An intelligent dust cleaning method for a fully enclosed driving sweeper

Abstract

The present application relates to the field of image recognition technology, and provides an intelligent dust cleaning method for a fully enclosed driving sweeper. The method comprises: collecting dirty images of the area to be cleaned; performing visual saliency recognition analysis on the dirty images to obtain dirty blocks; performing dry-wet discrimination analysis on the multi-dimensional image feature information of the dirty blocks to obtain discrimination results; when the discrimination result shows dry garbage pollution, introducing a predetermined dirt assessment interference function to perform interference impact analysis on the initial dirt index to obtain an effective dirt index; matching the corresponding cleaning plan based on the effective dirt index, and performing the cleaning operation using a target sweeper. The present application solves the technical problem that existing cleaning equipment is difficult to accurately distinguish dirt types and quantify pollution levels based on a single image recognition technology in dry and wet mixed garbage scenarios, resulting in poor adaptability of cleaning strategies and serious waste of resources, and achieves the technical effect of improving the matching accuracy and execution efficiency of cleaning plans.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to an intelligent dust cleaning method for a fully enclosed driving sweeper. Background Art

[0002] In closed or semi-enclosed environments like industrial plants, warehouses and logistics centers, and large parking lots, floor cleaning faces complex environmental challenges. On the one hand, mixed dry and wet waste (such as metal debris mixed with oil stains, dust and condensation) is common in the operating area, and traditional sweepers rely on single sensors or basic image recognition technology to struggle to accurately distinguish pollutant types. On the other hand, changes in ambient lighting, differences in floor materials, and fluctuations in temperature and humidity can easily lead to errors in dirt identification, directly affecting the adaptability of cleaning strategies and causing problems such as water waste and incomplete cleaning. Existing solutions often use fixed cleaning modes or rely on simple grayscale thresholds to determine pollutants. These solutions are unable to achieve intelligent decision-making in dynamic scenarios and are prone to misjudgment when dealing with complex situations such as reflective surfaces and shadow interference.

[0003] In this context, there is an urgent need for a cleaning decision-making system that integrates high-precision visual perception and environmental interference correction, and improves the scene adaptability and resource utilization of fully enclosed driving sweepers through multi-dimensional feature fusion and dynamic evaluation mechanism. Summary of the Invention

[0004] This application provides an intelligent dust cleaning method for a fully enclosed driving sweeper, aiming to solve the technical problem that existing cleaning equipment is difficult to accurately distinguish the type of dirt and quantify the degree of pollution based on a single image recognition technology in a dry and wet mixed garbage scenario, resulting in poor adaptability of cleaning strategies and serious waste of resources. The application achieves the technical effect of combining visual saliency recognition with multi-dimensional image feature analysis to achieve precise positioning of dirty blocks and intelligent dry and wet discrimination, and dynamically corrects the dirt index based on the interference function, thereby improving the matching accuracy and execution efficiency of cleaning plans.

[0005] The present application provides an intelligent dust cleaning method for a fully enclosed driving sweeper, the method comprising: dynamically acquiring a dirty image of an area to be cleaned by an image collector mounted on a target sweeper; performing visual saliency recognition analysis on the dirty image to obtain a dirty block; performing dry-wet discrimination analysis on multi-dimensional image feature information of the dirty block to obtain a discrimination result; when the discrimination result shows that the area to be cleaned is contaminated by dry garbage, introducing a predetermined dirt assessment interference function to perform interference influence analysis on an initial dirt index to obtain an effective dirt index; matching a corresponding cleaning plan based on the effective dirt index, and performing cleaning of the area to be cleaned by using the target sweeper according to the cleaning plan.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The above-mentioned intelligent dust cleaning method for a fully enclosed driving sweeper first uses the on-board image collector to capture the dirty image of the area to be cleaned in real time, providing a data basis for subsequent analysis. Subsequently, the collected dirty image is visually recognized to extract the significant dirty blocks to ensure that the polluted area is focused. After that, the dirty blocks are distinguished between dryness and wetness based on the multi-dimensional image feature information to clarify the type of pollution. If the judgment result is dry garbage pollution, a predetermined dirt assessment interference function is introduced to comprehensively consider environmental factors to correct the initial dirt index and obtain a more accurate effective dirt index. Finally, the corresponding cleaning plan is matched according to the effective dirt index to guide the sweeper to perform targeted cleaning operations to ensure the best cleaning effect. This method improves the adaptability and cleaning efficiency of sweepers in complex environments through intelligent analysis and dynamic decision-making.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0010] Figure 1 The present invention is a flow chart of an intelligent dust cleaning method for a fully enclosed driving sweeper in one embodiment.

