Management method and system for urban road occupation treatment

Abstract

The invention relates to the technical field of data management, in particular to a management method and system for urban road occupation management, and the method comprises the steps: obtaining a grid street road monitoring image based on an image obtained by a roadside camera; analyzing the street monitoring image to obtain the position and range of a street monitoring blind area; obtaining entrance and exit area information based on the street monitoring blind area range, and establishing a monitoring blind area associated image data set based on an image obtained by monitoring the entrance and exit area information; on the basis of the associated image data set, features of the road-occupying operation object are obtained, and identity marking is carried out; recording a time interval of entering and exiting the monitoring blind area of the lane-occupying operation object based on the identity mark, and judging a lane-occupying operation individual; and comparing and excluding information of shops where the abnormal staying individuals and the monitoring blind areas are located to obtain flowing lane occupation data. Therefore, the operation condition of the road shielding part can be analyzed in combination with the flow data, and the management efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data governance, and particularly to a management method and system for urban road occupation governance. Background Art

[0002] Urban road occupation governance refers to measures for regulating and managing the behavior of unauthorized occupation of roads, sidewalks, and other public spaces in cities. With the acceleration of urbanization, the increase in the number of vehicles, and the prosperity of commercial activities, the phenomenon of road occupation has become increasingly common, which not only affects traffic flow but also poses potential safety hazards to pedestrians.

[0003] Existing road occupation governance generally uses cameras to obtain street images and image recognition technology to identify the road occupation situation on the street to notify relevant personnel for handling. However, there are generally trees beside the street, and these trees may block the cameras, resulting in the inability to observe the road occupation situation in the blocked area and reducing the management efficiency. Summary of the Invention

[0004] The purpose of the present invention is to provide a management method and system for urban road occupation governance, aiming to combine mobile data analysis to analyze the business situation in the blocked part of the road, thereby improving management efficiency.

[0005] To achieve the above object, in a first aspect, the present invention provides a management method for urban road occupation governance, including obtaining a grid-based street monitoring image based on the image obtained by a roadside camera;

[0006] Analyzing the street monitoring image to obtain the position and range of the monitoring blind area of the street;

[0007] Obtaining import and export area information based on the range of the street monitoring blind area, and establishing a monitoring blind area associated image data set based on the images monitored based on the import and export area information;

[0008] Obtaining the characteristics of the objects operating without permission to occupy the road based on the associated image data set, and performing identity marking;

[0009] Recording the time interval for the objects operating without permission to occupy the road to enter and exit the monitoring blind area based on the identity marking, and judging the individuals operating without permission to occupy the road;

[0010] Comparing and excluding the individuals with abnormal stays and the store information where the monitoring blind area is located to obtain mobile road occupation data.

[0011] Wherein, after comparing and excluding the individuals with abnormal stays and the store information where the monitoring blind area is located to obtain mobile road occupation data, the method further includes extracting the image features in the mobile road occupation data, and matching and comparing the image features in the entire monitoring network to predict the mobile road occupation situation of the individuals.

[0012] Among them, the specific steps of obtaining the grid-based street road monitoring image from the images acquired by roadside cameras include:

[0013] Set the image acquisition frequency to capture the image information on the street;

[0014] Use a filtering algorithm to remove the noise in the image information;

[0015] Crop the image to focus on the street area;

[0016] Divide the street area into several grid cells of the same size;

[0017] Use image stitching technology to synthesize the grid cells generated from the monitoring images in all areas into a complete picture to obtain the street road monitoring image information.

[0018] Among them, the specific steps of analyzing the street road monitoring image to obtain the location and scope of the street monitoring blind area include:

[0019] Obtain the street road monitoring image information;

[0020] Based on the occlusion object recognition method, identify the stationary obstacles in the road monitoring image information to obtain the obstacle information;

[0021] Compare the road area marked on the street road monitoring image information with the position of the obstacle information to obtain the location and scope of the street monitoring blind area.

[0022] Among them, the specific steps of identifying the stationary obstacles in the road monitoring image information based on the occlusion object recognition method to obtain the obstacle information include:

[0023] Collect the characteristics of common obstacles in the target area;

[0024] Based on the characteristics of common obstacles, perform feature matching in the road monitoring image information to obtain the target obstacles;

[0025] Extract the obstacle information based on the target obstacles, and the obstacle information includes the obstacle type and location.

