Urban road sweeping quality evaluation method and device and storage medium

Abstract

The invention discloses an urban road sweeping quality evaluation method and device and a storage medium. The method comprises the steps of obtaining a swept urban road image; preprocessing the urban road image to obtain a target road image; feature extraction processing is carried out on the target road image to obtain target features, and the target features comprise edge features, corner features, texture features, color features or shape features; classifying and identifying the target road image to obtain a road garbage type, a road garbage quantity and road garbage distribution information; calculating a peak signal-to-noise ratio and a structural similarity index according to the target road image and the road image before cleaning; and according to the road garbage type, the road garbage quantity, the road garbage distribution information, the peak signal-to-noise ratio and the structural similarity index, performing sweeping quality evaluation to obtain a sweeping quality evaluation result. According to the invention, cleaning quality evaluation is realized, and accuracy is improved. The method can be widely applied to the technical field of urban environment maintenance.

Description

Technical Field

[0001] The present invention relates to the technical field of urban environment maintenance, and in particular to a method, device and storage medium for evaluating the quality of urban road cleaning. Background Art

[0002] With the rapid development of urbanization and the increase in urban population density, the cleaning and maintenance of urban roads have become particularly important. However, the current urban road cleaning and sanitation work mainly relies on manual labor. After the cleaning is completed, the traditional cleaning quality evaluation method mainly relies on manual observation and simple measurement, which is highly subjective, lacks scientificity, and has low evaluation accuracy, making it difficult to fully and objectively reflect the actual effect of cleaning and sanitation work.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention

[0004] The embodiments of the present invention provide a method, device and storage medium for evaluating the quality of urban road cleaning, which effectively improve the accuracy.

[0005] On the one hand, an embodiment of the present invention provides a method for evaluating the quality of urban road cleaning, comprising the following steps:

[0006] Acquire the cleaned urban road image;

[0007] Preprocessing the urban road image to obtain a target road image;

[0008] Performing feature extraction processing on the target road image to obtain target features, wherein the target features include edge features, corner features, texture features, color features or shape features;

[0009] According to the target features, the target road image is classified and identified to obtain road garbage type, road garbage quantity and road garbage distribution information;

[0010] Calculating a peak signal-to-noise ratio and a structural similarity index according to the target road image and the road image before cleaning;

[0011] A cleaning quality evaluation is performed based on the road garbage type, the road garbage amount, the road garbage distribution information, the peak signal-to-noise ratio and the structural similarity index to obtain a cleaning quality evaluation result.

[0012] In some embodiments, the preprocessing of the urban road image to obtain the target road image includes:

[0013] Performing denoising processing on the urban road image by using Gaussian filtering or median filtering to obtain a first road image;

[0014] Performing contrast enhancement processing on the first road image by using histogram equalization to obtain a second road image;

[0015] performing size normalization processing on the second road image to obtain a third road image;

[0016] The third road image is normalized to obtain the target road image.

[0017] In some embodiments, the step of performing feature extraction processing on the target road image to obtain target features includes:

[0018] Performing image segmentation on the target road image using a preset image segmentation method to obtain a segmented image, so that the road surface and road garbage in the segmented image are separated, the preset image segmentation method including threshold segmentation or region growing;

[0019] Extracting the edge features from the segmented image using a preset edge detection method;

[0020] Extracting the corner point features from the segmented image using a preset corner point detection method;

[0021] The texture features are extracted from the segmented image using a gray level co-occurrence matrix.

[0022] In some embodiments, the target road image is classified and identified based on the target features to obtain the type of road garbage, the amount of road garbage and the distribution information of road garbage, including:

[0023] According to the target features, a preset target detection method is used to perform target detection on the target road image to obtain a road garbage detection result, wherein the road garbage detection result includes a target road garbage object, road garbage location information or road garbage coverage information;

[0024] Using a support vector machine or a convolutional neural network to classify the target road garbage object to obtain a classification result;

[0025] Post-processing the classification result by using non-maximum suppression to obtain the road garbage type;

[0026] Calculating the amount of road garbage according to the target road garbage object and the road garbage location information;

[0027] The road garbage distribution information is calculated according to the target road garbage object, the road garbage location information, the road garbage coverage information and the road garbage type.

[0028] In some embodiments, the calculating the peak signal-to-noise ratio and the structural similarity index according to the target road image and the road image before cleaning includes:

[0029] Calculating a mean square error based on the target road image and the road image before cleaning;

[0030] Calculating the peak signal-to-noise ratio according to the mean square error and the maximum pixel value of the image;

[0031] Calculating image covariance based on the target road image and the road image before cleaning;

[0032] Calculating a mean value and a variance of a cleaned image according to the target road image;

[0033] Calculating the mean value and variance of the image before cleaning according to the image of the road before cleaning;

[0034] The structural similarity index is calculated based on the image covariance, the post-sweeping image mean, the post-sweeping image variance, the pre-sweeping image mean, the pre-sweeping image variance and a small constant.

[0035] In some embodiments, the cleaning quality evaluation is performed according to the road garbage type, the road garbage amount, the road garbage distribution information, the peak signal-to-noise ratio and the structural similarity index to obtain a cleaning quality evaluation result, including:

[0036] Calculating a first cleanliness level according to the road garbage type, the road garbage amount and the road garbage distribution information;

[0037] Calculating a second cleanliness level according to the peak signal-to-noise ratio and a preset signal-to-noise ratio threshold;

[0038] Calculating a third cleanliness level according to the structural similarity index and a preset similarity threshold;

[0039] The cleaning quality evaluation result is calculated based on the first cleanliness, the second cleanliness and the third cleanliness.

