A planning management and control method and system for urban landscape planning implementation effect

Abstract

The application relates to the technical field of style management and control, and discloses a planning management and control method for urban style planning implementation effect, which comprises the following steps: collecting multiple source style data such as planning drawings and on-site inspection records in real time, extracting core data, completing standardization through noise filtering and format unification, and optionally establishing semantic mapping rules to generate unified labels or correlating text data through AI segmentation of high-definition images; based on the standardized data, the style evaluation and dynamic monitoring capability are endowed through AI, the compliance evaluation standard and the deviation early warning threshold are set, the tasks are decomposed to the planning, city management and housing construction departments in combination with the implementation conditions of the AI monitoring, the control strategies are formulated through the collaborative platform, the departments execute the strategies, the data is collected in real time and input into the AI deviation judgment, the deviation is adjusted and the strategies are synchronized, the AI evaluates the effect and analyzes the effectiveness after the phased tasks, and the AI model parameters, the department task allocation and the control strategies are updated based on the results. The application can realize smooth connection between planning and implementation to improve the control effect.

Description

Technical Field

[0001] This application relates to the technical field of urban landscape management, and in particular to a planning management method and system for improving the implementation effect of urban landscape planning. Background Technology

[0002] In the current process of urban landscape planning implementation and control, it is necessary to integrate multiple landscape data sources such as planning drawings, on-site inspection records, high-definition images, and approval files. However, existing methods lack a unified standardized processing procedure for these data, which is prone to unstable basic data quality due to noise data and outlier interference. Furthermore, it is difficult to identify deviations between the actual landscape and planning requirements in real time, and it relies heavily on manual verification, resulting in low efficiency and a high rate of deviation omission.

[0003] Furthermore, the existing management and control model lacks a clear logic for breaking down complex tasks, has not formed an allocation mechanism that matches the responsibilities of the implementing departments, and has barriers to information sharing between departments, making it difficult to collaboratively formulate management and control strategies. After the completion of phased tasks, the evaluation of implementation effectiveness is mostly limited to the comparison of basic indicators, which cannot effectively optimize the parameters of the management and control model and the task allocation rules, resulting in poor coordination between planning and implementation and difficulty in continuously improving management and control effectiveness.

[0004] As can be seen from the above, how to achieve a smooth connection between planning and implementation to improve the effectiveness of management and control still needs to be addressed. Summary of the Invention

[0005] In order to achieve a smooth connection between planning and implementation and improve the control effect, this application provides a planning control method and system for the implementation effect of urban landscape planning.

[0006] Firstly, this application provides a planning control method for the implementation effect of urban landscape planning, which adopts the following technical solution:

[0007] A planning control method for evaluating the implementation effect of urban landscape planning includes:

[0008] Real-time collection of various landscape data sources from planning drawing databases, on-site inspection records, high-definition image data, and building approval archives; extraction of core landscape control data including building height, facade material, color scheme, and street interface form; noise filtering and outlier removal of the core data; unification of data format and description standards; completion of preliminary standardization processing; and obtaining corresponding standardized landscape data.

[0009] The AI algorithm based on the standardized style data gives the management module AI style evaluation capability and AI dynamic monitoring capability, the AI style evaluation capability is used to analyze the compliance of building and planning indicators and the coordination of style elements, and the AI dynamic monitoring capability is used to identify the deviation of real style and planning requirements in real time, and set the initial parameters including style compliance judgment standard and deviation warning threshold;

[0010] The corresponding city style management and control overall goal is obtained, based on the city style management and control overall goal and the current implementation condition obtained by the AI dynamic monitoring capability, the complex management and control task is divided into style compliance evaluation, deviation problem identification, and rectification scheme making subtasks, which are allocated to the corresponding executive departments of planning, urban management, and housing construction, and each executive department shares the corresponding style evaluation report and planning and real comparison results generated based on the AI style evaluation capability through the collaborative management platform, and cooperatively develops the corresponding collaborative management strategy in combination with the responsibilities of each executive department;

[0011] Each executive department executes the management action according to the collaborative management strategy, wherein the management action includes planning department checking planning indicators, urban management department handling style violation cases, and housing construction department supervising rectification implementation, and real style data is collected in the execution process, and the real style data is input into the management module with AI dynamic monitoring capability, and AI analysis is performed to determine whether there is deviation between real style and planning requirements, and if there is deviation, the management strategy is adjusted and synchronized to the collaborative management platform;

[0012] After completing the phased management and control task, the AI style evaluation capability is called to comprehensively evaluate the implementation effect, the style data and planning indicators before and after rectification in the phased task are compared, the effectiveness of the management and control measures is analyzed, the corresponding effectiveness analysis result is obtained, and the AI style evaluation model parameters, the executive department task allocation mechanism and the collaborative management strategy are updated based on the effectiveness analysis result.

[0013] Optionally, in the process of preliminarily standardizing the style management and control core data, the method further comprises:

[0014] A style attribute semantic mapping rule across data sources is established, and for the same name different meaning or same meaning different name expressions of building height and facade material in different data sources, a unified semantic label is generated based on a preset style attribute dictionary, wherein the semantic label includes absolute height and relative height distinction labels of building height, and main material and auxiliary material classification labels of facade material;

[0015] AI image segmentation processing is performed on the building form features in the high-definition image data, spatial feature data corresponding to building contours and window-to-wall ratios are extracted, and the spatial feature data are associated and matched with corresponding attributes in text data sources such as planning drawings and approval files.

[0016] Optionally, the AI style evaluation capability is realized through a multi-dimensional evaluation index system, including:

[0017] The multi-dimensional evaluation index system is divided into quantitative indicators and qualitative indicators, wherein the quantitative indicators include building height deviation rate, line fitting rate, and color deviation degree, the building height deviation rate is the difference between the actual height and the planning height divided by the planning height, the line fitting rate is the ratio of the length of the building road interface to the length of the road red line, and the color deviation degree is the difference in HSV color value between the actual color and the planning color keynote;

[0018] The qualitative indicators include facade material matching degree and style element coordination, the facade material matching degree is the degree of fit between the actual material and the planning required material, and the style element coordination is the adaptability of the building form to the surrounding historical context and natural environment;

[0019] The AI style evaluation capability also has an index weight dynamic allocation function, which adjusts the weight of each index according to different style area types in the city, and sets the material matching degree weight to 30% in the historical block and the line fitting rate weight to 25% in the commercial and business area.

[0020] Optionally, the collaborative management platform also integrates a department user hierarchical security module and an intelligent contract module, including:

[0021] The department user hierarchical security module is used to set differential permissions according to the responsibility and authority level differences of each executive department, the planning department has the editing permission and all department copy permission of the style evaluation report, the city management department has the viewing permission and supplementary editing permission of the deviation problem record, and the housing construction department has the viewing permission and progress updating permission of the rectification implementation record;

[0022] The intelligent contract module prewrites the responsibility list and task response time limit of each executive department, when the subtasks are decomposed, the intelligent contract automatically matches the executive department according to the task type and generates a task order with time limit, the style compliance evaluation corresponds to the planning department, and the illegal disposal corresponds to the city management department, if the executive department does not feedback the disposal result within the time limit, the intelligent contract automatically triggers the reminder mechanism and records the delay information.

[0023] Optionally, when the AI analysis judges that there is a deviation between the actual style and the planning requirement, the method further includes:

[0024] A deviation grading response mechanism is established, and the deviation is divided into first-level deviation, second-level deviation and third-level deviation according to the influence degree of the deviation on the city style, wherein the first-level deviation is to destroy the core style elements, the first-level deviation includes blocking the waterfront view corridor and tampering with the main structure of the historical building; the second-level deviation is to violate the general control index, the second-level deviation includes that the building height exceeds 5%-10% of the planning and the color deviation degree exceeds 15%; and the third-level deviation is that the local details are inconsistent, and the third-level deviation includes that the facade component is damaged and the street greenery is missing.

[0025] For different levels of deviation, corresponding resources are automatically allocated, the first-level deviation triggers cross-department joint response, and the planning and urban management departments are required to start collaborative disposal within 2 hours; the second-level deviation is formulated by the corresponding executive department within 24 hours; and the third-level deviation is completed by the local executive department within 72 hours.