[0026] Among them, the specific steps of obtaining the import and export area information based on the scope of the street monitoring blind area and establishing a monitoring blind area associated image dataset based on the images monitored by the import and export area information include:

[0027] Obtain the map data where the blind area is located;

[0028] Mark the road key points on the map data;

[0029] Obtain the corresponding camera numbers based on the road key points;

[0030] Based on the camera numbers, the images obtained by the relevant cameras are combined to form a monitoring blind area associated image dataset.

[0031] Among them, the specific steps for obtaining the characteristics of the objects occupying the road and performing identity marking based on the associated image dataset include:

[0032] Use the Canny algorithm to detect the contours of each object in the associated image dataset;

[0033] Apply the object detection algorithm to identify all object types in the object contour and frame their positions;

[0034] Perform identity marking on the objects.

[0035] Among them, the specific steps for recording the time interval of the objects occupying the road entering and leaving the monitoring blind area based on the identity marking and judging the individuals occupying the road include:

[0036] Use the motion detection algorithm to capture the events of each identity-marked object entering and leaving the monitoring blind area;

[0037] Record the entry timestamp for each detected entry event and the departure timestamp for each detected departure event;

[0038] For each entry and exit event, calculate the time interval that the object stays in the monitoring blind area;

[0039] Use the Z-score algorithm to identify the business objects with time intervals exceeding the normal range and obtain the individuals with abnormal stays.

[0040] Among them, the specific steps for comparing and excluding the individuals with abnormal stays and the store information where the monitoring blind area is located to obtain the mobile occupation data of the road include:

[0041] Obtain the store data in the monitoring area;

[0042] Based on the store data, confirm whether the individuals with abnormal stays belong to the store. If not, they are the mobile occupation data of the road.

[0043] In a second aspect, the present invention also provides a management system for urban road occupation governance, including a monitoring image acquisition module, a monitoring blind area detection module, an import and export area acquisition module, an identity marking module, an occupation judgment module, and an exclusion module;

[0044] The monitoring image acquisition module is used to obtain a grid-based street road monitoring image based on the images obtained by the roadside cameras;

[0045] The monitoring blind area detection module is used to analyze the street road monitoring image to obtain the position and range of the street monitoring blind area;

[0046] The import and export area acquisition module is used to acquire import and export area information based on the range of street monitoring blind spots, and establish a monitoring blind spot associated image dataset based on the images obtained by monitoring the import and export area information;

[0047] The identity marking module is used to acquire the characteristics of the objects occupying the road based on the associated image dataset and perform identity marking;

[0048] The road occupation judgment module is used to judge the individual occupying the road based on the time interval of the object occupying the road entering and leaving the monitoring blind spot recorded by the identity marking;

[0049] The exclusion module is used to compare and exclude the information of the individuals staying abnormally and the stores where the monitoring blind spots are located to obtain the mobile road occupation data.

[0050] A management method and system for urban road occupation governance according to the present invention captures the situations on the street in real time through cameras installed by the roadside. These cameras can cover a wide area, and the acquired images are processed into a grid to form a series of street monitoring images. Next, the system will perform in-depth analysis on these street monitoring images to determine where there are monitoring blind spots, that is, those areas that are difficult to directly observe due to limited viewing angles or object occlusion. To better understand and manage these monitoring blind spots, the system will further collect information about the import and export areas. By combining the image data at the import and export, an image dataset related to the blind spots is established. Using the above-mentioned associated image dataset, the system can identify the characteristics of the objects occupying the road, such as the shape, color, size, etc. of the stalls, and perform identity marking on them. This marking process is automated, which allows the system to quickly and accurately identify specific individuals occupying the road, even if they move to different positions. Subsequently, record the time intervals of these marked objects occupying the road entering and leaving the monitoring blind spot to judge which are individual operators. This helps to distinguish the difference between the behavior of long-term occupying public resources and temporary stays, so as to more accurately locate the targets that need to be intervened. For those individuals staying abnormally, the system will compare and exclude their information with the information of the surrounding stores. This process is to confirm whether there is a legal business activity or to confirm whether these behaviors belong to illegal road occupation. In this way, the operation situation of the blocked part of the road can be analyzed by combining the mobile data, thereby improving the management efficiency. The system can screen out the real mobile road occupation data, provide support for urban management decision-making, and at the same time protect the interests of legitimate merchants from being violated. Description of the Drawings

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0052] Figure 1 It is a flowchart of a management method for urban road occupation governance of the present invention.