[0011] Figure 2 The present invention is a flow chart of obtaining dirty blocks in an intelligent dust cleaning method of a fully enclosed driving sweeper in one embodiment. DETAILED DESCRIPTION

[0012] The embodiment of the present application provides an intelligent dust cleaning method for a fully enclosed driving sweeper to solve the technical problem that in the existing cleaning equipment, in the dry and wet mixed garbage scene, it is difficult to accurately distinguish the type of dirt and quantify the degree of pollution based on a single image recognition technology, resulting in poor adaptability of the cleaning strategy and serious waste of resources. The method achieves the technical effect of accurately locating the dirty blocks and intelligently distinguishing between dry and wet areas by combining visual saliency recognition with multi-dimensional image feature analysis, and dynamically correcting the dirt index based on the interference function, thereby improving the matching accuracy and execution efficiency of the cleaning plan.

[0013] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily 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] Examples, such as Figure 1 As shown, the present application provides an intelligent dust cleaning method for a fully enclosed driving sweeper, the method comprising:

[0016] The dirty image of the area to be cleaned is obtained by dynamically collecting the image of the area to be cleaned through the image collector mounted on the target sweeper.

[0017] In an embodiment of the present application, first, an image collector (such as a high-definition camera or an infrared sensor) installed on the body of the target sweeper is used to capture the ground image of the area to be cleaned in real time. These image collectors can cover all areas on the path of the target sweeper in the area to be cleaned in a dynamic scanning manner to ensure monitoring without blind spots. This target sweeper is an automatic cleaning device that integrates vacuuming, sweeping, and water spraying functions. It is designed for outdoor environments and is suitable for ground cleaning work in a variety of public places. Its application scenarios include outdoor streets, roads, community properties, indoor and outdoor floors of factories, peripheries of large shopping malls, railway stations, bus stations, large property plazas, and peripheries of airports. The device has a strong dust suction capacity, a cleaning width of up to 1900mm, and is equipped with internationally advanced cleaning and throwing technology. The theoretical value of the dust box utilization rate can reach 100%. The images collected by the image collector carried by the target sweeper not only contain visible dirt on the ground (such as dust, oil stains, debris, etc.), but also capture complex visual information caused by differences in lighting, shadows, or ground materials. This step generates high-resolution image data of the area to be cleaned, which is then stored in the dirt image, providing the basis for subsequent dirt identification and analysis. This dynamic acquisition method adapts to the movement of the sweeper in different environments, ensuring a real-time and comprehensive cleaning process.

[0018] Perform visual saliency recognition analysis on the dirty image to obtain dirty blocks.

[0019] In one embodiment, after acquiring a dirty image, a visual saliency analysis is performed on the image to accurately extract dirty blocks. This process first segments the entire dirty image into multiple small blocks, forming a block set. Feature extraction is then performed on each block in the block set based on a preset visual saliency metric to obtain relevant parameter information. Subsequently, a visual parameter map is generated based on the correspondence between the blocks and the parameters. This parameter map is then subjected to a differential analysis between the center and surrounding areas to generate a saliency map for each block. Finally, these saliency maps are superimposed and fused to create a complete dirt saliency map, which is then used to identify dirty blocks within the image. This process, through multi-level feature extraction and differential analysis, enables precise location and identification of dirty areas, providing accurate input for subsequent dry / wet discrimination and cleaning decisions.

