Urban governance data fusion analysis method and system
By combining fixed and mobile video acquisition equipment, the street image data of shops is collected and analyzed in real time, the road occupation abnormalities are identified and processed, rectification notices are sent and monitoring strategies are adjusted, and the detection frequency and arbitrary nature of existing urban governance methods are solved. The problem has achieved an efficient and accurate urban governance data fusion analysis method.
Patent Information
- Application Number
- CN202411452851.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-17
- Publication Date
- 2024-11-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing urban governance methods mainly rely on manual inspections, with low detection frequency and high arbitrary nature, resulting in low governance efficiency and difficulty in effectively supervising the problem of shops occupying public areas.
By combining fixed monitoring equipment and mobile video acquisition equipment, street image data of shops is collected in real time, evaluation parameters are established, abnormal data is identified, rectification notices are sent, and monitoring strategies are dynamically adjusted according to the monitoring level to achieve efficient monitoring of shop operations.
It reduces monitoring costs, improves monitoring efficiency, ensures the rational use of public spaces and the maintenance of urban order, and achieves efficient and precise supervision of commercial activities.
Smart Images

Figure CN118968322A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of digital data processing, and in particular to a method and system for fusion analysis of urban governance data. Background Art
[0002] In order to expand their business space, many street shops in cities place their goods on the sidewalks or occupy public areas. This behavior may hinder pedestrians and cause safety hazards. Specifically, occupying public areas includes merchants piling goods or setting up car blocking piles outside the store due to the small size of the store. The city management department will manage this problem.
[0003] However, the existing method mainly relies on manual driving to conduct inspections, which has a low inspection frequency and is more arbitrary, thus reducing management efficiency. Summary of the invention
[0004] The purpose of the present invention is to provide a method and system for fusion analysis of urban governance data, aiming to monitor the operation of stores by combining existing monitoring equipment with special patrol equipment, thereby reducing monitoring costs and improving monitoring efficiency.
[0005] To achieve the above-mentioned purpose, in a first aspect, the present invention provides a method for fusion analysis of urban governance data, comprising acquiring street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment; Establishing evaluation parameters for urban management and operation behaviors based on historical image data, wherein the evaluation parameters include the area of street occupied by shops, the time period of street occupied, the location of street occupied, and the type of street occupied object; Identify the collected street image data and record abnormal data that deviates from the evaluation parameters; When the number of abnormalities detected by the store reaches a preset value, a rectification notice will be sent to the store manager; Generate a corresponding monitoring level based on the frequency of rectification notices sent to the store during the assessment period; Dynamically adjust monitoring strategies based on monitoring levels.
[0006] The specific steps of obtaining street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment include: Continuously acquiring street images through fixed monitoring equipment, and stitching the images based on the position and shooting parameters of the monitoring equipment to obtain a stitched image; Get the position coordinates of the missing image in the stitched image; At every preset time period, the mobile video acquisition device is notified to acquire the second street image data based on the position coordinates, and the missing image is supplemented to obtain the street image data.
[0007] The specific steps of continuously acquiring street images through fixed monitoring equipment and stitching the images based on the position and shooting parameters of the monitoring equipment to obtain the stitched images include: Obtain images continuously collected by fixed monitoring equipment; Correct the acquired images to reduce lens distortion; Use SIFT algorithm to detect and match feature points in adjacent images. If there is no matching feature point, mark it as a missing image between two adjacent images. The transformation matrix is calculated based on the feature matching results, and the images are stitched into a panorama.
[0008] The specific steps of identifying the collected street image data and recording abnormal data that deviates from the evaluation parameters include: Identifying corresponding parameters in the street image data based on the evaluation parameters to obtain target parameters; Setting a threshold range for each evaluation parameter based on historical image data; The target parameter is compared with the threshold range of each evaluation parameter, and when it exceeds the threshold range, it is determined to be abnormal data.