[0053] Figure 2 It is a flowchart of obtaining a grid-based street road monitoring image from the images acquired by roadside cameras of the present invention.

[0054] Figure 3 It is a flowchart of analyzing the street road monitoring image to obtain the location and scope of the street monitoring blind area of the present invention.

[0055] Figure 4 It is a flowchart of identifying stationary obstacles in the road monitoring image information based on the occlusion recognition method to obtain obstacle information of the present invention.

[0056] Figure 5 It is a flowchart of obtaining import and export area information based on the street monitoring blind area scope and establishing a monitoring blind area associated image dataset based on the images monitored by the import and export area information of the present invention.

[0057] Figure 6 It is a flowchart of obtaining the characteristics of the objects occupying the road for business based on the associated image dataset and performing identity marking of the present invention.

[0058] Figure 7 It is a flowchart of recording the time interval for the objects occupying the road for business to enter and exit the monitoring blind area based on the identity marking and judging the individuals occupying the road for business of the present invention.

[0059] Figure 8 It is a flowchart of comparing and excluding the individuals staying abnormally with the store information where the monitoring blind area is located to obtain the mobile road occupation data of the present invention. Detailed implementation manners

[0060] The following will describe in detail the embodiments of the present invention. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below by referring to the drawings are exemplary and are intended to explain the present invention, and should not be construed as a limitation to the present invention.

[0061] The first embodiment

[0062] Please refer to Figures 1 to 8, the present invention provides a management method for urban road occupation governance, including:

[0063] S101 Obtain a grid-based street monitoring image from the image acquired by the roadside camera;

[0064] The specific steps include:

[0065] S201 Set the image acquisition frequency to capture the image information on the street;

[0066] Set the image acquisition frequency to capture the dynamic changes on the street. This step is crucial because it determines how quickly the system can respond to real-time events. The choice of acquisition frequency must balance the data volume and real-time requirements; overly frequent acquisition may lead to an excessive data processing burden, while too low an acquisition frequency may result in missing key events.

[0067] S202 Use a filtering algorithm to remove the noise in the image information;

[0068] Then use a filtering algorithm to remove the noise in the image. Due to environmental factors such as weather conditions, light changes, and the limitations of the camera itself, the image often contains various forms of noise. This noise will interfere with subsequent image analysis work. Adopting advanced filtering techniques, such as adaptive filters or deep learning models, can effectively reduce unnecessary interference and ensure the image quality.

[0069] S203 Crop the image to focus on the street area;

[0070] Then crop the image to focus on the street area. To improve the processing efficiency and focus on the area of interest, that is, the actual street part, it is necessary to crop the original image. This process may involve computer vision techniques such as edge detection to accurately locate the street boundary and exclude non-related backgrounds (such as the sky, buildings, etc.), so as to make the processing more efficient and targeted.

[0071] S204 Divide the street area into several grid cells of the same size;

[0072] Divide the street area into several grid cells of the same size. This is done to achieve finer-grained data management and analysis. Each grid cell represents a small segment of the street. By monitoring and analyzing each grid cell separately, more detailed and accurate street condition information can be obtained. This division method is also beneficial for subsequent spatial data analysis and pattern recognition.

[0073] S205 Use image stitching technology to synthesize the grid cells generated from the monitoring images in all areas into a complete picture to obtain the street monitoring image information.

[0074] Use image stitching technology to synthesize the grid cells generated from the monitoring images in all areas into a complete picture. This step involves image processing algorithms such as feature point matching and perspective transformation to ensure seamless connection between different grid cells. The final output is a high-resolution image that comprehensively reflects the entire street condition. This image is composed of multiple individual grid cell images stitched together, but appears as a coherent whole. The street monitoring image information obtained in this way not only helps traffic management departments with effective urban planning and management, but also provides the public with a more intuitive and detailed display of the street condition.

[0075] S102 Analyze the street monitoring image to obtain the location and scope of the blind spots in street monitoring;

[0076] The specific steps include:

[0077] S301 Obtain street monitoring image information;

[0078] This step relies on the previously captured and processed grid street images by cameras. These images provide the visual data basis of the street and its surrounding environment, which is the premise for further analysis.