[0020] Further, if Figure 2 As shown, the present application provides a method for performing visual saliency recognition analysis on the dirty image to obtain dirty blocks, including:

[0021] The method comprises extracting a first saliency indicator from predetermined visual saliency indicators; performing block processing on the dirty image to obtain a dirty block set, wherein the dirty block set includes a first block; collecting features of the first block based on the first saliency indicator to obtain a first indicator parameter; constructing a first parameter visual graph of the first saliency indicator based on a first correspondence between the first block and the first indicator parameter; performing center-periphery difference analysis on the first parameter visual graph to obtain a first saliency map; superimposing and fusing the first saliency maps to obtain a dirty saliency map of the dirty image; and determining the dirty block based on the dirty saliency map.

[0022] Preferably, when performing visual saliency recognition of a dirty image, first, from predefined visual saliency indicators (such as brightness, color, direction, etc.), the first visual indicator is selected as the first saliency indicator. This first saliency indicator will serve as the basis for subsequent analysis. Subsequently, the entire dirty image is divided into multiple small blocks, such as the first block, the second block, etc., according to the preset segmentation unit length, wherein the segmentation unit length is determined by the image resolution and actual application requirements. By adding these segmented blocks to a set at a time, a dirty block set is formed. Each block in the dirty block set is part of the image. This block processing helps to perform localized analysis of the image and improve the accuracy of dirt recognition. Afterwards, based on the first saliency indicator, feature extraction is performed on the first block to obtain the first indicator parameter. For example, if the first saliency indicator is brightness, the RGB value of each pixel in the first block will be obtained. By inputting the RGB value into the brightness calculation formula, the brightness value of each pixel is obtained. The brightness calculation formula is specifically as follows: L = wR ·R+w G ·G+w B ·B; where L represents the brightness value, R, G, and B represent the red channel value, green channel value, and blue channel value respectively, and w R 、w G 、w B The weights for the red, green, and blue channels are 0.299, 0.587, and 0.114, respectively. The calculated brightness values ​​of these pixels serve as the first indicator parameter of the first saliency index. If the first saliency index is color, the RGB values ​​of each pixel in the first block are directly stored as the first indicator parameter. If the first saliency index is direction, the gradient direction of each pixel is calculated using the Sobel operator and stored as the first indicator parameter. After obtaining the first indicator parameter, the correspondence between the first block and the first indicator parameter is combined. That is, each parameter in the first indicator parameter is mapped to the corresponding pixel in the first block, thereby generating a first parameter visual graph. This first parameter visual graph intuitively displays the distribution of the visual features of the block, providing visualization support for subsequent analysis. Then, with each pixel as the center, the central area and surrounding area of ​​each pixel are determined based on the pre-set central area range (such as 3*3, 5*5, etc.) and surrounding area range (such as 7*7, 9*9, etc.). The first indicator parameter in each determined area is averaged to obtain the first indicator parameter mean of each area. The absolute difference between the first indicator parameter mean of the central area and the first indicator parameter mean of the surrounding area is calculated, and then normalized using the maximum-minimum method to obtain the saliency of each pixel. These saliencies will together form the first saliency map, which represents the saliency of each pixel in the image. Finally, the first saliency map is superimposed and fused with other saliency maps (such as those generated based on other visual features). That is, for each pixel position, the saliency of each indicator is weighted using the weight of each indicator to generate a comprehensive dirt saliency map. This map integrates the analysis results of multiple visual features and can more comprehensively reflect the dirt areas in the image. Based on this dirt saliency map, the dirty blocks are accurately located by setting a threshold, providing a basis for subsequent processing. This process achieves accurate recognition of dirty images through steps such as extracting visual features, block processing, feature collection, visualization analysis, difference calculation, image fusion and block positioning.