[0040] In some embodiments, the method further comprises:

[0041] According to the cleaning quality evaluation result, a cleaning method is selected from the cleaning methods to be selected as the target cleaning method, and the cleaning methods to be selected include mechanized cleaning, vacuum cleaning, high-pressure water gun washing, classified cleaning, periodic cleaning, key area cleaning, night cleaning, weather-adaptive cleaning, intelligent path planning cleaning, multi-mode cleaning, autonomous navigation cleaning, interactive cleaning, environmentally friendly cleaning and customized cleaning;

[0042] A cleaning strategy is generated based on the cleaning quality evaluation results, accumulated data and the target cleaning method. The accumulated data includes historical cleaning data, environmental data, traffic flow data, garbage type and distribution data, cleaning equipment status data, citizen feedback data or model prediction data.

[0043] The beneficial effects of the present invention are as follows:

[0044] The embodiment of the present invention first acquires an urban road image and performs preprocessing to obtain a target road image, then performs feature extraction processing on the target road image to obtain target features, and classifies and identifies the target road image based on the target features to obtain road garbage type, road garbage quantity and road garbage distribution information, then calculates a peak signal-to-noise ratio and a structural similarity index based on the target road image and the road image before cleaning, and finally performs cleaning quality evaluation based on the road garbage type, road garbage quantity, road garbage distribution information, peak signal-to-noise ratio and structural similarity index to obtain a cleaning quality evaluation result, so that cleaning quality evaluation can be achieved by analyzing road garbage data, thereby improving accuracy.

[0045] On the other hand, an embodiment of the present invention provides a device for evaluating the quality of urban road cleaning, comprising:

[0046] A moving module, the moving module is used to adjust the position of the device;

[0047] A cleaning module, which is used to clean urban roads after adjusting the position of the device;

[0048] An image acquisition module, wherein the image acquisition module is used to acquire images of the cleaned urban road to obtain an image of the urban road;

[0049] An image processing module, the image processing module is used to generate a cleaning quality evaluation result according to the urban road image;

[0050] A display module, the display module is used to display the cleaning quality evaluation result;

[0051] A control module, the control module is used to control the operating status of the cleaning module and the image acquisition module;

[0052] A power supply module, wherein the power supply module is used to provide working power.

[0053] In some embodiments, the cleaning module includes:

[0054] A cleaning mechanical submodule, wherein the cleaning mechanical submodule is used to clean road garbage;

[0055] A garbage collection submodule is used to collect the road garbage.

[0056] The beneficial effects of the present invention are as follows:

[0057] An urban road cleaning quality evaluation device provided by an embodiment of the present invention includes a moving module, a cleaning and sanitation module, an image acquisition module, an image processing module, a display module, a control module and a power module. The moving module can be used to adjust the device position, and the cleaning and sanitation module can be used to clean the urban road after adjusting the device position. The image acquisition module can be used to acquire images of the cleaned urban road to obtain an urban road image. The image processing module can be used to generate a cleaning quality evaluation result based on the urban road image. The display module can display the cleaning quality evaluation result. The control module can be used to control the operating states of the cleaning and sanitation module and the image acquisition module. The power module can be used to provide working power. Therefore, after the urban road is cleaned, the cleaning quality evaluation can be realized by analyzing the road garbage data, thereby improving the accuracy.

[0058] On the other hand, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0059] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0061] Figure 1 This is a flow chart of a method for evaluating urban road cleaning quality according to an embodiment of the present invention;

[0062] Figure 2 The present invention is a schematic diagram of the structure of a device for evaluating the quality of urban road cleaning according to an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0064] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0065] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0067] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0068] Gaussian filtering: It is a linear smoothing filter that is suitable for eliminating Gaussian noise and is widely used in the noise reduction process of image processing. In layman's terms, Gaussian filtering is the process of weighted averaging the entire image. The value of each pixel is obtained by weighted averaging the value of itself and other pixels in the neighborhood. The specific operation of Gaussian filtering is: use a template (or convolution, mask) to scan each pixel in the image, and use the weighted average grayscale value of the pixels in the neighborhood determined by the template to replace the value of the central pixel of the template.

[0069] Median filtering: It is a nonlinear smoothing technique that sets the grayscale value of each pixel to the median of the grayscale values ​​of all pixels within a certain neighborhood window of that point.

[0070] Histogram equalization: It is a method of adjusting contrast using the image histogram in the field of image processing. In this way, the brightness can be better distributed on the histogram. This can be used to enhance the local contrast without affecting the overall contrast. Histogram equalization achieves this function by effectively expanding the commonly used brightness.

[0071] In the related technology, with the rapid development of urbanization and the increase in urban population density, the cleaning and maintenance of urban roads have become particularly important. A clean and tidy urban environment plays a vital role in improving the quality of life of citizens and the overall image of the city. However, the current urban road cleaning and maintenance work mainly relies on manual work, which is not only inefficient and costly, but also difficult to ensure the cleaning quality, and cannot meet the high requirements of modern cities for a clean environment. In addition, the traditional cleaning quality evaluation method mainly relies on manual observation and simple measurement. This method is highly subjective, lacks scientificity and accuracy, has low accuracy, and cannot fully and objectively reflect the actual effect of cleaning and maintenance work. Therefore, how to improve the efficiency and quality of cleaning and maintenance, and how to scientifically and objectively evaluate the cleaning quality, have become problems that need to be solved in the current urban road cleaning work.

[0072] In view of this, the embodiments of the present invention combine mechanical automation, image processing and data analysis technologies to improve the efficiency and quality of cleaning and sanitation, while providing scientific and objective quality evaluation standards and methods. It can more effectively maintain the cleanliness of urban roads and create a cleaner and more comfortable living environment for citizens.

[0073] The embodiment of the present application provides a method for evaluating the quality of urban road cleaning, which relates to the technical field of urban environmental maintenance. The embodiment of the present application provides a method for evaluating the quality of urban road cleaning, which can be applied to a terminal, a server, or a software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application that implements a method for evaluating the quality of urban road cleaning, etc., but is not limited to the above forms.

[0074] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0075] The following is a detailed explanation of the embodiments of the present application in conjunction with the accompanying drawings:

[0076] Figure 1 is an optional flow chart of a method for evaluating urban road cleaning quality provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S106.

