surface, and based on the formula Calculate the contour complexity of each stain target box on the vehicle body surface. Here, Complexity is the contour complexity of the stain target box on the vehicle body surface. Summarize the area, edge smoothness, and contour complexity of the calculated stain target boxes on the vehicle body surface to obtain the shape features of the stain area. Analyze the surface structure of the stain area based on the gray-level co-occurrence matrix (GLCM), and calculate contrast, homogeneity, energy, etc. Among them, contrast reflects the degree of change in pixel gray values in the image, homogeneity reflects the uniformity of image texture, and energy reflects the texture consistency of the image. These features will jointly form the texture features of the stain area. Finally, add the color features, shape features, and texture features of the stain area to a set to obtain a complete multi-dimensional feature set of vehicle body stains, providing data support for subsequent stain classification, area optimization, and cleaning plan design. In summary, through the step-by-step operations of gray-scale distribution recognition, stain comparison, edge detection, and region segmentation, the positioning and recognition of stains are refined, and through multi-dimensional feature extraction, the visual features (color, shape, texture) of stains are comprehensively analyzed. This information will contribute to subsequent clustering analysis and cleaning strategy matching, thereby improving the intelligent level and cleaning effect of car washing.

[0021] Perform clustering division on the multi-dimensional feature set of vehicle body stains to obtain N feature sets of vehicle body stain areas, and at the same time call the car washing plan parameter space according to the target automatic car washing equipment.

[0022] In one embodiment, perform clustering analysis on the extracted multi-dimensional feature set of vehicle body stains. Specifically, classify according to features such as the color, shape, and texture of the stain area, and group similar stain areas into one category. Through the clustering algorithm, divide different stain areas on the vehicle body surface into multiple stain area feature sets, and the stain areas in these feature sets have similar feature performances. In this way, all stain areas on the vehicle body surface can be divided into N area feature sets according to feature similarity. After completing the stain area division, the car washing plan parameter space of the target automatic car washing equipment will also be called according to the unique identifier of the target automatic car washing equipment, such as ID. This parameter space contains information such as the usage method, concentration, and dosage of different cleaning agents. According to the characteristics of different stain areas, the most suitable cleaning plan will be automatically selected to ensure that each stain area can receive the best cleaning treatment. In this way, not only can accurate identification be performed according to stain characteristics, but also targeted cleaning strategies can be formulated to improve car washing efficiency and cleaning effect.

[0023] Further, the present application provides the obtaining of N sets of body stain area features, including: According to the distribution characteristic information of the body stain multi-dimensional feature set, select a range of the number of clusters; sequentially select the number of clusters K within the range of the number of clusters to calculate the clustering error of the body stain multi-dimensional feature set, and obtain the sum of squared errors within clusters for multiple K values; generate a curve of the sum of squared errors within clusters according to the sum of squared errors within clusters for multiple K values, mark the elbow point on the curve of the sum of squared errors within clusters, and determine the target number of clusters K; perform clustering division on the body stain multi-dimensional feature set based on the target number of clusters K to obtain N sets of body stain area features.

[0024] Preferably, according to the distribution characteristics (such as the degree of dispersion of features) of the body stain multi-dimensional feature set, a range of the number of clusters is determined in combination with historical experience. This range refers to the interval of the expected possible number of clusters, which can avoid selecting too many or too few clusters. Subsequently, the number of clusters K is sequentially selected within this range, and clustering operations are performed on the stain feature set through the Euclidean distance to calculate the clustering error for each K value. A commonly used error calculation method is the sum of squared errors within clusters (WSS), that is, calculate the sum of the distances between the sample points within each cluster and the cluster center to reflect the compactness of the samples within each cluster. Then, according to the sum of squared errors within clusters corresponding to multiple calculated K values, a curve of the sum of squared errors is generated. On this curve, the X-axis represents the number of clusters K, and the Y-axis represents the sum of squared errors. By analyzing this curve, an elbow point is searched for, that is, the point where the sum of squared errors begins to slow down. This point usually indicates that the number of clusters reaches a reasonable balance, and adding more clusters will not significantly improve the clustering effect. The elbow point marking determines the target number of clusters K, and this K value represents the number of clusters most suitable for dividing the stain area. For example, when K = 4, the decreasing amplitude of the curve of the sum of squared errors significantly becomes smaller, then K = 4 is the target number of clusters. Finally, according to the determined target number of clusters K, the final clustering division operation is performed. In this process, the body stain multi-dimensional feature set will be divided into K preliminary stain feature areas according to this K value, and then by merging similar areas of the K preliminary stain feature areas, N sets of body stain area features are obtained. Each feature set contains similar stain features, and these stain area feature sets will be used as the basis for subsequent cleaning plan matching and control strategies. Generally speaking, through this process, the system effectively divides the stain areas on the body surface into multiple categories by selecting an appropriate number of clusters K, and each category represents a type of stain with similar features. In this way, corresponding cleaning plans can be formulated according to different stain types to achieve an efficient and accurate cleaning effect.