[0023] Furthermore, the present application provides that the predetermined visual saliency index at least includes a visual brightness index, a visual color index, and a visual direction index.

[0024] Optionally, the predetermined visual saliency indicators include at least a visual brightness indicator, a visual color indicator, and a visual direction indicator. The visual brightness indicator reflects the brightness variations between different regions in the image; dirty regions typically have a significant brightness difference from the surrounding ground. The visual color indicator reflects the color differences between different regions in the image; dirty regions typically have a significant color difference from the surrounding ground. The visual direction indicator reflects the directionality of the texture or edges between different regions in the image; dirty regions typically have a significant texture difference from the surrounding ground. By analyzing these indicators, dirty regions, such as stains, dust, and oil stains, can be efficiently identified on the ground, thereby improving cleaning efficiency and quality.

[0025] Furthermore, the present application provides a method for determining the dirty block according to the dirt saliency map, including:

[0026] Obtain an arbitrary block and record the arbitrary block as the central block; form a set of surrounding blocks for the central block and randomly extract any surrounding block from the set of surrounding blocks; compare and obtain a parameter difference between the central block and the arbitrary surrounding blocks; when the parameter difference is within a predetermined parameter difference threshold, add the arbitrary block to the dirty block.

[0027] Optionally, after obtaining the dirt saliency map, a block is randomly selected from the dirt saliency map as an arbitrary block. This arbitrary block will be recorded as the central block, and the blocks adjacent to the central block are extracted to form a set of surrounding blocks. Subsequently, a block is randomly selected from the set of surrounding blocks and recorded as an arbitrary surrounding block. The significance mean of the central block and the significance mean of the arbitrary surrounding blocks are calculated, and the absolute difference between the two means is calculated. If the calculated absolute difference is greater than or equal to the predetermined parameter difference threshold, it means that there is a significant difference between the central block and the surrounding blocks. At this time, the arbitrary block corresponding to the central block will be added to the dirty block. Repeat the above process until all blocks in the dirt saliency map are compared as central blocks. In summary, by comparing the parameter difference between the central block and the surrounding blocks, the dirty area on the ground can be effectively detected, and the cleaning accuracy and efficiency of the sweeper can be improved.

[0028] Performing dry-wet discrimination analysis on the multi-dimensional image feature information of the dirty block to obtain a discrimination result.

[0029] In one embodiment, after obtaining a contaminated area, image texture features are extracted from the multidimensional image features of the contaminated area. This is a key indicator for determining the nature of the contamination. Subsequently, a pre-set texture threshold is used as the classification standard for dry and wet waste. By comparing the extracted image texture features with the texture threshold, the current contamination is determined to be dry or wet waste, and a judgment result is generated based on the determined category. This process, by quantifying image texture features and combining them with pre-set judgment criteria, achieves automated and intelligent classification of contamination properties, providing an accurate basis for subsequent cleaning operations.

[0030] Furthermore, the present application provides a method for performing dry-wet discrimination analysis on the multi-dimensional image feature information of the dirty block to obtain a discrimination result, including:

[0031] Extracting the image texture feature value from the multidimensional image feature information; reading the predetermined texture limit value, and combining the predetermined texture limit value and the image texture feature value to obtain the discrimination result; wherein, reading the predetermined texture limit value, and combining the predetermined texture limit value and the image texture feature value to obtain the discrimination result, including: when the image texture feature value is greater than the predetermined texture limit value, the area to be cleaned is contaminated by dry garbage; when the image texture feature value is less than the predetermined texture limit value, the area to be cleaned is contaminated by wet garbage.