[0077] Step S101, obtaining a cleaned urban road image;

[0078] Step S102, preprocessing the urban road image to obtain a target road image;

[0079] Step S103, performing feature extraction processing on the target road image to obtain target features, where the target features include edge features, corner features, texture features, color features or shape features;

[0080] Step S104: classify and identify the target road image according to the target features to obtain the type of road garbage, the amount of road garbage and the distribution information of road garbage;

[0081] Step S105, calculating the peak signal-to-noise ratio and the structural similarity index according to the target road image and the road image before cleaning;

[0082] Step S106: Perform cleaning quality evaluation based on the road garbage type, road garbage quantity, road garbage distribution information, peak signal-to-noise ratio and structural similarity index to obtain a cleaning quality evaluation result.

[0083] Steps S101 to S106 shown in the embodiment of the present application realize cleaning quality evaluation and improve accuracy.

[0084] In step S101 of some embodiments, the image of the cleaned urban road can be obtained by an image acquisition module. The image of the cleaned urban road can also be obtained by other means, but is not limited thereto. For example, a high-definition camera can be used to collect images of the urban road surface at different time periods and under different lighting conditions to capture comprehensive information of the urban road surface.

[0085] In some embodiments, in step S102, preprocessing the urban road image to obtain the target road image may include but is not limited to the following steps:

[0086] Performing denoising processing on the urban road image by using Gaussian filtering or median filtering to obtain a first road image;

[0087] Performing contrast enhancement processing on the first road image by using histogram equalization to obtain a second road image;

[0088] performing size normalization processing on the second road image to obtain a third road image;

[0089] The third road image is normalized to obtain a target road image.

[0090] In some embodiments, after acquiring the urban road image, since the acquisition process may be disturbed by weather, noise, etc., an image quality assessment (IQA) algorithm can be introduced to filter out images of poor quality to ensure image quality before preprocessing. In the preprocessing, the urban road image can be first denoised using Gaussian filtering or median filtering to remove salt and pepper noise to obtain a first road image to reduce image noise and improve the accuracy of subsequent processing steps. Then, the first road image is contrast enhanced using histogram equalization to adjust the intensity distribution of pixels to obtain a second road image to enhance the image contrast and make the image contrast more uniform. The second road image is then size-standardized to scale the image to a uniform size to obtain a third road image to meet the input requirements of the subsequent algorithm. Finally, the third road image is normalized to scale the pixel values ​​to a range of 0-1 to obtain a target road image to improve the stability and convergence speed of the algorithm.

[0091] In some embodiments, in step S103, feature extraction processing is performed on the target road image to obtain target features, which may include but is not limited to the following steps:

[0092] Performing image segmentation on the target road image by using a preset image segmentation method to obtain a segmented image, so that the road surface and road garbage in the segmented image are separated. The preset image segmentation method includes threshold segmentation or region growing;

[0093] Extract edge features from the segmented image using a preset edge detection method;

[0094] Extract corner features from the segmented image using a preset corner detection method;

[0095] The gray-level co-occurrence matrix is ​​used to extract texture features from the segmented image.

[0096] In some embodiments, the target road image may be subjected to feature extraction processing to obtain target features, wherein the target features include edge features, corner features, texture features, color features or shape features, which are helpful for garbage identification and classification. The target road image may be first segmented using a preset image segmentation method to obtain a segmented image, so that the road surface and road garbage in the segmented image are separated. Among them, the preset image segmentation method may include threshold segmentation or region growing. Then, edge features are extracted from the segmented image using a preset edge detection method, and the edge features help to identify the contour of the object in the image. Among them, the preset edge detection method may include a Canny algorithm for identifying edges in an image. Then, corner features are extracted from the segmented image using a preset corner detection method, and the corner features are usually located at the corners or intersections of the edges of the object. Among them, the preset corner detection method may include a Harris corner detection algorithm for identifying key points in an image. Then, texture features are extracted from the segmented image using a gray level co-occurrence matrix (GLCM), which helps to distinguish different types of surfaces or materials.

[0097] In some embodiments, in step S104, the target road image is classified and identified according to the target features to obtain the type of road garbage, the amount of road garbage and the distribution information of road garbage, which may include but is not limited to the following steps:

[0098] According to the target features, a preset target detection method is used to perform target detection on the target road image to obtain a road garbage detection result, wherein the road garbage detection result includes the target road garbage object, road garbage location information or road garbage coverage information;

[0099] Use support vector machine or convolutional neural network to classify target road garbage objects and obtain classification results;

[0100] The classification results are post-processed using non-maximum suppression to obtain the road garbage type;

[0101] Calculate the amount of road garbage according to the target road garbage object and the road garbage location information;

[0102] The road garbage distribution information is calculated based on the target road garbage object, road garbage location information, road garbage coverage information and road garbage type.

[0103] In some embodiments, the target road image can be first detected by using a preset target detection method according to the target features to obtain a road garbage detection result. The road garbage detection result includes the target road garbage object, road garbage location information or road garbage coverage information. The preset target detection method can include YOLO (You Only Look Once) or Faster R-CNN in deep learning, which is used for target detection, identification and positioning of road garbage in the target road image. Then, the target road garbage object is classified using a support vector machine (SVM) or a convolutional neural network (CNN) to obtain a classification result. It can be understood that the support vector machine or the convolutional neural network can classify different types of garbage in the target road image and identify the type of road garbage. The classification result is then post-processed using non-maximum suppression to obtain the type of road garbage, so as to remove overlapping detection frames and improve the accuracy of recognition. Finally, according to the target road garbage object and the road garbage location information, the amount of road garbage is counted, and according to the target road garbage object, the road garbage location information, the road garbage coverage information and the road garbage type, the road garbage distribution information is calculated to obtain the distribution and amount of road garbage on the road surface.