[0025] Further, the present application provides the performing of clustering division on the body stain multi-dimensional feature set based on the target number of clusters K to obtain N sets of body stain area features, including: Perform iterative clustering on the multi-dimensional body stain feature set based on the target cluster number K to obtain K preliminary stain feature regions; perform feature similarity analysis on the K preliminary stain feature regions according to the body cleaning requirement target, and set a stain feature similarity threshold; obtain the distribution space information of the K preliminary stain feature regions, and merge the regions with adjacent spaces and similar features among the K preliminary stain feature regions based on the stain feature similarity threshold and the distribution space information to obtain N body stain merged regions; calculate the average feature of the N body stain merged regions in sequence based on the multi-dimensional body stain feature set to obtain the N body stain region feature sets.

[0026] Optionally, based on the number of target clusters K, K pixel point features are randomly selected from the multi-dimensional features of the vehicle body stains (such as color, shape, texture, etc.) as the initial clustering centers. These clustering centers represent the initial states of each cluster and will be continuously optimized in subsequent iterations. The selection of the initial clustering centers may be random or based on some heuristic rules (such as selecting representative sample points in the stain feature set). After initializing the clustering centers, each vehicle body stain area is assigned to the closest clustering center through the Euclidean distance. The clustering center with the smallest distance is considered the belonging cluster of the stain area. After all stain areas are assigned to the clusters, the center point of each cluster is recalculated. The center of each cluster is the feature mean of all stain areas within the cluster. The updated clustering centers can better represent the features of the stain areas within the cluster. After the clustering centers are updated, the pixel point clustering operation is performed again until the stop condition is met, that is, the clustering centers no longer change significantly or the change is less than a preset threshold, thus obtaining K preliminary stain feature areas. Each preliminary stain feature area represents a type of stain with similar features. After obtaining the K preliminary stain feature areas, the feature similarity analysis is performed on these areas according to the cleaning requirement objectives of the vehicle body (such as the cleaning intensity required for different stain types). By calculating the similarity between the clustering centers of each preliminary stain feature area, the area similarity between each two areas is obtained. At the same time, a stain feature similarity threshold is set according to the business requirements. This threshold determines which stain features are similar enough to be merged into the same group of areas. Subsequently, the distribution space information of each preliminary stain feature area is obtained, including the position and relative position relationship of the stain area in the image. Based on this distribution space information, the adjacent preliminary stain feature areas are determined, and the area similarity of these adjacent areas is obtained from the calculated multiple area similarities as the stain feature similarity of the adjacent areas. Then, these stain feature similarities are compared with the set stain feature similarity threshold. According to the comparison results, the adjacent preliminary stain feature areas with stain feature similarities less than or equal to the stain feature similarity threshold are merged to form N vehicle body stain merged areas. Each merged area contains one or more similar stain areas. After merging the areas, the feature average value is calculated for each merged stain area in turn. This process involves averaging the features such as color, shape, texture, etc. of all stain samples within each merged area. By calculating the mean of these features, a single value representing the features of each stain merged area can be created for each stain merged area, thus generating a more accurate set of N vehicle body stain area features, providing clear reference data for subsequent cleaning plans, dosing control, etc.In summary, through the above steps, starting from the initial stain feature regions, the system combines feature similarity and spatial distribution information, and finally obtains N merged vehicle body stain regions. These merged regions not only have high feature consistency but also consider the spatial distribution of the stains, making the cleaning requirements for each region clearer and more accurate. This process provides important data support for subsequent steps such as optimizing the cleaning plan and determining the amount of cleaning agent to be dispensed.

[0027] Based on the vehicle washing plan parameter space, analyze the characteristics sets of the N vehicle body stain regions to determine the cleaning plan parameters for the N stain regions.

[0028] In one embodiment, analyze and match the obtained characteristics sets of the N vehicle body stain regions according to the vehicle washing plan parameter space. Specifically, the vehicle washing plan parameter space contains cleaning strategies for different stain types, and these strategies include multiple factors such as the type, concentration, and dispensing amount of the cleaning agent, aiming to provide the most suitable cleaning plan for different types of stain regions. According to the characteristics of each stain region and combining the data in the vehicle washing plan parameter space, analyze the most suitable cleaning plan for each merged region. For example, for a merged region with more oil stains, a stronger cleaning agent can be selected and a higher concentration can be set; while for a merged region with more mud stains, a milder cleaning agent may be selected and a lower concentration may be set. Through this process, specific cleaning plan parameters are determined for each stain region, forming the cleaning plan parameters for the N stain regions. These parameters will be used as the basis for subsequent cleaning agent dispensing control to ensure that different types of stains are cleaned reasonably and effectively.