[0026] Optionally, in the process of calling the AI style evaluation capability to comprehensively evaluate the implementation effect, the method further comprises:

[0027] A public experience evaluation dimension and a long-term impact evaluation dimension are introduced, wherein the public experience evaluation dimension analyzes public feedback data through natural language processing technology, the public feedback data specifically includes government platform messages, questionnaire surveys, extracts key information such as satisfaction and improvement suggestions, and converts them into quantitative scores of 0-100 points;

[0028] The long-term impact evaluation dimension analyzes the positive influence coefficient of style control on regional development by comparing the style data and regional economic data 6-12 months after the completion of the stage task, and the regional economic data specifically includes business activity and tourism visit volume;

[0029] The index comparison results before and after the rectification, the public experience score and the long-term impact coefficient are fused according to the weight of 4:3:3 to generate a comprehensive implementation effect score.

[0030] Optionally, when updating the AI style evaluation model parameters, the task allocation mechanism of the executive department and the collaborative control strategy based on the effectiveness analysis result, a closed-loop iterative optimization mechanism is established, and the method further comprises:

[0031] The sample data corresponding to the index with a comprehensive implementation effect score lower than 80 points is included in the training set of the AI style evaluation model, the sample data includes building cases with excessive height deviation rate and style area data with low public satisfaction, and the model is retrained to improve the evaluation accuracy of such index;

[0032] According to the task completion rate and delay rate of the executive department recorded by the smart contract, the task allocation weight is adjusted, the departments with a completion rate of ≥90% are increased in the same task allocation proportion, and the departments with a delay rate of >15% are reduced in the task amount and pushed to optimization suggestions;

[0033] The improvement suggestions with high frequency in public experience evaluation are converted into supplementary clauses of the collaborative management strategy.

[0034] In a second aspect, the application provides a planning and management system for urban landscape planning implementation effect, which adopts the following technical solution:

[0035] A planning and management system for urban landscape planning implementation effect, comprising:

[0036] A multi-source landscape data acquisition and standardization module, which acquires in real time various types of landscape data sources from a planning drawing database, on-site patrol records, high-definition image data, and building approval archives, extracts landscape management core data including building height, facade material, color keynote, and street interface form, filters noise and removes outliers from the landscape management core data, unifies data format and description standards, completes preliminary standardization processing, and obtains corresponding standardized landscape data;

[0037] An AI landscape management capability and parameter configuration module, which, based on the standardized landscape data, gives the management module AI landscape evaluation capability and AI dynamic monitoring capability through AI algorithms, the AI landscape evaluation capability is used to analyze the compliance of buildings and planning indicators and the coordination of landscape elements, and the AI dynamic monitoring capability is used to identify the deviation of real landscape from planning requirements in real time, and initial parameters including landscape compliance evaluation standards and deviation early warning thresholds are set;

[0038] A management task decomposition and collaborative strategy module, which obtains corresponding urban landscape management overall goals, decomposes complex management tasks into sub-tasks of landscape compliance evaluation, deviation problem identification, and rectification scheme development based on the urban landscape management overall goals and the current implementation status obtained by the AI dynamic monitoring capability, and allocates the sub-tasks to corresponding executive departments of planning, urban management, and housing construction, and each executive department shares corresponding landscape evaluation reports and planning and real comparison results generated based on the AI landscape evaluation capability through a collaborative management platform, and collaboratively develops corresponding collaborative management strategies in combination with the responsibilities of each executive department;

[0039] A management execution and real-time deviation adjustment module, in which each executive department executes management actions according to the collaborative management strategies, wherein the management actions include planning department verification of planning indicators, urban management department handling of landscape violations, and housing construction department supervision of rectification implementation, real landscape data are collected in real time during the execution process, input into the management module with AI dynamic monitoring capability, and AI analysis is performed to determine whether there is a deviation between real landscape and planning requirements, and if there is a deviation, the management strategy is adjusted and synchronized to the collaborative management platform;

[0040] The implementation effect evaluation and control optimization module is used to comprehensively evaluate the implementation effect after completing the phased control task, compare the landscape data and planning indicators before and after rectification in the phased task, analyze the effectiveness of the control measures, obtain the corresponding effectiveness analysis result, update the AI landscape evaluation model parameters, the task allocation mechanism of the execution department, and the collaborative control strategy based on the effectiveness analysis result.

[0041] In a third aspect, the present application provides a planning and control system for urban landscape planning implementation effect, which adopts the following technical solution:

[0042] A planning and control system for urban landscape planning implementation effect, comprising a processor, wherein the processor runs a program of the planning and control method for urban landscape planning implementation effect according to any one of the above.

[0043] In a fourth aspect, the present application provides a storage medium, which adopts the following technical solution:

[0044] A storage medium storing a program of the planning and control method for urban landscape planning implementation effect according to any one of the above.

[0045] In summary, the present application includes at least one of the following beneficial technical effects:

[0046] Through multi-source landscape data standardization processing (unified format, eliminating abnormal values, establishing semantic mapping), the consistency of basic data such as planning drawings and approval archives and implementation data such as field inspection and high-definition images is ensured, and then relying on the AI landscape evaluation and dynamic monitoring capability, the planning indicators are compared with the real landscape in real time, the implementation deviation is accurately identified, and the data and technical foundation for the connection of planning and implementation is laid. At the same time, through task decomposition (compliance evaluation, deviation identification, rectification formulation), the responsibilities of planning, urban management and housing departments are adapted, combined with the collaborative platform of hierarchical security permissions and intelligent contract time limit control, each department can efficiently share planning basis and synchronize implementation progress, avoiding the disconnection of planning and implementation caused by information barriers, and preliminarily improving the connection efficiency of planning and implementation.

[0047] Through the deviation hierarchical response mechanism (one-level deviation cross-department 2-hour collaboration, two-level 24-hour scheme formulation, and three-level 72-hour rectification completion), the deviation problems in implementation are quickly disposed, and it is ensured that the implementation always meets the planning requirements. After the phased task, multi-dimensional evaluation (compliance + public experience + long-term impact) and closed-loop iterative optimization (updating AI model, adjusting task allocation, and supplementing control strategy) can continuously correct the connection details of planning and implementation, so that the control method can continuously adapt to the planning goal, and finally realize the smooth connection of the whole process of planning and implementation, and significantly improve the accuracy, timeliness and long-term effectiveness of urban landscape control. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 is a flow chart of a planning management method for urban landscape planning implementation effect according to an exemplary embodiment.

[0049] Figure 2 is a structural block diagram of a planning management system for urban landscape planning implementation effect according to an exemplary embodiment. DETAILED DESCRIPTION

[0050] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.

[0051] In the description of the present specification, the description of the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the described embodiments or examples are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0052] The embodiment of the present application discloses a planning management method for urban landscape planning implementation effect, referring to Figure 1 , comprising:

[0053] S100, real-time collection is performed on various landscape data sources from planning drawing databases, on-site patrol records, high-definition image data, and building approval archives, core data of landscape management and control including building height, facade material, color keynote, and street interface form are extracted, noise filtering and abnormal value elimination are performed on the core data of landscape management and control, unified data format and description standard are completed, preliminary standardized processing is completed, and corresponding standardized landscape data is obtained.

[0054] Among them, S100 is the "data basic construction link" of urban landscape management and control, which needs to be executed in four steps of "multi-source data collection-core data extraction-data cleaning-standardization processing" closed loop, and the specific process is as follows:

[0055] Step 1, Real-time collection of multi-source style data. For the planning drawing database, through the interface of the city planning management system, real-time synchronization of vector data (including building layout, elevation marking, etc.) such as control planning drawings and urban design schemes; field patrol records are collected through the grid member mobile APP, after the patrol personnel take photos of the building scene and record the material / color information on the spot, the system automatically uploads to the background, supporting the binding of positioning information (such as latitude and longitude) and records; high-definition video data is obtained through "periodic aerial photography by unmanned aerial vehicle + real-time snapshot by fixed monitoring point camera", aerial photography covers key style areas (such as historical blocks and waterfronts), the cycle can be set to once a week, and the fixed camera collects street images at a frequency of 1 frame per hour; building approval files are obtained by connecting to the administrative approval system, and the building height, facade material, color requirements and other text and drawing data at the time of project approval are obtained in real time, ensuring that the data source covers "planning source - implementation process - approval basis".

[0056] Step 2, Core data oriented extraction. For the multi-source data collected in the first step, four types of core data are extracted using a combination of "rule matching + AI recognition": building height is extracted from the elevation marking of the planning drawing, the height approval value of the approval file, and the field patrol record, and the legal height of the approval file is taken as the reference; facade material is analyzed by AI image recognition technology (such as convolutional neural network model) to analyze field patrol photos and high-definition videos, combined with the material design specification of the planning drawing, to determine the specific types of "stone / glass curtain wall / paint"; color keynote is extracted from the HSV color space value of the main building in the high-definition video through image recognition, and then compared with the color interval (such as "warm yellow color system") required by the planning to determine the core color information; street interface form is extracted from aerial video to extract key features such as building street continuity (whether the building along the street is interrupted), setback distance (distance between building exterior wall and road red line), and interface transparency (proportion of doors and windows to wall surface), to form structured data.