[0104] In some embodiments, in step S105, calculating the peak signal-to-noise ratio and the structural similarity index based on the target road image and the road image before cleaning may include but is not limited to the following steps:

[0105] Calculate the mean square error based on the target road image and the road image before cleaning;

[0106] Calculate the peak signal-to-noise ratio based on the mean square error and the maximum pixel value of the image;

[0107] Calculate image covariance based on the target road image and the road image before cleaning;

[0108] According to the target road image, the mean value and variance of the cleaned image are calculated;

[0109] According to the road image before cleaning, the mean value and variance of the image before cleaning are calculated;

[0110] The structural similarity index is calculated based on the image covariance, the mean of the image after cleaning, the variance of the image after cleaning, the mean of the image before cleaning, the variance of the image before cleaning and a small constant.

[0111] In some embodiments, the peak signal-to-noise ratio (PSNR) and the structural similarity index (SSIM) are two image quality evaluation indicators that can help evaluate the visual effect and structural similarity of the image, and then evaluate the cleaning effect. The mean square error between the road image before cleaning and the target road image can be calculated based on the target road image and the road image before cleaning, and then the peak signal-to-noise ratio is calculated based on the mean square error and the maximum pixel value of the image. The calculation formula of the peak signal-to-noise ratio is: Where PSNR is the peak signal-to-noise ratio, MAX is the maximum pixel value of the image (for example, for an 8-bit image, MAX is usually 255), and MSE is the mean square error. Then, based on the target road image and the road image before cleaning, the image covariance σ is calculated: xy , according to the target road image, calculate the mean μ of the image after cleaning y And the image variance after cleaning And according to the road image before cleaning, calculate the mean value μ of the image before cleaning x and the image variance before cleaning Finally, the structural similarity index is calculated based on the image covariance, the image mean after cleaning, the image variance after cleaning, the image mean before cleaning, the image variance before cleaning and a small constant. The calculation formula of the structural similarity index is: Where SSIM is the structural similarity index, x is the road image before cleaning, y is the target road image, σ xy is the image covariance, μ y is the mean value of the image after cleaning, is the image variance after cleaning, μ x is the mean value of the image before cleaning, is the image variance before cleaning, c1 and c2 are both small constants. More importantly, the evaluation index of road cleanliness can include road sensory quality, visible garbage on the road, stain density or road dust. By combining these indicators, the quality of road cleaning and maintenance can be evaluated in percentage. The input and output formats are: input RAW format road image, output processed RGB image, garbage location and area information, and road cleanliness level.

[0112] In some embodiments, in step S106, a cleaning quality evaluation is performed based on the road garbage type, road garbage amount, road garbage distribution information, peak signal-to-noise ratio, and structural similarity index to obtain a cleaning quality evaluation result, which may include but is not limited to the following steps:

[0113] Calculating a first cleanliness level according to the road garbage type, the road garbage amount and the road garbage distribution information;

[0114] Calculating a second cleanliness level according to the peak signal-to-noise ratio and a preset signal-to-noise ratio threshold;

[0115] Calculating the third cleanliness according to the structural similarity index and a preset similarity threshold;

[0116] The cleaning quality evaluation result is calculated based on the first cleanliness, the second cleanliness and the third cleanliness.

[0117] In some embodiments, the first cleanliness can be calculated based on the type of road garbage, the amount of road garbage and the distribution information of road garbage. For example, if the amount of road garbage is still large, the road garbage is scattered and the types of road garbage are many, low cleanliness can be used as the first cleanliness. Then, according to the peak signal-to-noise ratio and the preset signal-to-noise ratio threshold, the second cleanliness is calculated. In the evaluation of cleaning quality, the peak signal-to-noise ratio PSNR can be used to quantify the quality difference of the image before and after cleaning. A higher PSNR value indicates that the image quality is better. When the peak signal-to-noise ratio is greater than the preset signal-to-noise ratio threshold, it indicates that the cleaning effect is good, and high cleanliness can be used as the second cleanliness. Then, according to the structural similarity index and the preset similarity threshold, the third cleanliness is calculated. The structural similarity index SSIM can measure the similarity of the brightness, contrast and structure of the image. In the evaluation of cleaning quality, SSIM can be used to measure the visual quality changes of the road surface image before and after cleaning. The range of SSIM values ​​is -1 to 1. The closer the value is to 1, the more similar the two images are. When the structural similarity index is greater than the preset similarity threshold, it indicates that the cleaning effect is good, and high cleanliness can be used as the third cleanliness. Finally, according to the first cleanliness, the second cleanliness and the third cleanliness, the cleaning quality evaluation result is calculated. The cleaning quality evaluation result can be obtained by combining the first cleanliness, the second cleanliness and the third cleanliness. For example, if the first cleanliness, the second cleanliness and the third cleanliness are all highly clean, the highly cleanliness can be used as the cleaning quality evaluation result; if the first cleanliness, the second cleanliness and the third cleanliness are all low cleanliness, the low cleanliness can be used as the cleaning quality evaluation result. More, the image difference comparison technology can be used to quantify the target road image and the road image before cleaning to obtain a quantitative cleaning effect to evaluate the changes in the road surface before and after cleaning. By collecting PSNR and SSIM data for a long time, the trend and pattern of the cleaning effect can be analyzed, and then the cleaning strategy can be optimized. The cleaning effect can be quantitatively evaluated by calculating the PSNR and SSIM values ​​of the images before and after cleaning. If the PSNR and SSIM values ​​after cleaning are significantly higher than those before cleaning, it means that the cleaning effect is good. The real-time measurement results of PSNR and SSIM are fed back to the control platform so that the cleaning parameters can be automatically adjusted according to the current cleaning effect.

[0118] In some embodiments, the method further comprises:

[0119] According to the cleaning quality evaluation results, a cleaning method is selected from the cleaning methods to be selected as the target cleaning method. The cleaning methods to be selected include mechanized cleaning, vacuum cleaning, high-pressure water gun washing, classified cleaning, periodic cleaning, key area cleaning, night cleaning, weather-adaptive cleaning, intelligent path planning cleaning, multi-mode cleaning, autonomous navigation cleaning, interactive cleaning, environmentally friendly cleaning and customized cleaning;

[0120] A cleaning strategy is generated based on the cleaning quality evaluation results, accumulated data and target cleaning methods. The accumulated data include historical cleaning data, environmental data, traffic flow data, garbage type and distribution data, cleaning equipment status data, citizen feedback data or model prediction data.