[0079] S302 Identify the stationary obstacles in the road monitoring image information based on the occlusion recognition method to obtain the obstacle information;

[0080] To identify the objects that may affect the monitoring effect, such as parked vehicles, temporary structures, trees or other factors that block the line of sight.

[0081] The specific steps include:

[0082] S401 Collect the characteristics of common obstacles in the target area;

[0083] First, systematically collect a series of data samples of typical obstacles in the target area. This step aims to construct a detailed and representative database covering all types of objects that may affect road monitoring effects. Data collection can be completed in various ways, such as using high-definition cameras to capture high-resolution images or video sequences, using lidar (LiDAR) to obtain three-dimensional point cloud data, or combining drone photography, etc. For each sample, its appearance characteristics such as shape, size, color, texture, etc. should be recorded, and their positions and categories should be marked.

[0084] S402 Perform feature matching in the road monitoring image information based on the characteristics of common obstacles to obtain the target obstacles;

[0085] Apply machine learning or deep learning algorithms to real-time road monitoring images and conduct comparative analysis with a pre-established obstacle feature library. In this process, advanced pattern recognition algorithms such as convolutional neural networks (CNNs) and support vector machines (SVMs) will be used to achieve precise positioning of potential obstacles in the images. By comparing the newly acquired image features with the known features in the database, objects that exist in the road environment but do not belong to normal traffic elements can be effectively detected.

[0086] S403 Extract obstacle information based on the target obstacle, and the obstacle information includes the obstacle type and location.

[0087] Once the existence of suspected obstacles is determined, the next step is to extract detailed information about these obstacles. This not only involves identifying which category they belong to (such as vehicles, buildings, or naturally growing plants), but also includes precisely locating their spatial coordinates. To achieve this, computer vision techniques such as geometric transformation and perspective projection are usually combined to map the obstacles on the two-dimensional image back to the three-dimensional physical world, thereby obtaining a more accurate position description.

[0088] S303 Compare the road area marked on the street monitoring image information with the location of the obstacle information to obtain the location and range of the street monitoring blind area.

[0089] In this step, the known road layout is matched with the detected obstacle information to determine which areas are difficult to be effectively monitored due to the existence of obstacles. This comparison can be completed through a geographic information system (GIS) or other spatial analysis tools. The ultimate goal is to draw a map showing all potential blind areas. This not only helps to evaluate the effectiveness of the current monitoring system, but also can guide the decision-making of future camera deployment or obstacle removal, thereby improving the safety and management efficiency of the entire street.

[0090] S103 Obtain import and export area information based on the range of the street monitoring blind area, and establish a monitoring blind area associated image dataset based on the images monitored by the import and export area information;

[0091] The specific steps include:

[0092] S501 Obtain the map data where the blind area is located;

[0093] This step depends on the location and range of the blind area determined by the previous analysis (as described in S102). The map data not only includes the basic information of the road network, but may also cover detailed geographical features such as buildings and green belts. To accurately reflect the real situation, the latest high-resolution map data should be used.

[0094] S502 Mark the key points of the road on the map data;

[0095] These key points usually refer to strategically significant positions on the road, such as intersections, turning points, bridge entrances, or tunnel exits, etc. For blind spots in monitoring, it is particularly important to mark the access points to and from the blind spots - that is, the locations where vehicles or pedestrians enter and leave the blind spots. These key points will serve as reference coordinates in subsequent steps for associating camera positions and fields of view.

[0096] S503 Obtain the corresponding camera numbers based on the road key points;

[0097] Each camera installed near a key point has a unique number, which is used to identify and manage a large number of monitoring devices. Through the integration of GIS technology and the camera database, the closest camera can be automatically matched according to the position of the key point. If there is no suitable camera coverage near a certain key point, it is necessary to consider redeploying or adjusting the angles of existing cameras to ensure effective coverage.

[0098] S504 Compose the images obtained from relevant cameras into an associated image dataset for the blind spot in monitoring.

[0099] This step involves synchronizing images from different sources according to timestamps to ensure the temporal consistency between images. At the same time, it is also necessary to preprocess the images, such as correcting lens distortion, enhancing contrast, etc., in order to improve the image quality. The finally formed image dataset should be able to comprehensively reflect the situation of the blind spot in monitoring and its surrounding environment, providing a solid data foundation for subsequent analysis work.