[0029] Furthermore, this application provides the determination of the cleaning plan parameters for the N stain regions, including: Based on the vehicle washing plan parameter space, perform matching analysis on the characteristics sets of the N vehicle body stain regions respectively to determine the matching cleaning plans for the N stain regions. The matching cleaning plans for the N stain regions include the type of cleaning agent, the concentration of the cleaning agent, and the dispensing amount of the cleaning agent; perform priority analysis on the characteristics sets of the N vehicle body stain regions according to the importance of the cleaning regions to obtain the cleaning priorities of the vehicle body stain regions; based on the cleaning priorities of the vehicle body stain regions, perform influence correction on the matching cleaning plans for the N stain regions to determine the cleaning plan parameters for the N stain regions.

[0030] Optionally, according to the car wash plan parameter space, which includes various cleaning plan options such as detergent types (such as stain removers, degreasers, etc.), detergent concentrations, and detergent dosages, the obtained N sets of body stain area features are subjected to matching analysis. By matching each set of body stain area features with the stain feature ranges corresponding to each cleaning plan in the car wash plan parameter space, N matching cleaning plans for the stain areas are obtained. After determining the preliminary cleaning plan, by calculating the weights based on the cleanliness requirements of each stain area (determined by the position of the stain, such as the front windshield, windows, doors, etc.), the area of the stain, and the type of the stain, an importance index is assigned to each stain area. Then, corresponding priorities are assigned to each stain area according to these importance indexes, so as to obtain the cleaning priorities of each body stain area. The areas with higher cleaning priorities of the body stain areas are cleaned earlier to ensure that important areas and difficult-to-clean stains are treated promptly and thoroughly. Subsequently, based on these cleaning priorities of the body stain areas, the N matching cleaning plans for the stain areas initially determined are corrected and optimized, that is, the parameters of the stain area cleaning plan are adjusted according to the priority of each area and the influence of stain diffusion, ensuring that the most suitable cleaning plan is obtained for each stain area and achieving the best cleaning effect. In summary, through the above steps, the system not only accurately matches the cleaning plan but also adjusts the cleaning plan according to the characteristics and priorities of each stain area, ensuring the efficiency and effect during the cleaning process. In this way, the cleaning plan can be optimized according to the regional importance and actual needs, thereby improving the overall quality of car washing and resource utilization rate.

[0031] Furthermore, the present application provides an impact correction on the N matching cleaning plans for the stain areas based on the cleaning priorities of the body stain areas to determine the parameters of the N stain area cleaning plans, including: Performing a cleaning arrangement on the N sets of body stain area features based on the cleaning priorities of the body stain areas to obtain N sets of area features of the stain cleaning sequences; sequentially evaluating the degrees of stain diffusion influence among the N sets of area features of the stain cleaning sequences to obtain N stain area diffusion influence parameters; and performing a diffusion influence optimization on the N matching cleaning plans for the stain areas based on the N stain area diffusion influence parameters to determine the parameters of the N stain area cleaning plans.

[0032] Optionally, based on the previously obtained cleaning priorities of the vehicle body stain areas, the N vehicle body stain area feature sets are arranged for cleaning. In this process, the vehicle body stain area feature sets corresponding to the stain areas with higher priorities are arranged in the front and given priority for cleaning, while the vehicle body stain area feature sets corresponding to the stain areas with lower priorities are arranged in the back. After the arrangement, N stain cleaning sequence area feature sets are obtained, and these feature sets are sorted according to the cleaning priorities and cleaning requirements, which is convenient for subsequent processing. Subsequently, the impact of stain diffusion between the N stain cleaning sequence area feature sets is evaluated. Here, stain diffusion refers to the impact that a stain area may have on other stain areas during the cleaning process, especially the secondary pollution or stain transfer that may be caused by the cleaning agent or water flow during the cleaning process. To evaluate the impact of this stain diffusion, the reciprocal of the distance between every two stain areas is calculated to obtain the spatial diffusion factor, which is used to quantify the spatial diffusion impact; the preset diffusion ability corresponding to the pollution type is used as the stain type factor, which is used to quantify the diffusion ability of the stain type; the preset cleaning ability of the cleaning agent type is multiplied by the set concentration to obtain the cleaning agent diffusion factor, which is used to quantify the diffusion effect of the cleaning agent; the priority corresponding to the stain cleaning sequence area feature set is used as the sequential diffusion factor, which is used to quantify the diffusion impact of the area to be cleaned first on the subsequent areas; by performing a weighted sum of the spatial diffusion factor, the stain type factor, the cleaning agent diffusion factor, and the sequential diffusion factor, N stain area diffusion impact parameters are obtained, and each stain area diffusion impact parameter represents the degree of impact that the corresponding area may have on other areas during the cleaning process. After that, based on the obtained N stain area diffusion impact parameters, the N stain area matching cleaning plans are optimized. The purpose of this optimization is to reduce the possible diffusion effects during the stain cleaning process and ensure that the cleaning of each stain area does not have an adverse impact on other areas. For example, if the stain diffusion impact of a certain area is relatively large, the stain area matching cleaning plan for that area will be adjusted, such as increasing the concentration of the cleaning agent, increasing the dosage, or cleaning that area first and then cleaning other areas, to ensure that the stains are completely removed and diffusion is avoided. In this way, the cleaning plan parameters for each stain area are optimized for diffusion impact, and finally the optimal cleaning plan is determined as the cleaning plan parameters for the N stain areas. In summary, through the above steps, the system not only considers the cleaning priorities of each stain area, but also evaluates the mutual influence between the stain areas, and optimizes the cleaning plan to avoid unnecessary pollution or a decline in effect. This process ensures that the entire car washing process is more efficient, accurate, and maximizes the cleaning effect and resource utilization rate.