[0121] In some embodiments, with the accumulation and analysis of data, intelligent decision-making can be achieved. For example, the cleaning effect is predicted based on historical data and current environmental data, and the cleaning strategy is automatically adjusted to achieve the best cleaning effect; or the cleaning path and method are continuously optimized through machine learning algorithms to improve cleaning efficiency and quality. After obtaining the cleaning quality evaluation results, you can first select a cleaning method from the cleaning methods to be selected as the target cleaning method based on the cleaning quality evaluation results, and adjust the cleaning method according to different environmental conditions, garbage types and cleaning targets. Among them, the cleaning methods to be selected may include mechanized cleaning, vacuum cleaning, high-pressure water gun flushing, classified cleaning, periodic cleaning, key area cleaning, night cleaning, weather adaptability cleaning, intelligent path planning cleaning, multi-mode cleaning, autonomous navigation cleaning, interactive cleaning, environmentally friendly cleaning and customized cleaning. Mechanized cleaning is used to characterize the use of electric sweepers or automatic cleaning robots for large-scale cleaning, which is suitable for wide roads and squares; vacuum cleaning is used to characterize the use of vacuum equipment to absorb dust and fine particles on the road surface, which is suitable for areas that require fine cleaning; high-pressure water gun flushing is used to characterize the use of high-pressure water guns for flushing for stubborn stains or oil stains, which can more effectively remove dirt; classified cleaning is used to characterize the use of different cleaning tools or methods according to the type of garbage (such as organic matter, plastic, metal, etc.) to achieve classified collection of garbage; periodic cleaning is used to characterize the arrangement of regular cleaning plans based on the periodic laws of garbage generation and accumulation to keep the road surface continuously clean; key area cleaning is used to characterize the identification and marking of areas where garbage is concentrated, and these areas are cleaned as the focus to improve cleaning efficiency; night cleaning is used to characterize cleaning at night when traffic volume is small, reducing impact on traffic and citizens' lives The impact of weather-adaptive cleaning is used to characterize the adjustment of cleaning methods according to weather conditions (such as rainy days and snowy days). For example, more drainage work may be required on rainy days. Intelligent path planning cleaning is used to characterize the use of algorithms to optimize cleaning paths, avoid repeated cleaning, and improve cleaning efficiency. Multi-mode cleaning is used to characterize the combination of different cleaning tools and technologies, such as brooms, brushes, vacuum cleaners, and water guns, to adapt to different cleaning needs. Autonomous navigation cleaning is used to characterize the use of remote control or autonomous navigation technology to allow cleaning equipment to work with less or no human intervention. Interactive cleaning is used to characterize the ability of equipment to respond to real-time feedback, such as citizens' instant reports or warnings from monitoring systems, to perform interactive cleaning. Environmentally friendly cleaning is used to characterize the use of environmentally friendly materials and energy, such as solar-powered cleaning equipment, to reduce environmental impact. Customized cleaning is used to characterize the customization of cleaning strategies and equipment configurations according to the specific needs of different cities or regions. Through these diverse cleaning methods, cleaning and maintenance equipment can flexibly respond to various environments and conditions to achieve efficient and intelligent cleaning and maintenance work.

[0122] In some embodiments, accumulated data and advanced algorithms can be used to automatically adapt to different environments and conditions and optimize cleaning strategies. After selecting a target cleaning method, a cleaning strategy can be generated based on the cleaning quality evaluation results, accumulated data, and target cleaning method. The accumulated data may include historical cleaning data, environmental data, traffic flow data, garbage type and distribution data, cleaning equipment status data, citizen feedback data, or model prediction data. Historical cleaning data is used to characterize the collection of past cleaning effect data, including images before and after cleaning, cleaning efficiency, cleaning coverage area, etc., to analyze which areas have more garbage and which time periods have the highest garbage cleaning efficiency; environmental data is used to characterize the use of temperature and humidity sensors, wind speed sensors, etc. to collect current environmental data, and adjust the cleaning intensity and frequency according to weather conditions (such as wet roads after rain, garbage is more likely to adhere); traffic flow data is used to characterize the acquisition of road traffic flow information through the traffic monitoring system, and cleaning during periods of low traffic to reduce the impact on traffic; garbage type and distribution data is used to characterize the use of image recognition technology to identify different types of garbage and their distribution, and adjust cleaning tools according to garbage types, such as using different brooms or vacuum cleaners; cleaning equipment status data is used to characterize the monitoring of the operating status of cleaning equipment, such as battery power, degree of wear, etc., and arrange cleaning tasks when the power is sufficient or the equipment is in good condition. To ensure the cleaning effect; citizen feedback data is used to represent the collection of citizen feedback and complaints on cleaning quality, and to adjust the key cleaning areas or improve the cleaning methods based on the feedback; model prediction data is used to represent the use of machine learning algorithms to establish prediction models, predict the cleaning effects under different conditions, and adjust the cleaning plan based on the prediction results, such as cleaning in advance when the amount of garbage is predicted to increase; adaptive learning is used to represent that through continuous learning and optimization, the cleaning device can identify the pattern of poor cleaning effect and automatically adjust the strategy. For example, if a certain area repeatedly has incomplete garbage cleaning, the system can increase the number of cleaning times in the area or change the cleaning path; real-time adjustment is used to represent the real-time adjustment of cleaning routes and plans based on real-time collected data, such as the increase in garbage caused by emergencies; optimization algorithms are used to represent the application of optimization algorithms such as genetic algorithms and simulated annealing algorithms to find the best cleaning path and strategy, reduce cleaning time and increase coverage. By integrating these data and intelligent algorithms, the cleaning and sanitation device can achieve more efficient and intelligent cleaning work, constantly adapt to environmental changes, and improve cleaning quality and efficiency. Furthermore, an evaluation report can be generated based on the cleaning quality evaluation results and cleaning strategies, integrating the detected garbage information and cleanliness evaluation results into an evaluation report, which includes garbage statistics, cleanliness levels, and possible improvement suggestions. At the same time, the measurement results of PSNR and SSIM can be integrated into the cleaning quality evaluation report to provide operators with an intuitive cleaning effect evaluation.The evaluation results are displayed to operators in real time through the user interface, allowing them to intuitively understand the cleaning effect and perform manual intervention as needed. The cleaning quality evaluation report will not only be displayed to the operator, but will also be transmitted to the control platform for storage and analysis, and the evaluation report will be fed back to the control platform. The cleaning strategy will be adjusted according to the report results, the cleaning path and method will be optimized, and the collected images and evaluation results will be stored in the database for historical data analysis and long-term optimization of the cleaning strategy. By mining and analyzing historical data, problems and deficiencies in the cleaning process can be discovered, thereby guiding operators to improve cleaning strategies or adjust the working parameters of the device. Adaptive learning can also be performed to continuously optimize garbage detection algorithms and cleanliness evaluation standards based on historical data and real-time feedback.