[0009] The specific steps of identifying corresponding parameters in the street image data based on the evaluation parameters to obtain the target parameters include: Use the YOLO target detection model to detect objects in the image and mark the location of each object; Identify and classify objects occupying the road based on pattern matching; For the detected road-occupying objects, calculate their pixel areas in the image; Converting pixel area into actual area based on camera parameters; Get the location and time when the image was taken.
[0010] The specific steps of sending a rectification notice to the store manager when the number of abnormalities detected by the store accumulates to a preset value include: Each time an anomaly is detected, the target parameters of the anomaly event are recorded in the database; Track the number of abnormal events for each store; When the accumulation of abnormal events in a store reaches a preset threshold, a communication channel is selected and a rectification notice is sent to the store manager.
[0011] Among them, the rectification notice contains specific details of the violation and rectification suggestions.
[0012] The specific steps of generating the corresponding monitoring level based on the frequency of rectification notices sent to the shops during the evaluation period include: Define different compliance level ranges based on the number of rectification notices; Match the corresponding compliance level based on the number of rectification notices sent by the target store; The rectification notices are cleared at preset intervals, and the compliance levels of the shops are recalculated and adjusted.
[0013] In the second aspect, the present invention also provides a city governance data fusion and analysis system, including a data acquisition module, an evaluation parameter generation module, an abnormality judgment module, a notification module, a monitoring level matching module and a monitoring strategy adjustment module; the data acquisition module is used to obtain street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment; the evaluation parameter generation module is used to establish evaluation parameters of urban management and operation behaviors based on historical image data, and the evaluation parameters include the area occupied by shops, the time period of occupying the road, the location of occupying the road and the type of occupying objects; the abnormality judgment module is used to identify the collected street image data and record abnormal data that deviates from the evaluation parameters; the notification module is used to send a rectification notice to the shop manager when the number of abnormalities detected by the shop accumulates to a preset value; the monitoring level matching module is used to generate a corresponding monitoring level based on the frequency of rectification notices sent to the shop during the evaluation period; the monitoring strategy adjustment module is used to dynamically adjust the monitoring strategy according to the monitoring level.
[0014] The present invention provides a method and system for fusion analysis of urban governance data, including obtaining street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment; establishing evaluation parameters of urban management and operation behavior based on historical image data, wherein the evaluation parameters include the area occupied by shops, the time period occupied, the location occupied and the type of objects occupied; identifying the collected street image data and recording abnormal data that deviates from the evaluation parameters; sending rectification notices to shop managers when the number of abnormalities detected by shops accumulates to a preset value; generating corresponding monitoring levels based on the frequency of rectification notices sent to shops during the evaluation period; and dynamically adjusting monitoring strategies according to the monitoring levels. In this way, the operation of shops can be monitored by combining existing monitoring equipment with dedicated patrol equipment, which reduces monitoring costs and improves monitoring efficiency, so that urban management agencies can more efficiently and accurately supervise commercial activities in cities, ensure the rational use of public space, and maintain good urban order. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0016] Figure 1 It is a flow chart of a method for fusion analysis of urban governance data according to the first embodiment of the present invention.
[0017] Figure 2 It is a flow chart of the first embodiment of the present invention for acquiring street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment.
[0018] Figure 3 This is a flowchart of the first embodiment of the present invention, which continuously acquires street images through fixed monitoring equipment, and splices the images based on the position and shooting parameters of the monitoring equipment to obtain a spliced image.
[0019] Figure 4 It is a flowchart of the first embodiment of the present invention for identifying the collected street image data and recording abnormal data deviating from the evaluation parameters.
[0020] Figure 5 This is a flowchart of the first embodiment of the present invention for identifying corresponding parameters in street image data based on evaluation parameters to obtain target parameters.
[0021] Figure 6 This is a flowchart of sending a rectification notice to a store manager when the number of abnormalities detected by a store accumulates to a preset value according to the first embodiment of the present invention.
[0022] Figure 7 This is a flowchart of the first embodiment of the present invention for generating corresponding monitoring levels based on the frequency of rectification notices sent to shops within an evaluation period. DETAILED DESCRIPTION
[0023] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.