[0057] Step 3, noise filtering and outlier removal. For the "invalid interference data" in the collected data, first perform noise filtering: for high-definition video, use image sharpness evaluation algorithm (such as Laplacian operator gradient value) to screen out blurred, overexposed or severely obstructed images; for on-site patrol records, delete duplicate records through data deduplication algorithm (such as based on "time + location + content" triple verification); for text data (such as approval files), correct errors and format chaos through grammar correction algorithm. Then perform outlier removal: for building height, remove "label 1000 meters" and "negative height" and other obviously unreasonable data in combination with local planning limits (such as residential maximum 80 meters); for color tone, remove abnormal color values caused by light and shadow interference (such as heavy rain, strong light at night), and replace them with color data from normal images in the same area and period; for street interface form, remove misjudged data caused by shooting angle deviation (such as excessive tilting of drones), and correct them by referring to the average setback value of surrounding buildings.

[0058] Step 4, unified data format and description standard. In terms of format, convert data from different sources to a common format: convert planning drawings from CAD format to GIS vector format (such as SHP format) for easy spatial coordinate association; convert high-definition images from RAW format to JPEG format to reduce storage pressure; convert text data (such as patrol records, approval information) from Excel, PDF format to JSON format for easy system parsing. In terms of description standard, establish a unified "style data dictionary": unify the description of facade materials (such as "granite" instead of "stone" and "firestone"); use "absolute elevation" (based on Yellow Sea elevation) for building height to avoid confusion caused by "relative height"; describe color tone in the format of "main color (proportion ≥ 60%) + auxiliary color (proportion 20% ~ 40%) " (such as "main color beige + auxiliary color light gray"); describe street interface form with quantitative indicators such as "continuous segment length / broken segment length" and "average setback distance" to ensure that data of the same type from different sources can be compared and associated. The final output of standardized style data needs to be associated with a unique spatial identifier (such as plot number) and a timestamp (collection / update time) for easy subsequent tracing and dynamic updating.

[0059] S100 builds a "high-quality data base" for the entire style management plan through "full collection-precise extraction-purification treatment-unified standard": on the one hand, it solves the problem of "incompatible planning drawings and field data formats, chaotic description of core information such as material / height" in traditional management, ensuring that AI style evaluation and dynamic monitoring in S200 can be based on "unified standard" data for analysis, avoiding AI recognition bias caused by disordered data; on the other hand, the standardized style data contains "planning source data (drawings, approval)" and "implementation process data (patrol, image)", providing "planning benchmark" and "implementation status" comparable data for subsequent S300 task decomposition and S400 deviation identification, eliminating the "planning and implementation disconnection caused by inconsistent data" from the root, and providing reliable data support for subsequent management links, laying the foundation for the accuracy and synergy of the entire plan.

[0060] S200, based on standardized style data, uses AI algorithms to give the management module AI style evaluation and AI dynamic monitoring capabilities. AI style evaluation is used to analyze the compliance of buildings and planning indicators and the coordination of style elements. AI dynamic monitoring is used to identify real-time deviations between real style and planning requirements. Initial parameters including style compliance evaluation criteria and deviation warning thresholds are set.

[0061] S200, as the "AI capability construction and benchmark setting link" of urban style management, needs to follow the four-step process of "standardized data AI adaptation preprocessing-AI style evaluation capability construction-AI dynamic monitoring capability construction-initial parameter setting". The specific operation is as follows:

[0062] Step 1, AI adaptation preprocessing of standardized style data. For the standardized style data output by S100, "feature engineering" processing is performed to adapt to the input requirements of AI algorithms: for text data (such as "building height ≤ 80 meters" and "facade material is granite" in planning indicators), use natural language processing (NLP) entity recognition algorithm to extract key numerical values and attribute labels, convert "≤ 80 meters" to numerical threshold "80", and "granite" to material code "001"; for image data (such as high-definition images and on-site patrol photos), use image enhancement algorithms (such as histogram equalization and noise reduction filtering) to optimize image quality, and then use feature extraction networks to extract visual feature vectors such as building contours, color histograms, and material textures; for structured data (such as building height measured values and setback distances), perform normalization (scale numerical values to the [0, 1] interval) to avoid affecting AI model accuracy due to numerical magnitude differences. At the same time, the preprocessed data is divided into training set (used for model training), validation set (used for parameter tuning), and test set (used for effect verification) in a ratio of "7:2:1", providing adaptive data for subsequent AI capability construction.

[0063] Step 2, AI Landscape Assessment Capability Construction. This capability is achieved through "double module fusion", corresponding to "planning index compliance analysis" and "landscape element coordination analysis":

[0064] Planning Index Compliance Analysis Module: Gradient Boosting Tree (XGBoost) algorithm is used, with pre-processed standardized data as input (such as actual building height, material code, setback distance), and planning indicators as labels (such as "height compliance = 1 / non-compliance = 0" "material compliance = 1 / non-compliance = 0"), to train a classification model. The model outputs "compliance score" (0-100 points), for example, the actual building height is 75 meters (planning ≤80 meters), and the material is granite (in line with planning requirements), the compliance score is 90 points; if the actual height is 85 meters (5 meters higher than the planning), the score will drop to 60 points, and the specific deviation item ("height exceeds planning by 5 meters") will be output.

[0065] Landscape Element Coordination Analysis Module: Convolutional Neural Network (CNN) + Transformer fusion model is used, input "target building image + surrounding 300 meters range building image + historical landscape sample library", through CNN to extract building style features (such as pitched roof / flat roof, window hole ratio), through Transformer to capture the correlation of the surrounding environment (such as color echo degree, volume level), output "coordination score" (0-100 points). For example, the target building is modern glass curtain wall style, and the surrounding buildings are all traditional pitched roof buildings, the coordination score is 40 points, and the optimization suggestion is generated at the same time ("suggest adding local pitched roof elements to improve coordination").

[0066] Two-module results weighted fusion (such as compliance accounting for 60%, coordination accounting for 40%), forming the final AI landscape assessment results.

[0067] Step 3, AI dynamic monitoring capability construction. Focus on "real-time identification of real landscape and planning requirement deviation", using "real-time data access → dynamic comparison → deviation warning" real-time processing logic:

[0068] Real-time data access: build edge computing nodes, connect real landscape data collected by S400 executive departments in real time (such as on-site inspection photos, construction progress data, real-time video of fixed cameras), data transmission uses low-latency protocol (such as MQTT), to ensure that the delay from data collection to AI model access is ≤3 seconds.

[0069] Dynamic comparison algorithm: adopt the dual algorithm of "feature matching + anomaly detection" in coordination. For structured data (such as real-time collected building height), directly compare with planning index threshold, and mark deviation if out of range; for unstructured data (such as real-time image), use "frame difference method" to extract the change area of building form and color in the image, and then perform CNN feature matching with the planning image features (such as planning facade color, contour) in the standardized data, and if the similarity is less than 85%, it is determined as deviation.

[0070] Real-time early warning trigger: preset the automatic logic of "deviation identification → early warning generation" in the AI model, once the deviation is identified, automatically generate early warning information containing "deviation location (latitude and longitude), deviation type (height / material / color), deviation degree (more than 5% of planning)", and store in the control module for subsequent parameter matching and push.

[0071] Step 4, initial parameter setting. Around the two core parameters of "landscape compliance evaluation standard" and "deviation early warning threshold", combined with urban landscape control needs and industry standards:

[0072] Landscape compliance evaluation standard: reference "Urban Landscape Management Regulations" and local planning requirements, quantify the compliance range of various indicators, for example "building height deviation ≤ 5% is compliant" "facade material and planning requirement coincidence degree ≥ 90% is compliant" "color deviation degree (HSV value difference) ≤ 15 is compliant" "street interface line fitting rate ≥ 70% is compliant", the standard needs to be associated with specific landscape area type (such as historical street compliance standard is stricter than ordinary residential area).

[0073] Deviation early warning threshold: based on "compliance standard + risk level" setting, for example, the compliance standard is "height deviation ≤ 5%", the early warning threshold is set to "height deviation ≥ 8%" (trigger early warning, reserve rectification time); the color compliance standard is "HSV difference ≤ 15", the early warning threshold is set to "HSV difference ≥ 20"; at the same time, set the early warning priority for different deviation types (such as height deviation priority is higher than color deviation). All parameters are stored in the parameter configuration library of the control module, supporting subsequent dynamic update of effectiveness analysis results based on S500.