[0100] S104 Obtain the characteristics of the objects occupying the road and perform identity marking based on the associated image dataset;

[0101] The specific steps include:

[0102] S601 Use the Canny algorithm to detect the contours of each object in the associated image dataset;

[0103] In order to extract the boundary information of objects from a complex background, the Canny edge detection algorithm is used to process each frame of the image in the associated image dataset. The Canny algorithm is a multi-level edge detection method that can effectively suppress noise while retaining real edge information. The specific steps include: applying Gaussian blur to the original image to reduce noise. Calculating the intensity gradient magnitude and direction of each pixel point in the image to determine the position of the edge. Only retaining the gradients of local maxima as edges and removing other non-edge pixels. Using two thresholds, a high threshold and a low threshold, to trace strong edges and weak edges. Forming continuous contours by connecting strong edges and some weak edges.

[0104] After these steps, a clear outline of the object can be obtained, which provides a basis for subsequent object recognition.

[0105] S602 Apply an object detection algorithm to identify all object types in the object outline and frame their positions;

[0106] Next, apply advanced object detection algorithms (such as YOLO, SSD, or Faster R-CNN) to identify all object types within the object outline and accurately frame their positions. Object detection algorithms can not only distinguish different object categories (such as vehicles, pedestrians, stalls, etc.), but also accurately mark the position of each object in the image.

[0107] In addition, additional features such as color and texture can be combined to enhance the recognition effect, ensuring that even objects with similar appearances can be correctly classified.

[0108] S603 Mark the identity of the objects.

[0109] The last step is to mark the identity of the detected objects. This step is very important for long-term monitoring and management because it can help distinguish legal business activities from illegal occupation of road space. The identity marking can be achieved in the following ways: Compare the detected objects with the existing merchant registration information to confirm whether they belong to legal operators. By analyzing the appearance frequency and duration of objects in long-term sequence images at the same location, judge whether there is illegal occupation of road space. For situations that are difficult to determine automatically, introduce a manual review mechanism, and let the staff make a final judgment according to the actual situation. Generate a unique identity label for each detected object, including but not limited to object ID, type, position coordinates, first appearance time, and status (legal / illegal). These labels can be stored in the database for subsequent query and management.

[0110] S105 Based on the identity marking, record the time interval of the objects occupying the road space entering and leaving the monitoring blind area, and judge the individuals occupying the road space;

[0111] The specific steps include:

[0112] S701 Use a motion detection algorithm to capture the events of each identity-marked object entering and leaving the monitoring blind area;

[0113] Use advanced motion detection algorithms (such as background subtraction, optical flow method, or deep learning model) to capture the events of each identity-marked object entering and leaving the monitoring blind area. The motion detection algorithm can distinguish the static background from the moving objects and ensure accurate capture of each entry and exit moment. The key to this step lies in:

[0114] Build a stable background model to distinguish between the background and moving objects in the foreground. Set a reasonable motion detection threshold according to the specific application scenario to avoid false alarms. Detect the movement of objects by comparing the differences between consecutive frames. If the monitoring blind area is covered by multiple cameras, it is necessary to ensure the time synchronization between different cameras to guarantee the consistency of event capture.

[0115] S702 records the entry timestamp for each detected entry event and the exit timestamp for each detected exit event;

[0116] Once an event of an object entering or leaving the monitoring blind area is captured, immediately record the corresponding timestamp for each event. This involves ensuring that all cameras and processing units use the same time source to achieve millisecond-level timestamp recording. Create detailed log entries for each entry and exit event, including the object ID, type, location coordinates, and precise timestamps. Considering the large amount of timestamp data that may be generated, efficient data structures and compression techniques need to be adopted to optimize storage.

[0117] S703 calculates the time interval that the object stays in the monitoring blind area for each entry and exit event;

[0118] For each entry and exit event, calculate the time interval that the object stays in the monitoring blind area. This step can be completed through simple arithmetic operations, that is, subtracting the entry timestamp from the exit timestamp. To improve accuracy, the following factors should also be considered: correct the minor errors caused by network latency or device response time. Handle incomplete events (such as only entry without exit) caused by device failures or external interferences, and a reasonable time limit can be set as the default exit time.