[0024] First embodiment: See also Figure 1 to Figure 7 The present invention provides a method for fusion analysis of urban governance data, comprising: S101 obtains street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment; The specific steps include: S201 continuously acquires street images through a fixed monitoring device, and splices the images based on the position and shooting parameters of the monitoring device to obtain a spliced image; Configure the shooting parameters of the fixed monitoring device, including but not limited to resolution, frame rate, exposure time, etc., start the fixed monitoring device, and start continuously acquiring street images. The specific steps include: S301 obtains images continuously collected by fixed monitoring equipment; Turn on the fixed monitoring device and ensure that it is in normal working condition. Then receive the image data transmitted by the fixed monitoring device in real time through the network interface. Then save the image data in chronological order and record the timestamp of each image.
[0025] S302 corrects the collected image to reduce lens distortion; Load the parameters required for correction, such as focal length, principal point position, radial distortion coefficient, etc. Then apply the distortion correction algorithm to the image, such as the undistort function in OpenCV.
[0026] S303 uses SIFT algorithm to detect and match feature points in adjacent images, and if there is no matching feature point, marks the image between two adjacent images as a missing image; Run the SIFT algorithm on each image to extract feature points and their descriptors. Then use a feature matching algorithm (such as FLANN or Brute-Force matcher) to find common feature points between adjacent images. Use the RANSAC algorithm to filter out incorrect matching points to improve matching accuracy. If there are not enough matching feature points between adjacent images (for example, less than a certain number), mark the missing image between the two images. Finally, record the number and location information of the matching feature points.
[0027] S304 calculates a transformation matrix according to the feature matching result and stitches the images into a panoramic image.
[0028] Calculate the homography transformation matrix based on the matched feature points. Apply the transformation matrix to the images to align them. Fuse the aligned images to generate a panoramic image. Then crop or pad the panoramic image to get the final complete image. Save the stitched panoramic image and record the relevant metadata (such as stitching date, time, location, etc.).
[0029] S202 obtains the position coordinates of the missing image in the stitched image; Analyze the stitched image and identify the areas marked as missing images. For each missing area, record its location coordinates in the stitched image. Based on the location information of the monitoring equipment, calculate the geographic coordinates of the missing area in the actual scene. Organize the location coordinate information of the missing area into a data structure for subsequent processing.
[0030] S203: Notify the mobile video acquisition device to acquire the second street image data based on the location coordinates at every preset time period, and supplement the missing image to obtain the street image data.
[0031] A preset time interval is set to trigger the acquisition task of the mobile video acquisition device. When the preset time interval is reached, the missing image position coordinates recorded in step S202 are read. According to the position coordinates of the missing image, the actual position that the mobile video acquisition device needs to reach is calculated. A task instruction containing the position coordinates is sent to the mobile video acquisition device, instructing the device to go to the corresponding position. After receiving the instruction, the mobile video acquisition device navigates to the specified position and starts to acquire images. After the mobile video acquisition device completes image acquisition at the specified position, it transmits the image data back to the central processing system. The image data sent back by the mobile video acquisition device is received and a preliminary quality check is performed. The image data collected by the mobile device is matched and fused with the spliced image of the fixed monitoring device to supplement the missing image part. Use similar image splicing technology (such as SIFT algorithm, etc.) to fuse the newly collected image with the existing spliced image. Finally, the spliced image is updated to replace or fill the previous missing image area to generate complete street image data.
[0032] S102: establishing evaluation parameters of urban management and operation behaviors based on historical image data, wherein the evaluation parameters include the area of street occupied by shops, the time period of street occupied, the location of street occupied, and the type of street occupied object; The goal of this step is to extract key information from historical image data and establish a set of evaluation parameter systems to evaluate whether the business activities of shops in the city comply with management regulations. The evaluation parameters include the area occupied by shops, the time period occupied, the location occupied, and the type of objects occupied.