[0074] By "AI capability construction + initial parameter setting", an "intelligent core engine" is created for the entire style management plan, playing a key role in connecting the past and the future: on the one hand, it can accept standardized style data output by S100, and convert "static data" into "dynamic analysis capability" through AI algorithms, solving the pain points of "low efficiency of manual evaluation and lack of real-time monitoring" in traditional management and control - AI style evaluation capability can quickly quantify building compliance and coordination, replacing traditional manual scoring; AI dynamic monitoring capability realizes real-time identification of deviations, avoiding the lag of manual patrol. On the other hand, the setting of initial parameters provides a "judgment benchmark" for subsequent management and control: S300 task decomposition can clearly identify evaluation priorities for each department based on compliance standards, and S400 deviation adjustment can refer to the warning threshold to determine whether to start strategy optimization, ensuring that "planning indicators have AI analysis support and implementation deviations have warning standards to rely on" from a technical perspective, and completely breaking through the difficulties of "no quantification in evaluation, no real-time monitoring, and no benchmark in decision-making" in traditional management and control, laying the technical core for the precision and intelligent management and control of the entire plan.

[0075] S300, obtain the overall target of urban style management and control, based on the overall target of urban style management and control and the current implementation status obtained by AI dynamic monitoring capability, decompose the complex management and control task into sub-tasks of style compliance evaluation, deviation problem identification, and rectification scheme formulation, and assign them to the corresponding executive departments of planning, urban management, and housing construction. Each executive department shares the corresponding style evaluation report and planning and reality comparison results generated based on AI style evaluation capability through the collaborative management platform, and collaboratively formulates corresponding collaborative management strategies in combination with the responsibilities of each executive department.

[0076] Among them, S300 needs to be executed in a six-step closed loop of "overall target acquisition and analysis → target-current situation comparison → complex task decomposition → department task allocation → platform data sharing → cross-department strategy formulation", and the specific process is as follows:

[0077] Step 1, obtain and analyze the overall target of urban style management and control. By interfacing with official documents such as urban master plan files, special style management and control plans (such as historical block protection plans, waterfront area style design guidelines), and annual style improvement plans, extract overall targets such as "protecting historical building texture", "optimizing waterfront skyline form", and "improving commercial street interface continuity"; then decompose the overall target into quantifiable and verifiable sub-control indicators: for example, the "historical block protection" target is decomposed into "core area building height ≤ 9 meters", "traditional facade materials (blue bricks, black tiles) proportion ≥ 90%", and "street interface break length ≤ 5 meters"; the "waterfront skyline optimization" target is decomposed into "building height within 50 meters of water ≤ 40 meters", "view corridor width ≥ 80 meters", and "lakefront interface permeability ≥ 60%", ensuring that the target is transformed from "macro direction" to "specific requirements that can be implemented".

[0078] Step 2, comparison analysis of target and current implementation status. Call the current implementation status data output by the AI dynamic monitoring capability in S200, including "compliance rate of each style area (such as historical block building height compliance rate 82%, waterfront area sight corridor compliance rate 75%)", "deviation distribution data (such as 3 insufficient building setbacks in commercial street, 5 facade materials not in line with the requirements)", "implementation progress data (such as the proportion of completed rectification of deviation problems is 60%)"; compare the subdivided control indicators in the first step with the current implementation status one by one to form a "target-current status gap list", for example, "historical block 'facade traditional material proportion ≥ 90%' vs current 85% (gap 5%)", "waterfront area 'building height near water ≤ 40 meters' vs 2 buildings measured 45 meters and 48 meters (gap 5-8 meters)", and clearly define the core problems that need to be prioritized in control.

[0079] Step 3, decomposition of complex control tasks. Based on the "target-current status gap list", the complex control task of "narrowing the target gap and promoting style compliance" is decomposed into three types of subtasks according to the "logic chain of 'evaluating current compliance → locating specific deviations → developing rectification plans'".

[0080] Style compliance evaluation subtask: for buildings in each style area, check whether they meet the subdivided control indicators, and output "compliant building list (including compliance score) + non-compliant building details (annotating specific non-compliant indicators)".

[0081] Deviation problem identification subtask: for non-compliant buildings, further locate the deviation type (such as height exceeding regulations, material not in line with requirements, insufficient setback), deviation degree (such as 5% height exceeding regulations, 30% area of material not in line with requirements), spatial location (accurate to plot number or house number), and responsible subject (such as construction unit, operation and maintenance unit).

[0082] Rectification plan development subtask: for identified deviation problems, develop a feasible plan for "technical rectification path (such as structural adjustment scheme to reduce building height, construction process to replace facade material) + time node (such as submitting plan within 15 days, completing rectification within 30 days) + guarantee measures (such as rectification fund source, supervision unit responsibilities)".

[0083] Step 4, matching and distribution of subtasks and executive departments. According to the core responsibilities and professional capabilities of planning, urban management, and housing departments, establish "subtask-department" precise matching rules:

[0084] Style compliance evaluation subtask is assigned to the planning department: because the planning department masters the original planning indicators and is familiar with the development logic of subdivided control targets, it can accurately judge whether the building meets the planning intent, and can also supplement evaluation opinions combined with urban design requirements;

[0085] The deviation problem identification subtask is assigned to the urban management department: because the urban management department has the ability to conduct on-site patrol and law enforcement verification, it can verify the deviation problems identified by AI dynamic monitoring in the field and eliminate "image misjudgment" (such as interface discontinuity caused by temporary construction fences) to ensure the accuracy of the deviation information;

[0086] The rectification scheme development subtask is assigned to the housing department: because the housing department is familiar with the construction technical standards and the engineering rectification process, it can develop a technically feasible and time-controllable rectification scheme based on factors such as structural safety and construction period.

[0087] Step 5, data sharing preparation in the collaborative management platform. In the collaborative management platform, a "landscape data sharing zone" is built, and the core data generated by the AI landscape evaluation capability in S200 is uploaded and synchronized: including "landscape evaluation report (including compliance score of each building, coordination analysis conclusion)" "planning and reality comparison results (such as overlay comparison chart of planning drawings and field video, deviation position marking map)" "target-status gap list"; At the same time, set data access permissions (reference to the previous department hierarchical security mechanism): planning department can edit evaluation report, urban management department can supplement deviation verification record, housing department can view evaluation report and comparison results, to ensure that each department works based on "the same set of data", avoid information asymmetry.

[0088] Step 6, cross-departmental collaboration to develop collaborative management strategies. Each executing department conducts online consultation (or offline joint meeting) through the collaborative management platform: the planning department determines "which deviations need to be rectified first (such as height exceeding regulations affecting core landscape)" and "the planning bottom line that needs to be met (such as historical street material replacement needs to be recorded by the cultural relics department)"; The urban management department proposes "violation disposal process (such as the time limit for issuing rectification notice to the responsible party)" and "on-site verification frequency (such as weekly patrol during rectification)"; The housing department determines "rectification technical standards (such as environmental protection requirements for facade material replacement)" and "rectification acceptance process (such as submitting a third-party detection report after rectification)"; Finally, integrate the opinions of each department to form a unified collaborative management strategy that includes "task division, time node, technical standard, acceptance requirement", and synchronize it to the collaborative management platform for each department to execute.

[0089] Through the whole process design of "target decomposition - status comparison - task allocation - collaborative strategy", a "bridge for planning goals to land" is built for the whole landscape management plan:

[0090] On the one hand, the AI dynamic monitoring data of S200 converts the "macro target" into "department executable sub-tasks", solving the problem of "task boundary ambiguity, department shirking and wrangling" in traditional management and control - by clarifying the division of labor of "planning evaluation compliance, city management recognition bias, and housing construction scheme determination", the responsibilities of each department are clear and coordinated in an orderly manner;

[0091] On the other hand, through unified data sharing of the collaborative management platform, it ensures that each department formulates strategies based on "the same set of planning basis and the same set of current data", avoiding the problem of "planning requirements and implementation measures being disconnected" due to information fragmentation (such as planning requirements protecting traditional materials, while the housing department's rectification scheme misusing modern materials); The collaborative management strategy formed ultimately not only fits the overall goal of urban style management and control, but also adapts to the execution capabilities of each department, providing a "clear roadmap" for the implementation of S400 management actions, ensuring smooth connection between planning and implementation from the task coordination level, and laying a foundation for improving management and control effect in the future.