[0033] Based on the N stain area cleaning plan parameters, a control logic association is performed on the cleaning agent dispensing component of the target automatic car washing device to obtain N area cleaning agent dispensing control parameters.

[0034] In one embodiment, based on the determined cleaning solution parameters for N stain areas, control logic association is performed on the cleaning agent dispensing component of the target automated car wash equipment. That is, each cleaning solution parameter for a stain area is transmitted to the control system of the target automated car wash equipment corresponding to that stain area and converted into parameter instructions that can be recognized and called by the control system of the target automated car wash equipment, thereby obtaining N area cleaning agent dispensing control parameters. These area cleaning agent dispensing control parameters can ensure the precise dispensing of the cleaning agent in each stain area, thus improving the cleaning effect and minimizing resource waste.

[0035] Obtain the environmental temperature and humidity data stream in real time, and based on the N area cleaning agent dispensing control parameters and the environmental temperature and humidity data stream, the target automated car wash equipment performs cleaning agent dispensing control and feedback regulation optimization on the target vehicle.

[0036] In one embodiment, the environmental temperature and humidity data stream is monitored and obtained in real time through arranged temperature and humidity sensors to reflect the climate conditions of the current car wash environment. Subsequently, based on these temperature and humidity data, the previously determined N area cleaning agent dispensing control parameters are adjusted and corrected. For example, in a high humidity environment, the cleaning agent may take longer to take effect, and in a high temperature environment, the cleaning agent may evaporate faster. The system dynamically adjusts parameters such as the dispensing amount and concentration of the cleaning agent based on these environmental factors, and applies the corrected area cleaning agent dispensing control parameters to the cleaning agent dispensing control. Through this process, the dispensing control of the cleaning agent can be optimized in real time to ensure the best cleaning effect in each area, and at the same time, the cleaning plan is feedback-regulated according to environmental changes to further improve the car wash efficiency and reduce resource waste.

[0037] Furthermore, the present application provides the cleaning agent dispensing control and feedback regulation optimization for the target vehicle, including: Analyze the influence of the environmental temperature and humidity data stream on the N area cleaning agent dispensing control parameters to obtain N stain cleaning effect influence factors; use the N stain cleaning effect influence factors to adjust and correct the N area cleaning agent dispensing control parameters to determine N cleaning agent dispensing correction control parameters; based on the N cleaning agent dispensing correction control parameters, perform cleaning agent dispensing control and feedback regulation optimization on the target vehicle.

[0038] Preferably, after obtaining the environmental temperature and humidity data stream, the environmental temperature and humidity data stream is used to analyze the influence of the cleaning effect on the N regional cleaner dosing control parameters. Specifically, the current temperature is compared with the reference temperature corresponding to the N regional cleaner dosing control parameters, and the temperature deviation is calculated. This deviation is used for ratio calculation with the reference temperature to obtain the temperature influence factor. Similarly, the same operation is performed on the current humidity to obtain the humidity influence factor, and then the temperature influence factor and the humidity influence factor are added together to obtain the N stain cleaning effect influence factors. Subsequently, 1 is used to subtract each of the N stain cleaning effect influence factors, and then the N calculated differences are multiplied by the corresponding N regional cleaner dosing control parameters to adjust and correct the control parameters, thereby determining the N cleaner dosing correction control parameters. After that, based on the corrected N cleaner dosing correction control parameters, cleaner dosing control is performed on the target vehicle to accurately control the dosing amount, concentration, etc. of the cleaner in each stain area, ensuring the best cleaning effect under different environmental conditions. During the actual dosing process, the cleaning effect is also monitored in real time, and the N cleaner dosing correction control parameters are fine-tuned according to the feedback information. For example, if the cleaning effect in some areas does not meet the expectation, the dosing amount of the cleaner in that area is increased or the type of the cleaner is adjusted until the best cleaning effect is achieved. Through this feedback control process, the cleaner dosing control can be continuously optimized to ensure that the entire car washing process is efficient, energy-saving, and has the best cleaning effect.