[0033] Specifically, it is necessary to collect urban street surveillance image data from multiple sources over a period of time. The image data is then preprocessed, such as denoising and contrast enhancement, to improve image quality. The images are then classified and stored according to time sequence and geographic location.
[0034] Develop computer vision algorithms, such as edge detection and shape recognition, to identify the specific scope of shops occupying the road. Use machine learning methods (such as deep learning models) to automatically identify shops and their occupying behaviors. Combine with geographic information system (GIS) data to locate the specific location of shops.
[0035] For each identified road-occupied area, calculate the ground area it covers. Considering the size deviation that may be caused by shooting at different angles, use a correction algorithm to correct the area measurement value. Analyze the changing trend of the road-occupied area in different time periods.
[0036] By analyzing the image sequence, determine when shops start occupying the sidewalk and when they end. Use time series analysis methods to identify peak and trough times for sidewalk occupation. Record the start and end times of sidewalk occupation every day, and calculate the average sidewalk occupation time per week or month.
[0037] Use GPS data or map coordinates to locate the specific location of the store. Mark the street occupation area of each store on the map. Draw a heat map to show the frequency and density of street occupation in different areas.
[0038] Classify shops based on their appearance, business nature, etc. Identify different types of shops or vendors through image analysis.
[0039] Finally, the data on the area occupied by shops, the time period occupied, the location occupied and the type of objects occupied are summarized. A database is established to store these evaluation parameters.
[0040] S103 identifies the collected street image data and records abnormal data that deviates from the evaluation parameters; This step aims to analyze the collected street image data using the established evaluation parameters to identify and record abnormal data that deviates from the normal range. The specific steps include: S301 identifies corresponding parameters in the street image data based on the evaluation parameters to obtain target parameters; The specific steps include: S401 uses the YOLO target detection model to detect objects in the image and mark the location of each object; Load the YOLO (You Only Look Once) model. Preprocess the image to make it meet the model input requirements. Run the model to analyze the image and output the bounding box and category of each object. Record the location information of each recognized object.
[0041] S402 identifies and classifies objects occupying the road based on pattern matching; Use pattern matching technology to identify which objects are associated with road occupation. Categorize road occupation objects, such as whether they are fixed shops or mobile vendors.
[0042] S403, for the detected road-occupying object, calculating its pixel area in the image; Calculate the number of pixels occupied by each object in the image and save these pixel area data as the basis for subsequent analysis.
[0043] S404 converts the pixel area into an actual area based on the camera parameters; The scale factor is calculated based on the focal length, field of view and other parameters of the camera, as well as the distance at which the image is taken. The scale factor is applied to convert the pixel area into the actual area.
[0044] S405 obtains the location and time when the image was captured.
[0045] Use the camera's built-in GPS module or an external GPS device to obtain the shooting location. Record the timestamp of the captured image.
[0046] S302 sets a threshold range for each evaluation parameter based on historical image data; Analyze the statistical data in the historical image data, such as mean, standard deviation, etc. Set the upper and lower limits of the evaluation parameters based on the statistical data. Considering factors such as seasonal changes and holidays, it may be necessary to dynamically adjust the threshold range.
[0047] S303 compares the target parameter with the threshold range of each evaluation parameter, and determines it as abnormal data when it exceeds the threshold range.
[0048] For each target parameter, compare it to the corresponding threshold range. If the target parameter is outside the threshold range, record the data as an anomaly. You may also need to consider special circumstances outside the threshold range, such as temporary permissions during holidays or special events.
[0049] S104 sends a rectification notice to the store manager when the number of abnormalities detected by the store reaches a preset value; The specific steps include: S501 Each time an abnormality is detected, the target parameters of the abnormal event are recorded in the database; Whenever abnormal data is detected, relevant information about the abnormal data is recorded in the database. The recorded information should include but is not limited to the store ID, detection time, abnormal type (such as excessive road occupation area, non-compliant road occupation time period, etc.), deviation between the actual value and the threshold range, etc. Ensure that the recorded information is detailed enough for subsequent data analysis and tracking.