[0092] S400, each executing department executes management actions according to the collaborative management strategy, wherein the management actions include planning department checking planning indicators, city management department handling style violations, and housing department supervising rectification implementation. In the execution process, real style data is collected in real time, and the real style data is input into the management and control module with AI dynamic monitoring capability. The AI analyzes and judges whether there is a deviation between the real style and the planning requirements. If there is a deviation, the management and control strategy is adjusted and synchronized to the collaborative management platform.

[0093] Among them, S400 needs to execute in a five-step closed loop according to "department action details clear → real style data collected in real time → AI deviation analysis and judgment → deviation handling and strategy adjustment → adjusted strategy synchronization", the specific process is as follows:

[0094] Step 1, each executing department clarifies the management action details. Based on the collaborative management strategy formulated by S300, each department further refines the operation process and execution standard:

[0095] Planning department: Focus on "planning indicator checking", clarify the checking range (such as key style area built / in construction project), checking indicators (building height, setback distance, facade material, color keynote, etc. core data extracted by S100), checking tools (total station height measurement, GIS system setback comparison, material sample library verification facade), and determine the checking frequency (new project every construction stage 1 time, stock project every quarter 1 time), need to output 《planning indicator checking report》, mark "conformity / non-conformity" and specific deviation value (such as height exceeds planning 6 meters);

[0096] Urban management department: Focus on "disposal of style violations", clearly define violation identification standards (based on S200 compliance evaluation standards), disposal process (on-site evidence collection → issue of "violation rectification notice" → track rectification progress → rectification acceptance), disposal measures (such as illegal construction order to demolish, facade material inconsistent with replacement within a certain period), and configure mobile enforcement APP to support on-site upload of violation photos, locate violation location, and automatically associate corresponding planning requirements;

[0097] Housing department: Focus on "rectification implementation supervision", clearly define supervision objects (violation project construction unit, rectification construction unit), supervision content (rectification plan implementation progress, construction quality, safety specification), supervision method (weekly on-site inspection, monthly progress report review), and need to establish "rectification progress account" to record "planned completion time / actual completion time / cause of non-compliance".

[0098] Step 2, real-time collection of real style data. Departments collect data in real time through "automatic tool collection + manual auxiliary input" during the implementation of management activities:

[0099] Planning department: Use total station and laser range finder to collect actual building height, use high-definition camera to take facade material photos, and automatically synchronize data to collaborative management platform, associate corresponding planning index data (such as planning height, required material) of the plot;

[0100] Urban management department: Use APP to take photos / videos of violation scene, input violation type (such as height exceeding standard, material not meeting requirements), and violation degree (such as partial violation / overall violation), and upload data to platform in real time to trigger location labeling and timestamp recording;

[0101] Housing department: Use mobile device to fill out "rectification progress table" (such as "30% of material replacement completed" "structure adjustment pending acceptance"), take rectification site photos, and automatically associate and update data with "rectification progress account";

[0102] All data is transmitted in real time through 5G network or local area network, ensuring that the delay from collection to entering the control module is ≤10 minutes, meeting the "dynamic monitoring" requirement.

[0103] Step 3, AI dynamic monitoring module analyzes deviation. Input real-time collected real style data into AI dynamic monitoring module constructed by S200, and operate according to "data classification processing → deviation comparison analysis → deviation level determination" process:

[0104] Data classification processing: structured data (such as height measurement values, improvement progress percentage) directly enters the numerical comparison module; unstructured data (such as facade photos, violation videos) extracts features (such as material texture, building outline) through image recognition algorithm, and converts them into comparable structured feature values (such as material code, outline size);

[0105] Deviation comparison analysis: the module automatically retrieves the planning requirements (S100 standardized data planning height, material, color, etc.) of the corresponding land and the deviation warning threshold set in S200, and performs numerical comparison (such as measured height 86 meters vs. planned 80 meters, deviation 6 meters) and feature matching (such as measured material texture vs. planned material sample library, matching degree calculation);

[0106] Deviation level determination: according to the deviation influence degree standard set in S200, the deviation is divided into three levels - first-level deviation (such as height exceeding planning > 10%, damaging main body of historical building), second-level deviation (such as height exceeding planning 5% ~ 10%, material matching degree 70% ~ 90%), third-level deviation (such as color deviation degree 15% ~ 20%, improvement progress lag ≤ 7 days), and generates a real-time deviation analysis report, marking the deviation position, type, level and associated executive department.

[0107] Step 4, deviation treatment and control strategy adjustment. According to the deviation level, start the corresponding treatment mechanism and adjust the control strategy:

[0108] Third-level deviation (minor): adjust the strategy internally by the corresponding executive department, such as the housing and construction department increasing the frequency of on-site supervision by 1 time for improvement progress lag of 3 days, without the need for cross-departmental coordination;

[0109] Second-level deviation (general): led by the corresponding executive department, jointly review deviation data with the planning department (such as city management department finding material inconsistency, planning department reviewing planning requirements), adjust strategy content (such as extending the rectification period from 15 days to 20 days due to the need to customize special materials);

[0110] First-level deviation (serious): start cross-departmental joint treatment, hold a temporary coordination meeting by the planning, city management and housing departments, adjust core strategy (such as a project height exceeding planning by 15%, need to re-approve design scheme, adjust strategy to "suspend construction + submit new scheme within 7 days"), and clarify the adjusted responsibility division and time node.

[0111] Step 5, adjust the policy to the collaborative management platform. All adjusted control policies (including internal adjustment of departments, cross-departmental coordination adjustment) need to be uploaded through the "policy update module" of the collaborative management platform, automatically synchronized to the homepage of the relevant executing department's account, and trigger a reminder (such as APP push, SMS reminder). The platform records the "original content / adjusted content / adjustment reason / adjustment time / approver" of the policy adjustment, forming a traceable "policy adjustment ledger", ensuring that each department executes the latest policy and avoiding "mixing of new and old policies" leading to execution deviation.

[0112] Through the whole process design of "refining execution, real-time data collection, AI deviation judgment, and dynamic strategy adjustment", it becomes the "core execution layer" of the entire style control plan "strategy landing", playing three key roles:

[0113] First, it solves the problem of "strategy idling and disordered execution" in traditional control. By clearly defining the action details of each department, the collaborative strategy of S300 is transformed from "text requirements" to "operable specific actions", ensuring that planning index verification, violation disposal, and rectification supervision have rules to follow.

[0114] Second, it breaks the dilemma of "implementation lagging behind planning". Real-time collection of real data and AI dynamic deviation analysis can detect deviations between implementation and planning in the first time, avoiding the expansion of deviation caused by traditional manual verification "long cycle and many missed judgments".

[0115] Third, it realizes "strategy flexibility adaptation". By adjusting the strategy according to the deviation level, it ensures that the core planning requirements are not relaxed (such as strict cross-departmental disposal for first-level deviation), and can flexibly respond to special situations in implementation (such as reasonable extension of rectification period for second-level deviation), ultimately keeping the implementation process around the planning target, providing "dynamic execution data" for S500 effect evaluation, and ensuring real-time connection between planning and implementation.

[0116] S500, after completing the phased control task, calls the AI style evaluation capability to comprehensively evaluate the implementation effect, compares the style data before and after rectification in the phased task with the planning index, analyzes the effectiveness of the control measures, obtains the corresponding effectiveness analysis result, and updates the AI style evaluation model parameters, the execution department task allocation mechanism, and the collaborative control strategy based on the effectiveness analysis result.

[0117] Among them, S500 needs to perform a four-step closed loop of "evaluation preparation, AI comprehensive evaluation, effectiveness analysis, and model and mechanism update", the specific process is as follows:

[0118] Step 1, Preparation for periodic assessment and data collection. First, clearly define the assessment scope, which should be consistent with the scope of the periodic management tasks of S300-S400 (e.g., "3 square kilometers of core area of historical blocks" or "1.5 kilometers of waterfront section"). Avoid disconnection between the assessment scope and the task scope. Second, collect three types of core data: 1) "pre-renovation appearance data" (real appearance data collected before S400 execution, such as illegal building height, non-compliant material proportion), 2) "post-renovation appearance data" (latest data collected after S400 execution, such as renovated building height, replaced facade material), and 3) "reference data" (including S100 standardized planning indicators, such as building height limit, material requirements; S400 execution process management measures records, such as planning department verification frequency, city management department illegal disposal times, housing department supervision scheme). All data should be associated with a unique "phase identifier" (e.g., "2024Q2 historical block renovation phase") and spatial identifier (e.g., plot number), ensuring traceability and comparability of data.