[0039] Furthermore, the present application provides the cleaner dosing control and feedback control optimization for the target vehicle based on the N cleaner dosing correction control parameters, including: Performing cleaner dosing control monitoring on the target vehicle using the N cleaner dosing correction control parameters to obtain vehicle cleaning feedback parameters; dynamically adjusting and optimizing the N cleaner dosing correction control parameters based on the vehicle cleaning feedback parameters, and performing closed-loop cleaning control on the target vehicle through the optimized N cleaner dosing correction control parameters.

[0040] Optionally, according to the obtained N detergent dosing correction control parameters, the target vehicle is controlled for detergent dosing. During the cleaning process, the detergent dosing status of each area is continuously monitored, and the dosing amount, concentration, time, and area cleaning effect of the detergent are recorded. Through this monitoring, vehicle cleaning feedback parameters regarding each stain area can be collected, including detergent coverage rate, cleaning effect, and cleaning time. These feedback parameters provide data support for subsequent dynamic regulation and help determine the effectiveness of the current cleaning strategy. Subsequently, the N detergent dosing correction control parameters are dynamically regulated and optimized according to the obtained vehicle cleaning feedback parameters. For example, if the detergent coverage rate in certain areas is low and the cleaning effect is poor, the dosing amount of the detergent will be increased or the concentration of the detergent will be raised to ensure that the stains can be completely removed. For areas that have been sufficiently cleaned, the dosing amount of the detergent can be reduced to avoid wasting resources. After that, according to the optimized N detergent dosing correction control parameters, a closed-loop control of the cleaning will be executed, that is, according to the current optimization result, the detergent dosing for each stain area will be re-controlled, and precise monitoring and adjustment will be carried out to ensure that the cleaning effect of each area reaches the best, and the dosing amount and effect of the detergent maintain the best balance, avoiding waste of resources. At the same time, according to the continuously obtained feedback data, the detergent dosing is continuously adjusted until the expected cleaning effect is finally achieved. During this process, the detergent dosing control parameters will be continuously optimized according to the real-time feedback to ensure that each stain area of the vehicle is fully cleaned. In summary, through detergent dosing control monitoring, dynamic regulation optimization, and cleaning closed-loop control, it is ensured that the detergent dosing for each stain area is always in the best state. This process makes the cleaning effect more precise and efficient, while avoiding waste of resources and ensuring the automation and intelligence of the cleaning process, and ultimately providing customers with high-quality car washing services.

[0041] In summary, the embodiments of the present application at least have the following technical effects: In the embodiments of the present application, first, a high-precision camera is used to collect and obtain the body surface stain image of the target vehicle, and stain feature recognition is performed on the body surface stain image to obtain a multi-dimensional body stain feature set; subsequently, clustering division is performed on the multi-dimensional body stain feature set to obtain N body stain area feature sets, and at the same time, a car wash plan parameter space is called according to the target automatic car wash equipment; then, based on the car wash plan parameter space, cleaning plan analysis is performed on the N body stain area feature sets to determine the cleaning plan parameters for N stain areas; then, based on the cleaning plan parameters for N stain areas, control logic association is performed on the detergent dispensing component of the target automatic car wash equipment to obtain the detergent dispensing control parameters for N areas; finally, the environmental temperature and humidity data stream is monitored in real time, and based on the N area detergent dispensing control parameters and the environmental temperature and humidity data stream, the target automatic car wash equipment performs detergent dispensing control and feedback regulation optimization on the target vehicle. These technical effects together solve the technical problems that existing automatic car wash equipment lacks pertinence in detergent dispensing, cannot accurately control the amount of detergent dispensed according to the actual stain situation of the vehicle and environmental changes, resulting in poor car wash effect and detergent waste, and achieve the technical effects of accurately dividing stain areas according to the body stain characteristics of the vehicle, formulating personalized cleaning plans in combination with the car wash plan parameter space, and adjusting the detergent dispensing in real time according to the environmental temperature and humidity, improving the car wash quality, reducing detergent waste, and realizing intelligent and accurate control.