[0050] S502 tracks the number of abnormal events in each store; Create an abnormal event counter for each store. Whenever a new abnormal event is recorded, the corresponding store counter is incremented. Check the value of the store abnormal event counter regularly to ensure the accuracy of the data.
[0051] S503 When the abnormal events of a certain store have accumulated to a preset threshold, a communication channel is selected and a rectification notice is sent to the store manager.
[0052] The rectification notice contains specific details of the violation and rectification suggestions.
[0053] Set a preset threshold, such as 5 abnormal events. When the store's abnormal event counter reaches or exceeds the preset threshold, trigger the procedure of sending a rectification notice. Select an appropriate communication channel, such as email, SMS or application notification, based on the store manager's preference or availability. Prepare the content of the rectification notice, which should include: Specific details of the abnormal event, such as the type, time, and location of the violation.
[0054] An assessment of the severity of the violation.
[0055] Clear recommendations for corrective actions and deadlines.
[0056] Subsequent follow-up steps and contact information.
[0057] Send the above rectification notice to the store manager and record the sending time and other relevant information. Update the status of the store in the database, such as marking it as "rectification notice sent" for subsequent tracking and management.
[0058] S105 generates a corresponding monitoring level based on the frequency of rectification notices sent to the store during the evaluation period; The specific steps include: Define different compliance level ranges based on the number of rectification notices; define compliance level standards, for example, four levels can be set up: "Excellent", "Good", "Fair", and "Unqualified".
[0059] Set a range of rectification notices for each compliance level, for example: “Excellent” level: 0-1 rectification notices.
[0060] “Good” level: 2-3 rectification notices.
[0061] “General” level: 4-6 rectification notices.
[0062] “Unqualified” level: 7 or more rectification notices.
[0063] Ensure that the standards for these compliance levels are consistent and transparent across the system.
[0064] S602 matches the corresponding compliance level according to the number of rectification notices sent by the target store; Query the number of rectification notifications for the store in the current evaluation cycle. According to the compliance level range defined in S601, find the specific compliance level corresponding to the number of rectification notifications for the store. Update the compliance level status of the store and record the matching results.
[0065] S603 clears the rectification notice every preset period and recalculates and adjusts the compliance level of the store.
[0066] Set an evaluation cycle, for example, every three months. At the end of each evaluation cycle, reset the rectification notification counts for all shops. Re-count the number of rectification notifications for shops starting from the new evaluation cycle. Re-evaluate the compliance level of the shops according to the method in S602. Update the compliance level status of the shops and take corresponding measures, such as increasing or decreasing the supervision intensity, if necessary.
[0067] S106 dynamically adjusts the monitoring strategy according to the monitoring level.
[0068] Specifically, four monitoring strategies, A, B, C, and D, can be defined, and each monitoring strategy can be matched with the grades of "excellent", "good", "average", and "unqualified". Design monitoring strategies. For example, grade A shops are reviewed annually, grade B shops are reviewed every six months, grade C shops are reviewed quarterly, and grade D shops are reviewed monthly. Conduct on-site inspections and customer satisfaction surveys in accordance with the monitoring strategies. Conduct a comprehensive review of the compliance levels of all shops every six months. Adjust the monitoring strategy based on the review results. For example, if a shop is upgraded from grade B to grade A, adjust its review frequency from once every six months to once a year.
[0069] Second embodiment: The present invention also provides a city governance data fusion and analysis system, comprising a data acquisition module, an evaluation parameter generation module, an abnormality judgment module, a notification module, a monitoring level matching module and a monitoring strategy adjustment module; the data acquisition module is used to obtain street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment; the evaluation parameter generation module is used to establish evaluation parameters of urban management and operation behaviors based on historical image data, and the evaluation parameters include the area occupied by shops, the time period of occupying the road, the location of occupying the road and the type of occupying objects; the abnormality judgment module is used to identify the collected street image data and record abnormal data that deviates from the evaluation parameters; the notification module is used to send a rectification notice to the shop manager when the number of abnormalities detected by the shop accumulates to a preset value; the monitoring level matching module is used to generate a corresponding monitoring level based on the frequency of rectification notices sent to the shop during the evaluation period; the monitoring strategy adjustment module is used to dynamically adjust the monitoring strategy according to the monitoring level.