[0119] S704 uses the Z-score algorithm to identify business objects with time intervals beyond the normal range and obtain abnormal staying individuals.

[0120] Use the Z-score algorithm in statistics to identify business objects with time intervals beyond the normal range. The Z-score is a method to measure the degree of deviation of a data point from the average value and can help identify outliers.

[0121] The specific steps are as follows:

[0122] Data collection and preprocessing: Collect data on the time intervals that all objects stay for a period of time and remove obvious outliers or incorrect records.

[0123] Calculate the mean and standard deviation: Calculate the average value (μ) and standard deviation (σ) of all time interval data.

[0124] Calculate the Z-score: For each time interval t, calculate its Z-score: Z = σ(t - μ)

[0125] According to urban management and policy requirements, set a reasonable Z-score threshold, such as |Z| > 3, to determine whether it belongs to abnormal stay. Identify the objects corresponding to all time intervals with Z-scores exceeding the set threshold as individuals with abnormal stay, and generate a report for further review.

[0126] S106 Compare and exclude the information of the individuals with abnormal stay and the stores in the monitoring blind area to obtain the data of mobile occupation of road space.

[0127] The specific steps include:

[0128] S801 Obtain the store data within the monitoring area;

[0129] It is necessary to obtain all the legal store data within the monitoring area. This step involves multiple aspects of work:

[0130] Database integration: Obtain the latest store registration information from official channels such as the industrial and commercial management department and the urban planning department, and integrate it into a unified database.

[0131] Geographic Information System (GIS) integration: Ensure that the store data contains detailed geographical location information, such as latitude and longitude coordinates, street addresses, etc., for subsequent spatial analysis.

[0132] Regular update: Considering the opening, closing or change of stores, establish a regular update mechanism to ensure the timeliness and accuracy of the data.

[0133] Multi-source data fusion: In addition to official data, public resources such as social media and business review websites can also be combined to supplement the relevant information of stores.

[0134] S802 Based on the store data, confirm whether the individual with abnormal stay belongs to a store. If not, it is the data of mobile occupation of road space.

[0135] Based on the collected store data, confirm whether the individual with abnormal stay belongs to a legal store. This involves the following specific operations: Use GIS technology to perform spatial matching between the location of the individual with abnormal stay and the store location to determine whether it is within the legal business scope of a store. Use image processing technology to identify features such as logos and signs on the object and compare them with the picture materials in the store registration information.

[0136] For cases that are difficult to determine automatically, an artificial review mechanism is introduced, and the staff makes the final judgment based on the actual situation. Once it is confirmed that the abnormally staying individuals do not belong to any legal store, they are marked as mobile sidewalk occupation data. These data reflect unauthorized sidewalk occupation activities, which are the focus of urban management.

[0137] S107 Extract the image features in the mobile sidewalk occupation data, and match and compare the image features in the entire monitoring network to predict the mobile sidewalk occupation situation of individuals.

[0138] By extracting the image features in the mobile sidewalk occupation data and performing matching and comparison in the entire monitoring network, the mobile sidewalk occupation situation of individuals is predicted. This process aims to utilize the similarity of image features to track and predict the behavior patterns of sidewalk occupation individuals.

[0139] Specifically, image features that can effectively represent sidewalk occupation individuals can be selected, such as shape, color, texture, etc. Use a pre-trained deep learning model (such as a convolutional neural network CNN) to automatically extract high-level feature representations from the images. Convert the extracted features into fixed-length feature vectors for subsequent comparison and matching.

[0140] Search for images in the entire city's monitoring network that match the features of the known mobile sidewalk occupation data to expand the tracking scope. In a specific area (such as the same block or adjacent streets), focus on finding images with similar features to improve the matching efficiency. Combine the time and space dimensions to analyze the movement patterns between different locations and predict the new locations that individuals may go to. As new data is continuously added, continuously update and optimize the matching algorithm to ensure the adaptability and accuracy of the system.

[0141] Based on the historical movement patterns, predict the possible future paths and locations of sidewalk occupation individuals. Identify the areas where mobile sidewalk occupation behaviors occur frequently and deploy management measures in advance. Through long-term data analysis, understand the changing trends of mobile sidewalk occupation behaviors and provide a basis for policy formulation.