[0042] Embodiment 2, based on the same inventive concept as the intelligent detergent dispensing control method for an automatic car wash equipment in the foregoing embodiment, as Figure 2 shown, the present application provides an intelligent detergent dispensing control system for an automatic car wash equipment, and the system includes: a stain feature recognition module 11: collecting and obtaining the body surface stain image of the target vehicle through a high-precision camera, and performing stain feature recognition on the body surface stain image to obtain a multi-dimensional body stain feature set; a clustering division module 12: performing clustering division on the multi-dimensional body stain feature set to obtain N body stain area feature sets, and at the same time, calling a car wash plan parameter space according to the target automatic car wash equipment; a cleaning plan analysis module 13: performing cleaning plan analysis on the N body stain area feature sets based on the car wash plan parameter space to determine the cleaning plan parameters for N stain areas; a control logic association module 14: performing control logic association on the detergent dispensing component of the target automatic car wash equipment based on the cleaning plan parameters for N stain areas to obtain the detergent dispensing control parameters for N areas; a dispensing control module 15: monitoring and obtaining the environmental temperature and humidity data stream in real time, and based on the N area detergent dispensing control parameters and the environmental temperature and humidity data stream, the target automatic car wash equipment performs detergent dispensing control and feedback regulation optimization on the target vehicle.

[0043] Further, the stain feature recognition module 11 is also used to execute the following method: Identify the noise characteristics of the vehicle body surface stain image to obtain the vehicle body image noise characteristic information. According to the vehicle body image noise characteristic information, initialize the image filter; use the image filter to perform filtering and denoising processing on the vehicle body surface stain image to obtain a denoised vehicle body surface stain image; according to the specification model information of the target vehicle, obtain a standard vehicle body surface image; based on the standard vehicle body surface image, perform stain recognition and feature extraction on the denoised vehicle body surface stain image to obtain a multi-dimensional vehicle body stain feature set.

[0044] Further, the stain feature recognition module 11 is also used to execute the following method: Use the standard vehicle body surface image as the background area, identify the gray distribution of the background area to obtain the gray threshold of the vehicle body background area; based on the gray threshold of the vehicle body background area, perform stain comparison and recognition on the denoised vehicle body surface stain image to obtain a set of vehicle body surface stain areas; sequentially perform edge detection and region segmentation on the set of vehicle body surface stain areas, and mark to obtain a set of vehicle body surface stain target frames; respectively perform multi-dimensional feature extraction on the set of vehicle body surface stain target frames to obtain a multi-dimensional vehicle body stain feature set, and the multi-dimensional vehicle body stain feature set includes stain area color features, shape features, and texture features.

[0045] Further, the clustering and partitioning module 12 is also used to execute the following method: According to the distribution characteristic information of the multi-dimensional vehicle body stain feature set, select the range of the number of clusters; sequentially select the number of clusters K within the range of the number of clusters to calculate the clustering error of the multi-dimensional vehicle body stain feature set to obtain the sum of squared errors within the clusters for multiple K values; according to the sum of squared errors within the clusters for multiple K values, generate a curve of the sum of squared errors within the clusters, mark the elbow point of the curve of the sum of squared errors within the clusters to determine the target number of clusters K; based on the target number of clusters K, perform clustering and partitioning on the multi-dimensional vehicle body stain feature set to obtain N sets of vehicle body stain area features.

[0046] Further, the clustering and partitioning module 12 is also used to execute the following method: Performing iterative clustering on the multi-dimensional feature set of vehicle body stains based on the target cluster number K to obtain K preliminary stain feature regions; performing feature similarity analysis on the K preliminary stain feature regions according to the target of vehicle body cleaning requirements, and setting a stain feature similarity threshold; obtaining the distribution space information of the K preliminary stain feature regions, and merging the K preliminary stain feature regions with adjacent spaces and similar features based on the stain feature similarity threshold and the distribution space information to obtain N vehicle body stain merging regions; calculating the feature average values of the N vehicle body stain merging regions in sequence based on the multi-dimensional feature set of vehicle body stains to obtain the N vehicle body stain region feature sets.

[0047] Further, the cleaning plan analysis module 13 is further configured to execute the following method: Performing matching analysis on the N vehicle body stain region feature sets respectively based on the car washing plan parameter space to determine N stain region matching cleaning plans, where the N stain region matching cleaning plans include the type of cleaning agent, the concentration of cleaning agent, and the dosage of cleaning agent; performing priority analysis on the N vehicle body stain region feature sets according to the importance of the cleaning region to obtain the cleaning priority of the vehicle body stain regions; performing influence correction on the N stain region matching cleaning plans based on the cleaning priority of the vehicle body stain regions to determine the cleaning plan parameters of the N stain regions.