[0119] Step 2, Call AI appearance assessment capability to carry out comprehensive assessment. Based on the collected data, start the AI appearance assessment capability constructed in S200, and execute according to the three-layer logic of "compliance assessment → coordination assessment → multi-dimensional supplementary assessment":

[0120] Compliance assessment: AI model automatically compares "pre / post-renovation appearance data" with "planning indicators" to calculate the improvement rate of core indicators - such as building height compliance rate (75% before renovation → 92% after renovation), facade material compliance rate (68% before renovation → 89% after renovation), color tone consistency rate (72% before renovation → 90% after renovation), output "compliance improvement report", and mark the indicators that still do not meet the standards (such as the setback distance of a historical building still exceeding 2 meters);

[0121] Coordination assessment: AI model analyzes the adaptability of the renovated building and the surrounding appearance through image comparison technology - such as "material echo degree" "volume level" "color coordination" of the renovated building and adjacent traditional buildings in the historical block, outputs coordination score (e.g., 65 points before renovation → 88 points after renovation), and identifies remaining uncoordinated problems (such as a newly built building roof model not matching the traditional pitched roof);

[0122] Multi-dimensional supplementary assessment: Introduce public experience and long-term impact dimensions - analyze public feedback during the periodic task period (such as government messages, questionnaire surveys) through natural language processing technology, generate public satisfaction score (0-100 points, such as from 70 points to 85 points); compare regional economic data (such as tourism visit volume, business activity) 6-12 months after the completion of the periodic task, analyze the positive influence coefficient of appearance management on regional development (such as 20% increase in visit volume corresponds to influence coefficient 0.8), and finally form "comprehensive assessment report".

[0123] Step 3, Analysis of Effectiveness of Control Measures: Based on the "Comprehensive Evaluation Report" and the records of control measures in S400, establish a "measure-effect" correspondence, and analyze the actual effect of various control measures:

[0124] Effective measure identification: For example, the "monthly index check" of the planning department corresponds to a 17% increase in compliance rate of building height, the "illegal disposal within 3 days" of the city management department corresponds to a 21% increase in rectification completion rate, and the "weekly progress supervision" of the housing and construction department corresponds to a 10-day reduction in rectification period, which clearly defines the "effective logic" of such measures (e.g., high-frequency check can timely discover deviations);

[0125] Ineffective / ineffective measure analysis: For example, a rectification plan is delayed due to "long material procurement cycle", indicating that the housing and construction department's "no advance intervention in material preparation" supervision measure has loopholes; a violation problem repeatedly occurs due to the "insufficient follow-up review frequency" of the city management department, which clearly defines the optimization direction of such measures;

[0126] Root cause analysis: Distinguish the root causes of "technical level" (e.g., AI evaluation model has insufficient material identification accuracy leading to misjudgment), "process level" (e.g., long department coordination approval time), and "execution level" (e.g., incomplete rectification by construction units), and form an "effectiveness analysis result report" to mark "measures that need to be optimized" and "root problems that need to be solved".

[0127] Step 4, Iterative update based on effectiveness analysis results: Based on the "effectiveness analysis result report", update AI model parameters, department task allocation mechanism, and collaborative control strategy:

[0128] AI style evaluation model parameter update: Include sample data corresponding to indicators in the "comprehensive evaluation report" that still have deviations (e.g., building cases with excessive back-off distance, image samples with low coordination score) into the model training set, divide the training / validation / test set in a 7:2:1 ratio, retrain the model, and improve the identification accuracy of such problems (e.g., material identification accuracy from 88% to 95%);

[0129] Update of execution department task allocation mechanism: According to the effectiveness of measures of each department, for "effective measures" corresponding departments (e.g., planning department index check, city management department illegal disposal), increase the allocation proportion of similar tasks (e.g., from 40% to 55%); for "ineffective measures" corresponding departments (e.g., housing and construction department material preparation supervision), reduce the task amount and add "optimization requirements" (e.g., add "material procurement progress advance check" sub-task); simultaneously refer to the task completion rate in S400 (e.g., departments with completion rate ≥90% are preferentially allocated core tasks);

[0130] Cooperative management strategy update: Convert the vulnerabilities found in the effectiveness analysis and the high-frequency needs of public feedback into strategy supplements, such as "long material procurement cycle", add "housing department needs to check and rectify material preparation 15 days in advance" in the strategy; for public feedback "insufficient accessibility of waterfront promenade", supplement "city management department needs to check promenade opening every month" in the strategy, the updated strategy is synchronized to the collaborative management platform, covering the original version and triggering department reminders.

[0131] Through the closed-loop design of "comprehensive assessment - effectiveness analysis - iterative update", it becomes the "core driver of continuous optimization" of the entire style control scheme, playing three key roles: First, it solves the pain points of traditional control "heavy execution, light assessment", through multi-dimensional assessment (compliance + coordination + public + economy), it fully tests whether the phased tasks achieve the planning goals, avoiding "no feedback after implementation, long-term deviation"; Second, it breaks the "measures and effects are out of touch" dilemma, through "measure-effect" corresponding analysis, it accurately locates the root cause of inefficient measures, providing a clear direction for subsequent optimization, avoiding "blind adjustment of strategy"; Third, it realizes the dynamic adaptation of the control system, through updating the AI model to improve recognition accuracy, optimizing task allocation to improve department efficiency, supplementing strategy clauses to cover vulnerabilities, so that the entire control scheme can continuously adapt to changes in planning goals, adjustments in implementation environment, and upgrades in public demand, ultimately forming a complete closed loop of "planning-implementation-evaluation-optimization", ensuring the long-term smooth connection of planning and implementation, and promoting the continuous improvement of urban style control effect.

[0132] Taking "historical block core area style control" as an example, first, through S100, build a solid data connection foundation: collect the planning drawings of the block (clearly indicate that the building height is ≤9 meters and the facade material is blue brick), historical building approval files, grid employee on-site patrol records (mark 3 suspected high-rise construction), and unmanned aerial vehicle aerial images, after noise filtering (remove blurred images) and standardization processing (unify building height to Huanghai elevation, material description to "blue brick / non-blue brick"), form a standardized data set of "planning source data + implementation process data", avoiding subsequent implementation due to different data standards and planning disconnection; then through S200, build AI connection capability, AI style evaluation model can quantify "height compliance rate" and "material matching degree", dynamic monitoring model can capture construction deviations in real time, and set the initial parameter "height over-standard ≥8% triggers early warning", so that planning indicators are converted into real-time monitoring implementation benchmarks.

[0133] In the execution phase, the scheme realizes the dynamic connection of tasks and strategies through S300-S400: based on the overall goal of "raising the compliance rate of the block to 95%" and the current situation of "three high-rise construction" found by AI monitoring, S300 decomposes the task into "planning department verifies height compliance, urban management department identifies illegal location, and housing construction department formulates rectification scheme", and each department shares the AI evaluation report (marks that a construction point has been built to 10 meters) through the collaboration platform; during the execution, the urban management department confirms the illegal construction on site, the housing construction department urges the construction party to suspend work, and at the same time, real-time collection of rectification data (such as new design scheme height 8.5 meters) is input into the AI monitoring module, and after finding the secondary deviation of "material procurement is imitation blue brick (non-compliant blue brick)", the strategy is immediately adjusted (requires the planning department to review the material standard, and the housing construction department to check the material in advance), and is synchronized to the collaboration platform, to ensure that the implementation always meets the planning requirements.

[0134] After the phased rectification, S500 optimizes the connection of the closed loop through evaluation: calling the AI evaluation capability to compare the data before and after rectification - the height compliance rate increases from 65% to 98%, and the material compliance rate increases from 70% to 95%, while analyzing the effectiveness of the control measures, finding that "the housing construction department did not check the material in advance" led to material deviation, so the task allocation mechanism is updated (adding a "material procurement compliance check" sub-task for the housing construction department), and the AI material identification model parameters are optimized (improving the accuracy of distinguishing imitation blue bricks from real blue bricks); ultimately forming a complete connection chain of "planning indicators → data support → task execution → dynamic adjustment → evaluation optimization", not only solving the disconnection problem of "planning controlling planning, implementation controlling implementation" in traditional control, but also making the connection of subsequent block style control more accurate, and the control effect continues to improve.

[0135] In the embodiment of the present application, in the process of preliminarily standardizing the core data, the method further comprises:

[0136] Step 1, establish a cross-data-source style attribute semantic mapping rule:

[0137] For the differences in the expression of building height and facade material in different data sources (such as "building elevation 9 meters" in planning drawings, "relative ground height 8.5 meters" in patrol records, or both referring to "facade main material" but writing "blue brick" and "clay brick" respectively), first, a preset style attribute dictionary (containing standard expressions and associated explanations of "building height" and "facade material") is constructed; then, based on the dictionary, a unified semantic label is generated: for building height, mark "absolute height" (such as Huanghai elevation 9 meters) or "relative height" (such as 8.5 meters above ground) to distinguish the labels; for facade material, mark "main material" (such as exterior wall blue brick) or "auxiliary material" (such as window frame wood) to classify the labels, to ensure that the same attribute is expressed uniformly and can be compared in different data sources.