[0142] Second Embodiment

[0143] The present invention also provides a management system for urban road occupation governance, including a monitoring image acquisition module, a monitoring blind area detection module, an import and export area acquisition module, an identity marking module, an illegal occupation judgment module, and an exclusion module; the monitoring image acquisition module is used to obtain a grid-based street road monitoring image based on the images acquired by roadside cameras; the monitoring blind area detection module is used to analyze the street road monitoring image to obtain the position and range of the street monitoring blind area; the import and export area acquisition module is used to obtain import and export area information based on the street monitoring blind area range, and establish a monitoring blind area associated image dataset based on the images monitored by the import and export area information; the identity marking module is used to obtain the characteristics of the objects occupying the road for business based on the associated image dataset, and perform identity marking; the illegal occupation judgment module is used to judge the individuals occupying the road for business based on the time intervals of the objects occupying the road for business entering and leaving the monitoring blind area recorded by the identity marking; the exclusion module is used to compare and exclude the information of the individuals staying abnormally and the stores where the monitoring blind area is located to obtain mobile road occupation data.

[0144] In this embodiment, the monitoring image acquisition module collects images from the cameras installed on the street and converts these images into grid-based street road monitoring images. Set an appropriate image acquisition frequency to capture the dynamic changes on the street; use a filtering algorithm to remove image noise and improve the image quality.

[0145] The monitoring blind area detection module, based on the acquired street road monitoring image, through a series of image processing steps, identifies the areas with insufficient monitoring and their influence ranges caused by obstacles and other reasons. Use long-time series images to construct a stable background model, compare the differences between consecutive frames to detect static obstacles, and apply an edge detection algorithm to find non-background contours.

[0146] The import and export area acquisition module obtains import and export area information based on the monitoring blind area range and establishes a monitoring blind area associated image dataset. Specifically, obtain map data containing the blind area position and mark key points such as intersections and turning points on the map. Based on the key point positions, obtain the corresponding camera numbers to ensure effective coverage of the blind area and its surroundings.

[0147] The identity marking module automatically detects and classifies the objects occupying the road for business appearing on the street based on the associated image dataset, and performs identity marking on them. Use the Canny algorithm or other edge detection methods to extract the object contours. Apply a deep learning model (such as YOLO, SSD) to identify the object types and frame their positions. Generate a unique identity label for each detected object, including information such as object ID, type, position coordinates, first appearance time, and status.

[0148] The illegal occupation of public space judgment module records the time intervals when objects occupying public space enter and leave the monitoring blind area, and through time management and statistical analysis, identifies individuals with abnormal stays. It uses motion detection algorithms to capture object entry and exit events and records accurate timestamps. It calculates the time intervals that objects stay within the monitoring blind area. The Z-score algorithm is used to identify time intervals that exceed the normal range, and individuals with abnormal stays are obtained.

[0149] The exclusion module compares individuals with abnormal stays with legal store information to distinguish between legal operations and illegal occupation of public space, and finally obtains mobile illegal occupation of public space data.

[0150] In summary, through the collaborative work of each module, this intelligent management system realizes the refined management of urban illegal occupation of public space behavior. It not only helps to maintain the cleanliness and safety of the city, but also provides strong technical support for relevant law enforcement departments, improving the efficiency and intelligence level of urban management.

[0151] The above-disclosed is only a preferred embodiment of the present invention. Of course, it cannot be used to limit the scope of the rights of the present invention. Those of ordinary skill in the art can understand all or part of the processes of implementing the above embodiments, and the equivalent changes made according to the claims of the present invention still fall within the scope covered by the invention.

Claims

1. A management method for urban road occupation management, It is characterized in that Including: obtaining gridded street road monitoring images based on images acquired by roadside cameras; Analyze street monitoring images to obtain the location and range of street monitoring blind spots; The import and export area information is obtained based on the street monitoring blind area range, and the monitoring blind area associated image data set is established based on the images obtained by the import and export area information monitoring; Acquire the features of road-occupying business objects based on the associated image dataset and perform identity tagging; Based on the identity tag, the time interval between the objects occupying the road and entering and exiting the monitoring blind area is recorded to identify the individuals occupying the road; Compare and exclude the information of abnormally staying individuals and stores in the monitoring blind spots to obtain the mobile road occupancy data.