[0123] In some embodiments, the overall process of generating a cleaning strategy can first obtain the image of the urban road after cleaning, and perform preprocessing to obtain the target road image, then perform feature extraction processing to obtain the target feature, and classify and identify the target road image to obtain road garbage information, and then calculate the peak signal-to-noise ratio and structural similarity index based on the target road image and the road image before cleaning, and perform cleaning quality evaluation to obtain a cleaning quality evaluation result. Finally, after selecting the target cleaning method, a cleaning strategy is generated based on the cleaning quality evaluation result, accumulated data and target cleaning method.

[0124] The beneficial effects of implementing the embodiments of the present invention include: the embodiments of the present invention first acquire a city road image and perform preprocessing to obtain a target road image, then perform feature extraction processing on the target road image to obtain target features, and classify and identify the target road image based on the target features to obtain road garbage type, road garbage quantity and road garbage distribution information, and then calculate the peak signal-to-noise ratio and structural similarity index based on the target road image and the road image before cleaning, and finally perform cleaning quality evaluation based on the road garbage type, road garbage quantity, road garbage distribution information, peak signal-to-noise ratio and structural similarity index to obtain a cleaning quality evaluation result, so that cleaning quality evaluation can be achieved by analyzing road garbage data, thereby improving accuracy.

[0125] like Figure 2 As shown, the embodiment of the present invention also provides a device for evaluating the quality of urban road cleaning, comprising:

[0126] A moving module 201, which is used to adjust the position of the device;

[0127] A cleaning module 202, which is used to clean the city roads after adjusting the position of the device;

[0128] An image acquisition module 203 is used to acquire images of the cleaned urban roads to obtain urban road images;

[0129] An image processing module 204, which is used to generate a cleaning quality evaluation result based on the urban road image;

[0130] Display module 205, the display module is used to display the cleaning quality evaluation result;

[0131] A control module 206, which is used to control the operation status of the cleaning module and the image acquisition module;

[0132] The power module 207 is used to provide working power.

[0133] In some embodiments, a device for evaluating the cleaning quality of urban roads provided in this embodiment includes a mobile module, a cleaning and sanitation module, an image acquisition module, an image processing module, a display module, a control module and a power module. The position of the device can be adjusted by the mobile module, wherein the mobile module, as the bearing basis of the entire device, can be an electric vehicle or other form of mobile device, responsible for the movement and positioning of the device, and can be driven by electricity or fuel to provide the necessary mobility so that the device can cover a wide range of cleaning areas. The urban roads can be cleaned by the cleaning and sanitation module after adjusting the position of the device. The image acquisition module can be used to collect images of the cleaned urban roads to obtain urban road images. The images of the cleaned urban roads can be captured by a high-definition camera to ensure high resolution and high definition of the images, so that various details on the road can be accurately identified. Other sensors, such as temperature and humidity sensors, wind speed sensors, etc., can also be used to collect environmental data to assist in the evaluation of cleaning quality. The cleaning quality evaluation results can be generated by the image processing module based on the urban road images. Exemplarily, the collected images can be deeply analyzed, and the cleaning quality can be quantitatively evaluated by identifying the quantity, type and distribution of garbage residues. The cleaning quality evaluation results can be displayed by the display module to provide intuitive feedback for the cleaning staff. The operating status of the cleaning and maintenance module and the image acquisition module can be controlled by the control module. It is understandable that the control module is responsible for the control and coordination of the entire device, including the start and stop of the cleaning and maintenance module, the sending of the acquisition instruction of the image acquisition module, and the collaborative work between the modules, and can monitor and control various parameters in the cleaning process. The power module can be used to provide working power for each module of the device of this embodiment, which can include a battery pack, a charging interface and a power management unit to ensure that the device can continue to operate stably during continuous work. The solar panel can be installed on the roof to charge the battery during the day to provide clean energy. The lithium battery can be used as an energy storage device to ensure that the device can continue to work even without sunlight. In addition, the collected urban road images and environmental data can be transmitted to the image processing module through the signal transmission module. Signal transmission can be wired or wireless, such as using Wi-Fi, Bluetooth or 4G / 5G network for data transmission. In order to ensure real-time and stability, the transmission protocol can be optimized to reduce transmission delays and data loss. The lifting device can be used to switch the cleaning device between different working modes, such as switching from driving mode to cleaning mode, which improves the maneuverability and versatility of the machine and can adapt to different cleaning tasks. In the operation process of the device, environmental data can be collected and analyzed first, and then the cleaning module or the sweeping and suction road sweeper can be controlled to start working according to the analysis results and the preset cleaning plan. During the cleaning process, the lifting device can adjust the working state of the cleaning device as needed.After the cleaning is completed, the image acquisition module captures the road surface image for analysis by the quality evaluation module. In some embodiments, the device design should consider easy maintenance and cleaning to ensure long-term stable operation. And it can continuously optimize the cleaning strategy and equipment performance by collecting operation data and user feedback. The device of this embodiment not only improves the cleaning efficiency and quality, but also provides strong support for the continuous improvement of the urban environment through integrated intelligent systems and environmental protection technologies.