[0048] Further, the cleaning plan analysis module 13 is further configured to execute the following method: Performing cleaning arrangement on the N vehicle body stain region feature sets based on the cleaning priority of the vehicle body stain regions to obtain N stain cleaning sequence region feature sets; evaluating the stain diffusion influence degree between the N stain cleaning sequence region feature sets in sequence to obtain N stain region diffusion influence parameters; performing diffusion influence optimization on the N stain region matching cleaning plans based on the N stain region diffusion influence parameters to determine the cleaning plan parameters of the N stain regions.

[0049] Further, the dosing control module 15 is further configured to execute the following method: Performing analysis on the influence of cleaning effect on the dosing control parameters of cleaning agents in the N regions based on the environmental temperature and humidity data stream to obtain N stain cleaning effect influence factors; using the N stain cleaning effect influence factors to perform regulation and correction on the dosing control parameters of cleaning agents in the N regions to determine N dosing correction control parameters; performing dosing control and feedback regulation optimization on the target vehicle based on the N dosing correction control parameters.

[0050] Further, the dosing control module 15 is further configured to execute the following method: Execute the cleaning agent delivery control monitoring on the target vehicle by using the N cleaning agent delivery correction control parameters, and obtain the vehicle cleaning feedback parameters; dynamically adjust and optimize the N cleaning agent delivery correction control parameters based on the vehicle cleaning feedback parameters, and perform closed-loop control on the cleaning of the target vehicle by using the optimized N cleaning agent delivery correction control parameters.

[0051] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above specific embodiments of this specification have been described. The processes depicted in the drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0052] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0053] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. An intelligent cleaner dosing control method for automated car washing equipment, characterized in that, The method includes: Collecting the body surface stain image of the target vehicle through a high-precision camera, performing stain feature recognition on the body surface stain image to obtain a multi-dimensional feature set of body stains; Performing clustering division on the multi-dimensional feature set of body stains to obtain N body stain area feature sets, and simultaneously calling the car wash plan parameter space according to the target automatic car wash equipment; Analyzing the cleaning plan for the N body stain area feature sets based on the car wash plan parameter space to determine the cleaning plan parameters for N stain areas; Associating the control logic of the detergent dispensing component of the target automatic car wash equipment based on the cleaning plan parameters for N stain areas to obtain the detergent dispensing control parameters for N areas; Real-time monitoring and obtaining the environmental temperature and humidity data stream, and the target automatic car wash equipment performs detergent dispensing control and feedback regulation optimization on the target vehicle based on the detergent dispensing control parameters for N areas and the environmental temperature and humidity data stream.

2. The intelligent cleaner dosing control method for automated car washing equipment according to claim 1, characterized in that, The obtaining of the multi-dimensional feature set of body stains includes: Performing noise characteristic recognition on the body surface stain image to obtain the body image noise characteristic information, and initializing the image filter according to the body image noise characteristic information; Performing filtering and denoising processing on the body surface stain image using the image filter to obtain the denoised body surface stain image; Obtaining the standard body surface image according to the specification and model information of the target vehicle; Performing stain recognition and feature extraction on the denoised body surface stain image based on the standard body surface image to obtain the multi-dimensional feature set of body stains.

3. The intelligent cleaner dosing control method for automated car washing equipment according to claim 2, characterized in that, The obtaining of the multi-dimensional feature set of body stains includes: Taking the standard body surface image as the background area, performing gray-scale distribution recognition on the background area to obtain the gray-scale threshold of the body background area; Performing stain comparison and recognition on the denoised body surface stain image based on the gray-scale threshold of the body background area to obtain the set of body surface stain areas; Successively performing edge detection and region segmentation on the set of body surface stain areas, and marking to obtain the set of body surface stain target frames; Performing multi-dimensional feature extraction on the set of body surface stain target frames respectively to obtain the multi-dimensional feature set of body stains, and the multi-dimensional feature set of body stains includes stain area color features, shape features, and texture features.

4. The intelligent cleaner dosing control method for automated car washing equipment according to claim 1, characterized in that, The obtaining of N body stain area feature sets includes: Selecting the cluster number range according to the distribution characteristic information of the multi-dimensional feature set of body stains; Successively selecting the cluster number K within the cluster number range to calculate the clustering error of the multi-dimensional feature set of body stains, and obtaining the sum of squared errors within clusters for multiple K values; Generating a curve of the sum of squared errors within clusters according to the sum of squared errors within clusters for multiple K values, performing elbow point marking on the curve of the sum of squared errors within clusters, and determining the target cluster number K; Performing clustering division on the multi-dimensional feature set of body stains based on the target cluster number K to obtain N body stain area feature sets.