2. A management method for urban road occupation management as claimed in claim 1, characterized in that: After comparing and eliminating the abnormally staying individuals and the store information in the monitoring blind spots to obtain the mobile road occupation data, the method also includes extracting image features in the mobile road occupation data, matching and comparing the image features in the entire monitoring network to predict the individual's mobile road occupation situation.

3. A management method for urban road occupation management as claimed in claim 2, characterized in that: The specific steps of obtaining a gridded street monitoring image based on the image acquired by the roadside camera include: Set the image acquisition frequency to capture image information on the street; Use filtering algorithms to remove noise from image information; The image was cropped to focus on the street area; Divide the street area into a number of grid cells of uniform size; Image stitching technology is used to combine the grid units generated by monitoring images in all areas into a complete picture to obtain street road monitoring image information.

4. A management method for urban road occupation management as claimed in claim 3, characterized in that: The specific steps of analyzing the street monitoring image to obtain the location and range of the street monitoring blind area include: Obtain street monitoring image information; Based on the occlusion recognition method, the stationary obstacles in the road monitoring image information are identified to obtain the obstacle information; The road area and obstacle information positions marked on the street monitoring image information are compared to obtain the location and range of the street monitoring blind spot.

5. A management method for urban road occupation management as claimed in claim 4, characterized in that: The specific steps of identifying the stationary obstacles in the road monitoring image information based on the occlusion recognition method to obtain the obstacle information include: Collect common obstacle features in the target area; Based on the common obstacle features, feature matching is performed on the road monitoring image information to obtain the target obstacle; Obstacle information is extracted based on the target obstacle, where the obstacle information includes obstacle type and position.

6. A management method for urban road occupation management as claimed in claim 5, characterized in that: The specific steps of acquiring the import and export area information based on the street monitoring blind area range, and establishing the monitoring blind area associated image data set based on the images obtained by monitoring the import and export area information include: Obtain map data of the blind spot location; Mark key road points on map data; Get the corresponding camera number based on the road key point; Based on the camera numbers, the images obtained by related cameras are combined into a monitoring blind spot associated image data set.

7. A management method for urban road occupation management as claimed in claim 6, characterized in that: The specific steps of acquiring the features of the road-occupying business object based on the associated image data set and performing identity marking include: Use the Canny algorithm to detect the contours of each object in the associated image dataset; Apply object detection algorithms to identify all object types in the object outline and frame their locations; Label objects with their identities.

8. A management method for urban road occupation management as claimed in claim 7, characterized in that: The specific steps of recording the time interval of the road-occupying business object entering and exiting the monitoring blind area based on the identity tag and determining the road-occupying business individual include: Use motion detection algorithms to capture every time an ID-tagged object enters or exits a blind spot; Record an entry timestamp for each detected entry event and a departure timestamp for each detected departure event; For each entry and exit event, calculate the time interval that the object stays in the monitoring blind area; The Z-score algorithm is used to identify operating objects with time intervals beyond the normal range and obtain abnormal stay individuals.

9. A management method for urban road occupation management as claimed in claim 8, characterized in that: The specific steps of comparing and eliminating the abnormally staying individuals and the store information in the monitoring blind area to obtain the mobile road occupation data include: Get store data within the monitoring area; Based on the store data, confirm whether the abnormal staying individual belongs to the store. If not, it is mobile road occupation data.

10. A management system for urban road occupation management, applied to a management method for urban road occupation management as claimed in any one of claims 1 to 9, characterized in that: It includes monitoring image acquisition module, monitoring blind spot detection module, import and export area acquisition module, identity marking module, road occupation business judgment module and exclusion module; The monitoring image acquisition module is used to obtain a gridded street monitoring image based on the image acquired by the roadside camera; The monitoring blind spot detection module is used to analyze the street monitoring image to obtain the location and range of the street monitoring blind spot; The import and export area acquisition module is used to obtain the import and export area information based on the street monitoring blind area range, and establish a monitoring blind area associated image data set based on the images obtained by the import and export area information monitoring; The identity marking module is used to obtain the characteristics of the objects occupying the road for business based on the associated image data set and perform identity marking; The road occupation business judgment module is used to record the time interval of the road occupation business object entering and exiting the monitoring blind area based on the identity tag, and judge the road occupation business individual; The exclusion module is used to compare and exclude the information of abnormally staying individuals and stores in the monitoring blind area to obtain mobile road occupation data.

Images