[0134] In some embodiments, the cleaning module includes:

[0135] The cleaning machine submodule is used to clean road garbage;

[0136] The garbage collection submodule is used to collect road garbage.

[0137] In some embodiments, the cleaning and sanitation module includes a cleaning mechanical submodule and a garbage collection submodule. Among them, the cleaning mechanical submodule may include a cleaning mechanism for cleaning road garbage, and an electric sweeper or a vacuum cleaner may be used for efficient cleaning. A dust suction system may be used, and the core component is an air suction fan, which absorbs dust and garbage on the road surface by generating a strong suction force. The suction port is directly in contact with the road surface, and is designed to maximize the suction area and efficiency. A filter bag is used to filter solid particles in the inhaled air to keep the air clean. A hose is used to connect each component, allowing the position of the suction port to be flexibly adjusted. A wind system may also be used, driven by a secondary engine, to work in conjunction with a power transmission system and a fan to form a negative pressure dust suction. A hydraulic system may also be used, with a dual pump design to improve the system's working efficiency and the utilization rate of hydraulic oil. The garbage collection submodule may include a garbage recycling mechanism for collecting road garbage, which may automatically follow the cleaning mechanical submodule to promptly recycle the cleaned garbage into a container, thereby realizing the integrated operation of cleaning and recycling.

[0138] In some embodiments, the workflow of road cleaning and cleaning quality evaluation can start the cleaning and sanitation module first, so that the cleaning mechanism can clean the road thoroughly, and the garbage recycling mechanism can recycle the cleaned garbage in time. Among them, the cleaning mechanism uses efficient cleaning technology to clean the urban roads comprehensively. The garbage recycling mechanism follows closely, automatically collects and processes the garbage generated during the cleaning process, ensuring that the road is clean and avoiding secondary pollution. Then the quality evaluation module is started, the image acquisition device captures the image of the road after cleaning, and the image processing unit performs high-precision analysis on the image to evaluate the cleaning quality. Among them, the image acquisition module captures high-definition images of the road after cleaning to ensure that every detail can be accurately recorded. The image processing module then performs in-depth processing and analysis on the captured images, and accurately calculates the amount, type and distribution of garbage remaining on the road through advanced image recognition technology. Finally, the evaluation results are displayed to the cleaning staff in real time through the result display screen, providing them with immediate feedback and guidance. Not only does it provide operators with immediate feedback on the cleaning effect, but it also provides them with targeted improvement suggestions, thereby ensuring the continuous optimization and improvement of urban road cleaning work. This embodiment can not only significantly improve the efficiency and quality of urban road cleaning, but also provide scientific and objective data support for the continuous improvement of the urban environment. At the same time, the flexibility and scalability of this embodiment also enable it to adapt to the cleaning needs of different cities and different roads, and it has broad application prospects and market potential. This embodiment can provide scientific and objective cleaning quality evaluation standards, providing strong support for the continuous improvement of the urban environment. Furthermore, this embodiment can be connected to a remote server through a network to achieve remote monitoring and management of data. Through the remote monitoring function, work data can be uploaded to the server in real time, so that managers can monitor the working status of the device anytime and anywhere, improving the convenience and efficiency of management.

[0139] In some embodiments, during the road cleaning process, tools can be prepared first, and all cleaning equipment (such as electric sweepers, vacuum cleaners, sweeping and suction road sweepers, etc.) can be checked and ensured to work properly, and necessary consumables such as filter bags, batteries, detergents, etc. can be prepared. Then, the areas in charge are cleaned in a cycle, and cleaning plans and routes are formulated according to the divided areas and responsibilities, and the cycle cleaning tasks are performed to ensure that all designated areas are covered. And assist in the removal of domestic garbage, collect garbage generated during the cleaning process, and transport it to the designated garbage disposal station to ensure the classification and compliance of garbage. At the same time, public equipment can be cleaned, and equipment in public areas such as benches, trash cans, traffic lights, etc. can be cleaned and maintained to ensure the cleanliness and normal function of public facilities. The status of equipment and facilities can be checked regularly, problems can be reported in time, maintenance needs can be recorded, and repair work can be arranged in coordination with the maintenance team. After the cleaning is completed, the image acquisition device can be used to evaluate the cleaned road surface, and the collected images can be analyzed using image processing algorithms to generate a quality evaluation report. During the evaluation process, data recording and analysis can be performed to record data during the cleaning process, including cleaning time, cleaning area, garbage volume, etc., and analyze the data to optimize the cleaning strategy and improve efficiency. Feedback and adjustment can be performed. According to the quality evaluation report and data records, the cleaning plan and method can be adjusted, feedback from operators can be collected, and the work process can be continuously improved. In addition, training and guidance can be provided to cleaning personnel, and operators can be trained regularly to improve their operating skills and safety awareness, provide guidance and support, and ensure that all personnel understand the work process and operating standards. Supervision and inspection can be performed, and on-site supervision and inspection can be performed regularly to ensure the execution of the work process and quality standards, and corrective measures can be taken for non-compliance with the standards. In addition, report and document management can be performed, work reports can be prepared, cleaning activities, problem solving and improvement measures can be recorded, and related documents can be managed to ensure the accuracy and traceability of information. Continuous improvement can be performed. According to feedback, data analysis and supervision results, the work process can be continuously improved, and new technologies and methods can be adopted to improve the efficiency and quality of cleaning and sanitation work. This embodiment can achieve efficient and systematic cleaning and sanitation operations while ensuring the quality and continuous improvement of the operations.