[0070] In this embodiment, The data acquisition module obtains the image data of the street where the shops are located collected in real time by fixed monitoring equipment and mobile video acquisition equipment. The evaluation parameter generation module establishes the evaluation parameters of urban management and operation behavior based on historical image data.
[0071] The evaluation parameter of the shop area measures the actual area of public space occupied by shops. The occupation time records the time period of the shop occupying the public space. The occupation location determines the specific location occupied by the shop. The occupation object type classifies the reasons why the shop occupies the public space (such as placing goods, setting up temporary stalls, etc.). The model is trained using machine learning algorithms to extract features from historical image data and generate evaluation parameters. The evaluation parameters are continuously optimized and adjusted according to actual conditions and management needs.
[0072] The abnormality judgment module identifies the collected street image data and records the abnormal data that deviates from the evaluation parameters. It uses image processing technology and pattern recognition algorithms to compare the evaluation parameters and automatically identify the road occupation behaviors that do not comply with regulations. For any abnormal situations found, the system will record the specific deviation information, including time, location, violation type, etc.
[0073] The notification module sets the threshold of abnormal detection times according to the actual situation. Once the threshold is reached, the system will automatically send a notification containing the violation details to the store manager to remind him to make corrections.
[0074] The monitoring level matching module generates the corresponding monitoring level based on the frequency of rectification notices sent to the store during the evaluation period. When working, it counts the number of rectification notices received by the store during an evaluation period. According to the preset rules, the violation frequency of the store is mapped to different monitoring levels. As the evaluation period ends, the system will re-evaluate the violation of the store and adjust its monitoring level.
[0075] The monitoring strategy adjustment module sets different monitoring strategies for different monitoring levels, such as increasing or decreasing the monitoring frequency, strengthening or weakening on-site inspections, etc. When in use, the actual monitoring strategy is adjusted according to the monitoring level of the store.
[0076] Through the collaborative work of the above modules, the system can effectively improve urban management efficiency and realize intelligent monitoring and management of shops occupying the road for business operations.
[0077] What is disclosed above is only a preferred embodiment of the present invention, and it certainly cannot be used to limit the scope of rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.
Claims
1. A method for urban governance data fusion analysis, It is characterized in that include: Obtain street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment; Establishing evaluation parameters for urban management and operation behaviors based on historical image data, wherein the evaluation parameters include the area of street occupied by shops, the time period of street occupied, the location of street occupied, and the type of street occupied object; Identify the collected street image data and record abnormal data that deviates from the evaluation parameters; When the number of abnormalities detected by the store reaches a preset value, a rectification notice will be sent to the store manager; Generate a corresponding monitoring level based on the frequency of rectification notices sent to the store during the assessment period; Dynamically adjust monitoring strategies based on monitoring levels.
2. A method for fusion analysis of urban governance data as claimed in claim 1, It is characterized in that The specific steps of acquiring street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment include: Continuously acquiring street images through fixed monitoring equipment, and stitching the images based on the position and shooting parameters of the monitoring equipment to obtain a stitched image; Get the position coordinates of the missing image in the stitched image; At every preset time period, the mobile video acquisition device is notified to acquire the second street image data based on the position coordinates, and the missing image is supplemented to obtain the street image data.
3. A method for fusion analysis of urban governance data as claimed in claim 2, It is characterized in that The specific steps of continuously acquiring street images by fixed monitoring equipment and stitching the images based on the position and shooting parameters of the monitoring equipment to obtain the stitched images include: Obtain images continuously collected by fixed monitoring equipment; Correct the acquired images to reduce lens distortion; Use SIFT algorithm to detect and match feature points in adjacent images. If there is no matching feature point, mark it as a missing image between two adjacent images. The transformation matrix is calculated based on the feature matching results, and the images are stitched into a panorama.