[0032] Preferably, when performing dry and wet discrimination, the image texture feature value, such as the gradient amplitude (calculated by the Sobel operator), is first extracted from the multi-dimensional image feature information (such as brightness, color, direction, etc.) of the dirty block. Subsequently, a preset texture limit, such as the gradient amplitude limit, is read from a preset database or configuration file. This limit is set based on historical experience and expert decision-making, and is used to distinguish the texture feature threshold of dry garbage from wet garbage. Afterwards, the extracted image texture feature value is compared with the predetermined texture limit. If the image texture feature value is greater than the predetermined texture limit, it means that the dirt in the area has a higher granularity, and it is therefore determined to be dry garbage pollution. If the image texture feature value is less than the predetermined texture limit, it means that the dirt in the area has a lower granularity, and it is therefore determined to be wet garbage pollution. Through this process, the nature of the dirt in the area to be cleaned can be quickly and accurately judged, providing a scientific basis for subsequent cleaning operations.

[0033] When the discrimination result shows that the area to be cleaned is contaminated by dry garbage, a predetermined contamination assessment interference function is introduced to perform interference impact analysis on the initial contamination index to obtain an effective contamination index.

[0034] In one embodiment, when feature analysis identifies the area to be cleaned as being contaminated by dry waste, a predetermined contamination assessment interference function is introduced to correct the initial contamination index. This initial contamination index, determined based on multidimensional image feature information and the material of the area to be cleaned, is used to preliminarily assess the degree of contamination. Subsequently, the initial contamination index is input into the predetermined contamination assessment interference function to correct the contamination index. This balances the influence of light intensity, light direction, and ambient temperature on the contamination assessment, resulting in an effective contamination index that better supports dry waste contamination assessment and cleaning decisions.

[0035] Furthermore, the present application provides that when the discrimination result shows that the area to be cleaned is contaminated by dry garbage, a predetermined contamination assessment interference function is introduced to perform interference impact analysis on the initial contamination index to obtain an effective contamination index, including:

[0036] Extract the image tone feature value from the multidimensional image feature information; perform weighted analysis on the image tone feature value and the image texture feature value to obtain an image feature value; obtain the regional material of the area to be cleaned, and match the material image feature value of the regional material; compare the image feature value with the material image feature value to obtain a contrast feature difference, and record the contrast feature difference as the initial dirtiness index.

[0037] Preferably, color features are extracted from the multidimensional image feature information and converted from the RGB color space to the HSV color space to separate hue, saturation, and brightness information. The color space conversion is performed using a mathematical formula, including normalizing the RGB values, calculating maximum and minimum values, and calculating the hue. Subsequently, the calculated hue value is used as the image hue feature value, and the image hue feature value and image texture feature value are normalized using the maximum and minimum method to bring them to the same dimension. The image hue feature value and image texture feature value are then weighted and calculated based on weights determined based on actual needs and expert decision-making to obtain an image feature value. Subsequently, the regional image of the area to be cleaned is input into a material classification channel for labeling, and the material type of the area to be cleaned, such as cement, asphalt, metal, wood, etc., is obtained. The standard hue feature value and texture feature value of the material are then obtained from a predefined material feature library, and the standard hue feature value and texture feature value are weighted and calculated using the weights determined above to obtain the material image feature value. Then, by comparing the image eigenvalues ​​and the material image eigenvalues, the difference between the image eigenvalues ​​and the material image eigenvalues ​​is calculated, and the difference is used as the initial dirtiness index. This index reflects the difference between the area to be cleaned and the standard material, providing a quantitative basis for subsequent dirtiness assessment and cleaning decisions.