[0140] The beneficial effects of implementing the embodiments of the present invention include: the embodiment of the present invention provides a device for evaluating the cleaning quality of urban roads, including a mobile module, a cleaning and sanitation module, an image acquisition module, an image processing module, a display module, a control module and a power module. The device position can be adjusted by the mobile module, the cleaning and sanitation module can be used to clean the urban road after adjusting the device position, the image acquisition module can be used to acquire images of the cleaned urban road to obtain an image of the urban road, the image processing module can be used to generate a cleaning quality evaluation result based on the urban road image, the display module can be used to display the cleaning quality evaluation result, the control module can be used to control the operating status of the cleaning and sanitation module and the image acquisition module, the power module can be used to provide working power, so that after the urban road is cleaned, the cleaning quality evaluation can be realized by analyzing the road garbage data, thereby improving the accuracy.

[0141] The embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, Figure 1 The method shown.

[0142] The contents of the above method embodiments are all applicable to the present storage medium embodiments. The functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0143] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for evaluating the quality of urban road cleaning, characterized in that: The following steps are involved: Acquire the cleaned urban road image; Preprocessing the urban road image to obtain a target road image; Performing feature extraction processing on the target road image to obtain target features, wherein the target features include edge features, corner features, texture features, color features or shape features; According to the target features, the target road image is classified and identified to obtain road garbage type, road garbage quantity and road garbage distribution information; Calculating a peak signal-to-noise ratio and a structural similarity index according to the target road image and the road image before cleaning; A cleaning quality evaluation is performed based on the road garbage type, the road garbage amount, the road garbage distribution information, the peak signal-to-noise ratio and the structural similarity index to obtain a cleaning quality evaluation result.

2. The method according to claim 1, characterized in that The preprocessing of the urban road image to obtain a target road image includes: Performing denoising processing on the urban road image by using Gaussian filtering or median filtering to obtain a first road image; Performing contrast enhancement processing on the first road image by using histogram equalization to obtain a second road image; performing size normalization processing on the second road image to obtain a third road image; The third road image is normalized to obtain the target road image.

3. The method according to claim 1, characterized in that The step of performing feature extraction processing on the target road image to obtain target features includes: Performing image segmentation on the target road image using a preset image segmentation method to obtain a segmented image, so that the road surface and road garbage in the segmented image are separated, the preset image segmentation method including threshold segmentation or region growing; Extracting the edge features from the segmented image using a preset edge detection method; Extracting the corner point features from the segmented image using a preset corner point detection method; The texture features are extracted from the segmented image using a gray level co-occurrence matrix.

4. The method according to claim 1, characterized in that The target road image is classified and identified according to the target features to obtain road garbage type, road garbage quantity and road garbage distribution information, including: According to the target features, a preset target detection method is used to perform target detection on the target road image to obtain a road garbage detection result, wherein the road garbage detection result includes a target road garbage object, road garbage location information or road garbage coverage information; Using a support vector machine or a convolutional neural network to classify the target road garbage object to obtain a classification result; Post-processing the classification result by using non-maximum suppression to obtain the road garbage type; Calculating the amount of road garbage according to the target road garbage object and the road garbage location information; The road garbage distribution information is calculated according to the target road garbage object, the road garbage location information, the road garbage coverage information and the road garbage type.

5. The method according to claim 1, characterized in that The step of calculating the peak signal-to-noise ratio and the structural similarity index according to the target road image and the road image before cleaning includes: Calculating a mean square error based on the target road image and the road image before cleaning; Calculating the peak signal-to-noise ratio according to the mean square error and the maximum pixel value of the image; Calculating image covariance based on the target road image and the road image before cleaning; Calculating a mean value and a variance of a cleaned image according to the target road image; Calculating the mean value and variance of the image before cleaning according to the image of the road before cleaning; The structural similarity index is calculated based on the image covariance, the post-sweeping image mean, the post-sweeping image variance, the pre-sweeping image mean, the pre-sweeping image variance and a small constant.

6. The method according to claim 1, characterized in that The cleaning quality evaluation is performed according to the road garbage type, the road garbage amount, the road garbage distribution information, the peak signal-to-noise ratio and the structural similarity index to obtain a cleaning quality evaluation result, including: Calculating a first cleanliness level according to the road garbage type, the road garbage amount and the road garbage distribution information; Calculating a second cleanliness level according to the peak signal-to-noise ratio and a preset signal-to-noise ratio threshold; Calculating a third cleanliness level according to the structural similarity index and a preset similarity threshold; The cleaning quality evaluation result is calculated based on the first cleanliness, the second cleanliness and the third cleanliness.

7. The method according to claim 1, characterized in that The method further comprises: According to the cleaning quality evaluation result, a cleaning method is selected from the cleaning methods to be selected as the target cleaning method, and the cleaning methods to be selected include mechanized cleaning, vacuum cleaning, high-pressure water gun washing, classified cleaning, periodic cleaning, key area cleaning, night cleaning, weather-adaptive cleaning, intelligent path planning cleaning, multi-mode cleaning, autonomous navigation cleaning, interactive cleaning, environmentally friendly cleaning and customized cleaning; A cleaning strategy is generated based on the cleaning quality evaluation results, accumulated data and the target cleaning method. The accumulated data includes historical cleaning data, environmental data, traffic flow data, garbage type and distribution data, cleaning equipment status data, citizen feedback data or model prediction data.

8. A device for evaluating the quality of urban road cleaning, characterized in that: include: A moving module, the moving module is used to adjust the position of the device; A cleaning module, which is used to clean urban roads after adjusting the position of the device; An image acquisition module, the image acquisition module is used to acquire images of the cleaned urban road to obtain an image of the urban road; An image processing module, the image processing module is used to generate a cleaning quality evaluation result according to the urban road image; A display module, the display module is used to display the cleaning quality evaluation result; A control module, the control module is used to control the operating status of the cleaning module and the image acquisition module; A power supply module, wherein the power supply module is used to provide working power.

9. The device according to claim 8, characterized in that The cleaning module comprises: A cleaning mechanical submodule, wherein the cleaning mechanical submodule is used to clean road garbage; A garbage collection submodule is used to collect the road garbage.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images