[0138] Step 2, AI image segmentation and spatial feature association matching:

[0139] For the building form in high-definition images, an AI image segmentation algorithm (such as the U-Net model) is used to separate the building main body from the background in the image, extract spatial feature data such as building contour (such as eave line, wall boundary), door and window proportion (such as total door and window area / total wall area), and the like; then, the spatial features are associated and matched with the corresponding attributes of the text data sources such as planning drawings (such as building contour design drawings), approval files (such as door and window size approval values), and the like (such as comparison of the contour size extracted from the image with the size marked on the drawing, comparison of the door and window proportion with the approval requirement), to ensure that the spatial feature data is consistent with the text attributes, and to further improve the integrity and accuracy of the standardized data.

[0140] In the embodiments of the present application, the AI style evaluation capability is realized through a multi-dimensional evaluation index system, which includes:

[0141] Step 1, construction of a quantitative index system:

[0142] Three types of core quantitative indicators and calculation methods are defined: building height deviation rate = (actual height-planned height) / planned height x 100% (such as actual 9 meters, planned 8 meters, deviation rate 12.5%); line fitting rate = building road interface length / road red line length x 100% (such as road interface 100 meters, road red line 120 meters, line fitting rate 83.3%); color deviation degree = actual color HSV value and planned color HSV value Euclidean distance (such as planned warm yellow HSV is (30, 50, 90), and actual is (35, 55, 85), and the deviation degree is calculated to be 7.1), to ensure that the indicators can be directly quantified and compared.

[0143] Step 2, construction of a qualitative index system:

[0144] Two types of qualitative index evaluation logic are set: the facade material matching degree is compared by AI image recognition of the actual material and the texture, density and other characteristics of the planned material, and the output is 0-100 points of coincidence (such as actual blue brick and planned blue brick matching degree 95 points, and imitation blue brick 60 points); the style element coordination is analyzed by analyzing the association of building form (such as pitched roof / flat roof) and surrounding historical context (such as traditional street scale), natural environment (such as waterfront topography), and the output is an adaptability score (such as 80 points for adaptation to historical block texture, and 40 points for conflict).

[0145] Step 3, dynamic allocation of index weights:

[0146] According to the city style area type (historical block, commercial business district, etc.) preset weight adjustment rule: historical block highlights material protection, sets the facade material matching degree weight to 30% (higher than other indicators); commercial business district emphasizes interface continuity, sets the curve fitting rate weight to 25%; at the same time, based on the S500 evaluation result fine-tuning (such as the color deviation of a historical block is frequent, the color deviation degree weight can be temporarily increased to 20%), ensure that the evaluation focus matches the regional style management and control demand.

[0147] The multi-dimensional evaluation index system combines the design of "quantitative index data benchmark, qualitative index evaluation depth, dynamic weight adaptation to regional characteristics", so that the AI style evaluation not only avoids the one-sidedness of single indicators, but also accurately matches the management and control focus of different style areas (such as historical blocks emphasize material, commercial areas emphasize interface), provides scientific and quantitative evaluation basis for subsequent task allocation and strategy formulation, ensures the accurate benchmarking of planning requirements and implementation effect from the evaluation dimension, and improves the pertinence and effectiveness of style management and control.

[0148] In the embodiment of the application, the collaborative management platform also integrates a department user hierarchical security module and an intelligent contract module, including:

[0149] Step 1, department user hierarchical security module configuration and operation:

[0150] Permission setting basis: first extract the core responsibilities of planning, urban management and housing departments (such as planning department leading evaluation report preparation, urban management department responsible for deviation record, housing department following up rectification), and clearly define the data boundaries of "need to view / need to operate" for each department;

[0151] Differential permission allocation: configure the planning department with "style evaluation report editing right + all department copy right" (can modify the evaluation conclusion and synchronize to all departments), configure the urban management department with "deviation problem record viewing right + supplementary editing right" (can view historical deviation and add on-site verification results), and configure the housing department with "rectification implementation record viewing right + progress updating right" (can view rectification scheme and input construction progress);

[0152] Permission taking effect and management: input the permission rules in the background of the collaborative management platform, bind the department account, and automatically match the permission when the user logs in (such as the housing department account cannot edit the evaluation report), at the same time, record the permission operation log (such as who viewed the deviation record, who updated the rectification progress), for easy traceability.

[0153] Step 2, intelligent contract module deployment and execution:

[0154] Contract content preset: Write "department responsibility list" (such as planning department responsible for appearance compliance assessment, urban management department responsible for illegal disposal) and "task response time limit" (such as compliance assessment needs to be completed within 3 days, illegal disposal needs to be filed within 24 hours) in the smart contract system;

[0155] Task matching and work order generation: when S300 decomposes subtasks (such as appearance compliance assessment, illegal disposal), the smart contract automatically reads the task type, matches the corresponding department (compliance assessment → planning department, illegal disposal → urban management department), generates an electronic work order with a unique number, task content and deadline, and pushes it to the corresponding department account;

[0156] Time limit monitoring and reminding: the smart contract system monitors the status of the work order in real time, and if the executing department does not provide feedback before the deadline (such as the urban management department does not provide feedback on the progress of illegal disposal within 24 hours), the system automatically triggers "APP push + SMS reminder", and records "delay time, delay task type" in the background to form a department performance archive.

[0157] Through the above process, the department user hierarchical security module accurately divides the permissions, avoiding information confusion caused by unauthorized access of data, and ensuring that each department can efficiently obtain the required data, realizing "data sharing on demand, safe and controllable flow"; the smart contract module reduces the cost of manual coordination, avoids departmental shirking or response delay, and guarantees the efficient landing of tasks from allocation to execution. The combination of the two provides the collaborative management platform with a "data security + execution guarantee" double foundation, allowing the planning, urban management and housing construction departments to promote control work in an environment of information interconnection without barriers and task execution with constraints, further strengthening the efficiency of planning and implementation, and improving the orderliness and reliability of overall appearance control.

[0158] In the embodiments of the present application, when the AI analysis and judgment finds that there is a deviation between the real appearance and the planning requirements, the method further comprises:

[0159] Step 1, establish deviation grading standards and judgment rules:

[0160] Determine the core dimension of classification: take "the influence degree of deviation on urban appearance" as the core, and divide it into "core element destruction", "general index violation" and "local detail inconsistency";

[0161] Clearly define the level of deviation: the first level of deviation is to "destroy the core landscape elements", including blocking the waterfront view corridor (such as building to block the view of the lake), altering the main structure of historical buildings (such as removing the load-bearing wall of historical buildings); the second level of deviation corresponds to "violation of general control indicators", including building height exceeding planning by 5%-10% (such as planning 80 meters, actual 84-88 meters), color deviation exceeding 15% (HSV color value difference > 15); the third level of deviation is limited to "local details not in line with", including facade component damage (such as external wall tiles falling off), lack of street greenery (such as planned green space being occupied);

[0162] Set AI judgment logic: enter the characteristic parameters of each level of deviation (such as height over-planning ratio threshold, HSV difference threshold) into the AI dynamic monitoring module, and automatically match the corresponding level after AI recognizes the deviation (such as detecting the removal of the load-bearing wall of a historical building, directly determining the first level of deviation).

[0163] Step 2, resource allocation and response execution for each level of deviation:

[0164] First level of deviation response: trigger cross-department joint resource allocation, automatically push collaborative disposal instructions to planning department (responsible for checking the extent of core element destruction) and urban management department (responsible for on-site law enforcement), system forces to set "2 hours to start disposal" time limit, and synchronously open core data shared by two departments (such as historical building protection red line map, waterfront view corridor planning map);

[0165] Second level of deviation response: automatically match the corresponding executing department (such as height over-planning matching planning department, color deviation matching urban management department), allocate "develop rectification plan" resources (such as plan template, relevant planning standards), and require 24 hours to complete plan preparation and upload to collaborative platform;

[0166] Third level of deviation response: allocate tasks to local executing departments (such as local urban management team, street housing office), allocate local rectification resources (such as facade repair materials list, greenery replanting standards), set 72-hour rectification completion time limit, and upload on-site photos for verification after rectification.