[0038] For the material classification pipeline, sample region image data with regional material identifiers is first collected. The sample region image data is then partitioned into training and validation sets. Subsequently, a convolutional neural network (CNN) is used to construct the material classification pipeline, consisting of an input layer, convolutional layers, activation functions, pooling layers, fully connected layers, and an output layer. The weights of the convolutional and fully connected layers are initialized using random initialization methods (such as Xavier initialization or He initialization), and the bias terms are initialized to 0 or small random numbers. The region images from the training set are then fed into the CNN. The CNN passes through the convolutional layers, activation functions, pooling layers, fully connected layers, and output layers in sequence. The predicted probabilities for each class are calculated, and the cross-entropy loss function is used to calculate the difference between the predicted results and the true labels. The backpropagation algorithm then calculates the gradient of the loss with respect to the weights layer by layer, and the Adam optimizer is used to update the weights to minimize the loss function. This process of forward propagation, loss calculation, backpropagation, and weight update is repeated until the maximum number of iterations is reached or the loss converges. Finally, the validation set is used to test the channel performance and calculate the accuracy of material classification. If the accuracy meets the expected requirements, the parameters of the current material classification channel are saved. If the accuracy does not meet the expected requirements, the hyperparameters such as learning rate, batch size, and network depth are adjusted and the model is retrained.

[0039] Furthermore, the present application provides the following expression for the predetermined dirt assessment interference function:

[0040] Among them, I eff refers to the effective dirt index, I0 refers to the initial dirt index, F interf Refers to the dirt assessment interference coefficient, μ i is the weight of the i-th interference factor in the predetermined interference factor set, satisfying f i is the interference coefficient of the i-th interference factor, ranging from 0 to 1, indicating the degree of interference of this factor on the dirtiness index.

[0041] Preferably, the predetermined dirt assessment interference function is used to adjust the initial dirt index to obtain an effective dirt index, thereby more accurately assessing the degree of dirt that the sweeper needs to clean. The predetermined dirt assessment interference function is specifically as follows: Among them, I eff It refers to the effective dirtiness index, which is the dirtiness index finally used for cleaning decision-making and is the value corrected by interference factors; I0 refers to the initial dirtiness index, which is the original calculated dirtiness index and has not been corrected by environmental factors; F interf It refers to the pollution assessment interference coefficient, which represents the comprehensive impact of all environmental interference factors; μ iis the weight of the i-th interference factor in the predetermined interference factor set, indicating the importance of this factor relative to other factors, and the sum of all weights is 1; f i is the interference coefficient of the i-th interference factor, ranging from 0 to 1, indicating the degree of interference with the dirtiness index. It is calculated by comparing the interference factor's value with the standard value. n is the total number of interference factors, such as light, temperature, and humidity. The effective dirtiness index is determined by the weight and influence of environmental factors. If certain environmental factors (such as light intensity and humidity) have a significant impact on the dirtiness index, the effective dirtiness index will be reduced, making the cleaning strategy more accurate.

[0042] Furthermore, the present application provides that the predetermined interference factor set includes light intensity, light direction, ambient temperature, ambient humidity and ambient pollution.

[0043] Optionally, the predetermined set of interference factors includes at least light intensity, light direction, ambient temperature, ambient humidity and ambient pollution. These factors will have a significant impact on the dirt assessment results. Among them, light intensity refers to the brightness of the light irradiated on the surface of the object. Strong light may make the dirt more obvious, while weak light may cover up some of the dirt. It can be directly measured by a light sensor (such as a photometer); light direction refers to the angle at which the light irradiates the surface of the object. Different light directions will affect the visual effect of the dirt, for example, producing shadows or reflections. It can be directly measured using a light direction sensor or a multi-angle light measurement device; ambient temperature refers to the temperature of the surrounding air. Temperature changes will affect the physical state of the stains. For example, high temperature causes grease to diffuse, and low temperature causes stains to solidify. It can be directly measured using a temperature sensor; ambient humidity refers to the water vapor content in the air. Humidity changes will affect the adhesion and cleaning difficulty of the stains. For example, high humidity makes dust more easily adhered. It can be directly measured by a humidity sensor; environmental pollution refers to the concentration of pollutants in the air. The degree of pollution directly affects the accumulation speed and severity of dirt. It can be directly obtained through an environmental monitoring station or an air quality monitoring network. By comprehensively considering these interference factors and their weights and coefficients, the degree of soiling can be assessed more accurately, leading to the development of more effective cleaning strategies.