4. A method for fusion analysis of urban governance data as claimed in claim 3, It is characterized in that The specific steps of identifying the collected street image data and recording abnormal data that deviates from the evaluation parameters include: Identifying corresponding parameters in the street image data based on the evaluation parameters to obtain target parameters; Setting a threshold range for each evaluation parameter based on historical image data; The target parameter is compared with the threshold range of each evaluation parameter, and when it exceeds the threshold range, it is determined to be abnormal data.
5. A method for fusion analysis of urban governance data as claimed in claim 4, It is characterized in that The specific steps of identifying corresponding parameters in the street image data based on the evaluation parameters to obtain the target parameters include: Use the YOLO target detection model to detect objects in the image and mark the location of each object; Identify and classify objects occupying the road based on pattern matching; For the detected road-occupying objects, calculate their pixel areas in the image; Converting pixel area into actual area based on camera parameters; Get the location and time when the image was taken.
6. A method for fusion analysis of urban governance data as claimed in claim 5, It is characterized in that The specific steps of sending a rectification notice to the store manager when the number of abnormalities detected by the store accumulates to a preset value include: Each time an anomaly is detected, the target parameters of the anomaly event are recorded in the database; Track the number of abnormal events for each store; When the accumulation of abnormal events in a store reaches a preset threshold, a communication channel is selected and a rectification notice is sent to the store manager.
7. A method for fusion analysis of urban governance data as claimed in claim 6, It is characterized in that The rectification notice contains specific details of the violation and rectification suggestions.
8. A method for fusion analysis of urban governance data as claimed in claim 7, It is characterized in that The specific steps of generating the corresponding monitoring level based on the frequency of rectification notices sent to the shops during the evaluation period include: Define different compliance level ranges based on the number of rectification notices; Match the corresponding compliance level based on the number of rectification notices sent by the target store; The rectification notices are cleared at preset intervals, and the compliance levels of the shops are recalculated and adjusted.
9. A city governance data fusion analysis system, It is characterized in that It includes data acquisition module, evaluation parameter generation module, abnormality judgment module, notification module, monitoring level matching module and monitoring strategy adjustment module; The data acquisition module is used to obtain street image data of shops collected in real time from fixed monitoring equipment and mobile video acquisition equipment; The evaluation parameter generation module is used to establish evaluation parameters of urban management and operation behaviors based on historical image data, wherein the evaluation parameters include the area of street occupied by shops, the time period of street occupied, the location of street occupied and the type of street occupied object; The abnormality judgment module is used to identify the collected street image data and record abnormal data that deviates from the evaluation parameters; The notification module is used to send a rectification notification to the store manager when the number of abnormalities detected by the store accumulates to a preset value; The monitoring level matching module is used to generate a corresponding monitoring level based on the frequency of rectification notices sent to the store during the evaluation period; The monitoring strategy adjustment module is used to dynamically adjust the monitoring strategy according to the monitoring level.
Citation Information
Patent Citations
Superstore operation analysis method and device based on image deep learning algorithm
CN111353436A
Arthroscope-based joint injury identification method and system and storage medium
CN117173144A
Close-shot image splicing method for crop dynamic phenotype extraction
CN117635898A
Mobile vendor management method in urban operation based on image analysis
CN118609036A
Optical element surface pollutant rapid detection method based on visual imaging
CN118691953A
Cited By
Street-along shop treatment task distribution method and system based on multi-modal scheduling
CN119809282A
Multi-modal data fusion method and system based on identification analysis
CN119851081A
Multimodal Data Fusion Method and System Based on Identification Resolution
CN119851081B
Exhibition intelligent security and protection monitoring system and method based on Internet of Things
CN119889010A
Urban comprehensive law enforcement management and control system based on deep learning
CN120612653A