5. The intelligent cleaner dosing control method for automated car washing equipment according to claim 4, characterized in that, The performing of clustering division on the multi-dimensional feature set of body stains based on the target cluster number K to obtain N body stain area feature sets includes: Perform iterative clustering on the multi-dimensional feature set of the vehicle body stains based on the target cluster number K to obtain K preliminary stain feature regions; Perform feature similarity analysis on the K preliminary stain feature regions according to the vehicle body cleaning requirement target, and set a stain feature similarity threshold; Obtain the distribution space information of the K preliminary stain feature regions, and merge the regions that are spatially adjacent and feature-similar among the K preliminary stain feature regions based on the stain feature similarity threshold and the distribution space information to obtain N vehicle body stain merged regions; Calculate the feature average values of the N vehicle body stain merged regions in sequence based on the multi-dimensional feature set of the vehicle body stains to obtain the N vehicle body stain region feature sets.

6. The intelligent cleaner dosing control method for automated car washing equipment according to claim 5, characterized in that, The determination of the cleaning scheme parameters for N stain regions includes: Perform matching analysis on the N vehicle body stain region feature sets respectively based on the car washing scheme parameter space to determine the matching cleaning schemes for the N stain regions, and the matching cleaning schemes for the N stain regions include the type of cleaning agent, the concentration of the cleaning agent, and the dosage of the cleaning agent; Perform priority analysis on the N vehicle body stain region feature sets according to the importance of the cleaning regions to obtain the cleaning priorities of the vehicle body stain regions; Perform influence correction on the matching cleaning schemes for the N stain regions based on the cleaning priorities of the vehicle body stain regions to determine the cleaning scheme parameters for the N stain regions.

7. The intelligent cleaner dosing control method for an automatic car washing device according to claim 6, characterized in that, The performing influence correction on the matching cleaning schemes for the N stain regions based on the cleaning priorities of the vehicle body stain regions to determine the cleaning scheme parameters for the N stain regions includes: Perform cleaning arrangement on the N vehicle body stain region feature sets based on the cleaning priorities of the vehicle body stain regions to obtain N stain cleaning sequence region feature sets; Evaluate the stain diffusion influence degrees among the N stain cleaning sequence region feature sets in sequence to obtain N stain region diffusion influence parameters; Perform diffusion influence optimization on the matching cleaning schemes for the N stain regions based on the N stain region diffusion influence parameters to determine the cleaning scheme parameters for the N stain regions.

8. The intelligent cleaner dosing control method for an automatic car washing device according to claim 1, characterized in that, The performing cleaning agent dosing control and feedback regulation optimization on the target vehicle includes: Perform cleaning effect influence analysis on the cleaning agent dosing control parameters for the N regions based on the environmental temperature and humidity data stream to obtain N stain cleaning effect influence factors; Use the N stain cleaning effect influence factors to perform regulation correction on the cleaning agent dosing control parameters for the N regions to determine N cleaning agent dosing correction control parameters; Perform cleaning agent dosing control and feedback regulation optimization on the target vehicle based on the N cleaning agent dosing correction control parameters.

9. The intelligent cleaner dosing control method for an automatic car washing device according to claim 8, characterized in that, The performing cleaning agent dosing control and feedback regulation optimization on the target vehicle based on the N cleaning agent dosing correction control parameters includes: Use the N cleaning agent dosing correction control parameters to perform cleaning agent dosing control monitoring on the target vehicle to obtain vehicle cleaning feedback parameters; Perform dynamic regulation optimization on the N cleaning agent dosing correction control parameters based on the vehicle cleaning feedback parameters, and perform cleaning closed-loop control on the target vehicle through the optimized N cleaning agent dosing correction control parameters.

10. An intelligent cleaner dosing control system for an automatic car washing device, characterized in that, The system is used to execute the intelligent cleaner dispensing control method for the automatic car washing equipment according to any one of claims 1-9, and the system includes: Stain feature recognition module: Acquire the body surface stain image of the target vehicle through a high-precision camera, perform stain feature recognition on the body surface stain image, and obtain a multi-dimensional feature set of body stains; Clustering and partitioning module: Perform clustering and partitioning on the multi-dimensional feature set of body stains, obtain N feature sets of body stain regions, and at the same time call the car washing scheme parameter space according to the target automatic car washing equipment; Washing scheme analysis module: Analyze the washing scheme for the N feature sets of body stain regions based on the car washing scheme parameter space to determine the washing scheme parameters for N stain regions; Control logic association module: Perform control logic association on the cleaner dispensing components of the target automatic car washing equipment based on the washing scheme parameters for N stain regions to obtain the cleaner dispensing control parameters for N regions; Dispensing control module: Real-time monitor and obtain the environmental temperature and humidity data stream, and based on the cleaner dispensing control parameters for N regions and the environmental temperature and humidity data stream through the target automatic car washing equipment, perform cleaner dispensing control and feedback regulation optimization on the target vehicle.

Images