[0044] Furthermore, the present application provides for adding a water-oil ratio to the predetermined interference factor set when the discrimination result shows that the area to be cleaned is contaminated by wet garbage.

[0045] Optionally, when the judgment result shows that the area to be cleaned is contaminated by wet garbage, the characteristics of the wet garbage (such as moisture and grease content) will have a significant impact on the dirt assessment and cleaning strategy. Therefore, the water-oil ratio needs to be added to the predetermined interference factor set to more accurately assess the degree of dirtiness and formulate a cleaning plan. The water-oil ratio refers to the relative content of water and grease in wet garbage. A high moisture content may cause stains to spread more easily and increase the difficulty of cleaning. Especially when water is mixed with dust or other pollutants, a high grease content may make stains more difficult to remove, especially in low-temperature environments where grease easily solidifies and forms stubborn stains. This can be determined by the saturation distribution of the image, that is, the saturation of each pixel is compared with the water-oil distinction threshold to determine the pixel ratio of the low-saturation area (presumed water stains) to the high-saturation area (presumed oil stains). This pixel ratio is the water-oil ratio.

[0046] A corresponding cleaning plan is matched based on the effective dirtiness index, and the target sweeper is used to clean the area to be cleaned according to the cleaning plan.

[0047] In one embodiment, after obtaining the effective dirtiness index, a corresponding cleaning plan is matched according to the index, wherein the cleaning plan is a different cleaning strategy divided according to the effective dirtiness index, which is usually divided into multiple levels, for example, light pollution (I eff ≤0.3), simple cleaning, use normal cleaning mode; moderate pollution (0.3 less than I eff ≤0.6), deep cleaning, use enhanced cleaning mode, and may require local focused cleaning; heavy pollution (I eff >0.6), a comprehensive cleaning using a high-intensity cleaning mode may require multiple cleanings or special cleaning agents. Subsequently, based on the matching cleaning plan, the target sweeper is adjusted to the corresponding operating mode for cleaning, ensuring that the sweeper can select the most appropriate cleaning method based on the type and severity of pollution, improving cleaning efficiency and effectiveness.

[0048] In summary, the embodiments of the present application have at least the following technical effects:

[0049] The embodiment of the present application first obtains a dirty image of the area to be cleaned by dynamically acquiring it through an image collector mounted on a target sweeper; performs visual saliency recognition analysis on the dirty image to obtain a dirty block; performs dry-wet discrimination analysis on the multi-dimensional image feature information of the dirty block to obtain a discrimination result; when the discrimination result shows that the area to be cleaned is contaminated by dry garbage, introduces a predetermined dirt assessment interference function to perform interference impact analysis on the initial dirt index to obtain an effective dirt index; matches the corresponding cleaning plan based on the effective dirt index, and uses the target sweeper to clean the area to be cleaned according to the cleaning plan. These technical effects jointly solve the technical problem that existing cleaning equipment cannot accurately distinguish the type of dirt and quantify the degree of pollution based on a single image recognition technology in a dry and wet mixed garbage scene, resulting in poor adaptability of the cleaning strategy and serious waste of resources. The technical effect is achieved by combining visual saliency recognition with multi-dimensional image feature analysis to achieve accurate positioning of the dirty block and intelligent dry-wet discrimination, and dynamically correcting the dirt index based on the interference function, thereby improving the matching accuracy and execution efficiency of the cleaning plan.

[0050] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0051] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0052] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. An intelligent dust cleaning method for a fully enclosed driving sweeper, characterized in that: include: The dirty image of the area to be cleaned is obtained by dynamically collecting the image collector mounted on the target sweeper; Performing visual saliency recognition analysis on the dirty image to obtain dirty blocks; Performing dry-wet discrimination analysis on the multi-dimensional image feature information of the dirty block to obtain a discrimination result; When the discrimination result shows that the area to be cleaned is contaminated by dry garbage, a predetermined contamination assessment interference function is introduced to perform interference impact analysis on the initial contamination index to obtain an effective contamination index; Matching a corresponding cleaning plan based on the effective dirtiness index, and performing cleaning of the area to be cleaned using the target sweeper according to the cleaning plan; The multi-dimensional image feature information of the dirty block is subjected to dry-wet discrimination analysis to obtain a discrimination result, including: Extracting image texture feature values ​​from the multidimensional image feature information; Reading a predetermined texture limit value, and combining the predetermined texture limit value with the image texture feature value to obtain the discrimination result; The method further comprises: reading a predetermined texture limit value, and combining the predetermined texture limit value with the image texture feature value to obtain the discrimination result, including: when the image texture feature value is greater than the predetermined texture limit value, the area to be cleaned is contaminated by dry garbage; and when the image texture feature value is less than the predetermined texture limit value, the area to be cleaned is contaminated by wet garbage; The expression of the predetermined dirt assessment interference function is as follows: in, Refers to the effective dirtiness index, Refers to the initial dirtiness index, Refers to the dirt assessment interference coefficient, It is the first of the predetermined interference factors. The weight of the interference factors satisfies , f i It is The interference coefficient of an interference factor ranges from 0 to 1, indicating the degree of interference of the factor on the dirtiness index.

2. The intelligent dust cleaning method for a fully enclosed driving sweeper according to claim 1, characterized in that: Performing visual saliency recognition analysis on the dirty image to obtain dirty blocks includes: extracting a first saliency indicator from predetermined visual saliency indicators; Performing block processing on the dirty image to obtain a dirty block set, wherein the dirty block set includes a first block; Collecting features of the first block based on the first significance indicator to obtain a first indicator parameter; constructing a first parameter visual graph of the first significance indicator based on a first correspondence between the first block and the first indicator parameter; Performing a center-periphery difference analysis on the first parameter visualization map to obtain a first significance map; superimposing and fusing the first saliency maps to obtain a dirt saliency map of the dirt image; The dirty block is determined according to the dirt saliency map.

3. The intelligent dust cleaning method for a fully enclosed driving sweeper according to claim 2, characterized in that: The predetermined visual saliency index includes at least a visual brightness index, a visual color index, and a visual direction index.

4. The intelligent dust cleaning method for a fully enclosed driving sweeper according to claim 2, characterized in that: Determining the dirty block according to the dirt saliency map includes: Obtain an arbitrary block, and record the arbitrary block as a central block; Forming a set of surrounding blocks of the central block, and randomly extracting any surrounding block from the set of surrounding blocks; Comparing and obtaining parameter differences between the central block and any surrounding blocks; When the parameter difference value is within a predetermined parameter difference threshold, the arbitrary block is added to the dirty blocks.

5. The intelligent dust cleaning method for a fully enclosed driving sweeper according to claim 1, characterized in that: When the discrimination result shows that the area to be cleaned is contaminated by dry garbage, a predetermined contamination assessment interference function is introduced to perform interference impact analysis on the initial contamination index to obtain an effective contamination index, including: extracting an image hue feature value from the multidimensional image feature information; Performing weighted analysis on the image hue feature value and the image texture feature value to obtain an image feature value; Acquire the area material of the area to be cleaned and match the material image feature value of the area material; The image feature value is compared with the material image feature value to obtain a comparison feature difference, and the comparison feature difference is recorded as the initial dirt index.

6. The intelligent dust cleaning method for a fully enclosed driving sweeper according to claim 1, characterized in that: The predetermined interference factor set includes light intensity, light direction, ambient temperature, ambient humidity and ambient pollution.

7. The intelligent dust cleaning method for a fully enclosed driving sweeper according to claim 6, characterized in that: When the discrimination result shows that the area to be cleaned is contaminated by wet garbage, the water-oil ratio is added to the predetermined interference factor set.

Images