[0167] The deviation grading response mechanism achieves "precise grading and standard setting, resource allocation according to level and efficient response" by "precise grading and standard setting, resource allocation according to level and efficient response", which not only clearly defines the disposal boundaries of different deviations (avoiding overreaction to small problems or insufficient disposal of big problems), but also ensures response efficiency through rigid time limit and resource matching (such as 2-hour response of cross-department for first level of deviation, 72-hour rectification of local for third level of deviation), so that core landscape elements (such as historical buildings, waterfront views) are given priority protection, and general indicators and local details are solved in an orderly manner, achieving "problem grading precision, resource investment rationality, response efficiency and timeliness" in the aspect of deviation disposal, further strengthening the rapid connection between planning requirements and rectification implementation, and improving the emergency response capability and overall effectiveness of landscape control.

[0168] In the process of calling AI style evaluation capability comprehensive evaluation implementation effect in the embodiment of the application, the method further comprises:

[0169] Step 1, public experience evaluation dimension execution:

[0170] Public feedback data collection: summarize government affairs platform messages (such as citizen complaints, suggestions), offline questionnaire survey results (including "style satisfaction" and "improvement needs" items);

[0171] Natural language processing analysis: extract key information (such as "satisfaction" and "dissatisfaction" emotional words, "increase green" specific suggestions) through NLP technology, and quantify satisfaction (such as 80% of positive feedback corresponding to 80 points);

[0172] Generate quantitative score: convert the analysis result into a public experience score of 0-100 points (such as no negative feedback and reasonable suggestions, score 90 points).

[0173] Step 2, long-term impact evaluation dimension execution:

[0174] Data selection and comparison: collect style data (such as compliance rate after rectification) and regional economic data (business activity, tourism visit volume) 6-12 months after the completion of the stage task, and compare the baseline data before the task;

[0175] Positive influence coefficient calculation: obtain the influence coefficient through correlation analysis (such as 30% increase in visit volume mainly due to style improvement, coefficient 0.9).

[0176] Step 3, comprehensive implementation effect score fusion:

[0177] According to the preset weight (comparison of indicators before and after rectification 40%, public experience score 30%, long-term influence coefficient 30%), weighted calculation, such as rectification comparison score 85 points, public score 90 points, long-term coefficient 0.8 (corresponding to 80 points), comprehensive score = 85x0.4+90x0.3+80x0.3=85 points.

[0178] By including public feelings and long-term benefits, the evaluation is limited to only reaching the target, and the 4:3:3 weight fusion makes the comprehensive score reflect not only the direct effect of planning implementation, but also the public experience and regional development value, making the evaluation more comprehensive and objective, and providing a more practical basis for subsequent optimization.

[0179] Both by public experience dimension anchoring the actual needs of citizens, avoiding the disconnection between management and public feelings, and by extending the evaluation vision through the long-term impact dimension, ensuring that management goes beyond short-term index compliance, and better supports regional economic and tourism development; the comprehensive score weight of 4:3:3 balances direct results and multiple values, making the evaluation results more in line with actual management needs, providing accurate direction for subsequent AI model optimization, task allocation adjustment, and strategy iteration, further strengthening the practicality of planning and implementation linkage.

[0180] In the embodiments of the present application, when updating the AI landscape evaluation model parameters, executing the department task allocation mechanism and the collaborative management strategy based on the effectiveness analysis results, a closed-loop iterative optimization mechanism is established, and the method further comprises:

[0181] Step 1, AI landscape evaluation model parameter update:

[0182] Sample data screening: from the effectiveness analysis results, extract sample data corresponding to "comprehensive implementation effect score less than 80 points", such as building cases with high deviation rate exceeding the standard (actual height exceeding the planning by more than 10%), and public satisfaction score less than 70 points of landscape area image / text data;

[0183] Model retraining: include the screened samples into the model training set, divide the training / validation / test set in the ratio of 7:2:1, retrain along with the original AI model architecture (such as CNN+Transformer), and focus on optimizing the identification algorithm for low-score indicators (such as improving the accuracy of height deviation rate calculation);

[0184] Precision verification: verify the evaluation accuracy of the trained model with the test set to ensure that the evaluation error of low-score indicators is reduced by more than 10%, and the parameter update is completed.

[0185] Step 2, execute department task allocation mechanism adjustment:

[0186] Data extraction: retrieve the "task completion rate" (such as planning department compliance evaluation completion rate 92%, city management department violation disposal completion rate 85%) and "delay rate" (such as housing construction department rectification supervision delay rate 18%) of each department from the smart contract module;

[0187] Allocation weight adjustment: for departments with completion rate ≥ 90% (such as the planning department), increase the allocation proportion of similar tasks by 10%~15% (such as from 40% to 55%); for departments with delay rate > 15% (such as the housing construction department), reduce the amount of similar tasks by 20%, and push optimization suggestions (such as "prepare 7 days in advance to check rectification materials").

[0188] Step 3, collaborative management strategy supplement and update:

[0189] Public suggestion extraction: from the public experience evaluation results, screen out high-frequency mentioned improvement suggestions (such as "poor accessibility of waterfront walkway" "lack of street seats");

[0190] Clause transformation and synchronization: transform high-frequency suggestions into supplementary clauses of collaborative management strategies (such as "city management department checks the opening of waterfront walkway every month, and housing and construction department adds 1 place of street rest facility every quarter"), and synchronize to the collaborative management platform, cover the original strategy version and remind each department to execute.

[0191] The closed-loop iterative optimization mechanism solves the problems of inaccurate pre-evaluation, uneven execution efficiency, and insufficient strategy fit by "targeting optimization AI model precision, dynamically adjusting department task allocation, and accurately supplementing management strategies", and enables AI capabilities, department collaboration, and management strategies to continuously adapt to actual needs (such as public suggestions and execution performance), forming a virtuous cycle of "evaluating to find problems-optimizing to solve problems-re-evaluating and re-optimizing", which strengthens the depth of planning and implementation from the three dimensions of technology, execution, and demand, and ensures that the effect of style management and control can be long-term and stable.

[0192] The embodiment of the application discloses a planning and control system for the implementation effect of urban style planning, referring to Figure 2 , comprising:

[0193] The multi-source style data acquisition and standardization module 001 acquires various style data sources from the planning drawing database, field patrol records, high-definition image data, and building approval archives in real time, extracts style control core data including building height, facade material, color keynote, and street interface form, filters noise and removes outliers from the style control core data, unifies data format and description standards, completes preliminary standardization processing, and obtains corresponding standardized style data;

[0194] The AI style management and control capability and parameter configuration module 002 gives the control module AI style evaluation capability and AI dynamic monitoring capability based on standardized style data through AI algorithm, the AI style evaluation capability is used to analyze the compliance of buildings and planning indicators and the coordination of style elements, the AI dynamic monitoring capability is used to identify the deviation of real style and planning requirements in real time, and initial parameters including style compliance judgment standard and deviation early warning threshold are set;

[0195] The control task decomposition and coordination strategy module 003 obtains the overall cityscape control target, decomposes the complex control task into sub-tasks of cityscape compliance evaluation, deviation problem identification, and rectification scheme formulation based on the overall cityscape control target and the current implementation condition obtained by the AI dynamic monitoring capability, and assigns the sub-tasks to the corresponding execution departments of planning, urban management, and housing construction. The execution departments share the corresponding cityscape evaluation report and planning and reality comparison result generated based on the AI cityscape evaluation capability through the collaborative management platform, and collaboratively formulate the corresponding collaborative control strategy in combination with the responsibilities of the execution departments.

[0196] The control execution and real-time deviation adjustment module 004 executes the control action according to the collaborative control strategy, wherein the control action includes checking the planning index by the planning department, disposing the cityscape violation situation by the urban management department, and supervising the rectification implementation by the housing construction department. Real-time cityscape data are collected during the execution process, and the real-time cityscape data are input into the control module with AI dynamic monitoring capability. The AI analyzes and judges whether there is deviation between the real cityscape and the planning requirement. If there is deviation, the control strategy is adjusted and synchronized to the collaborative management platform.

[0197] The implementation effect evaluation and control optimization module 005 evaluates the implementation effect comprehensively after completing the phased control task, compares the cityscape data and planning index before and after rectification in the phased task, analyzes the effectiveness of the control measures, obtains the corresponding effectiveness analysis result, and updates the AI cityscape evaluation model parameter, the execution department task allocation mechanism, and the collaborative control strategy based on the effectiveness analysis result.

[0198] The embodiment of the application further discloses a cityscape planning implementation effect planning control method, comprising a processor, and the processor runs a program of the cityscape planning implementation effect planning control method of any one of the above.

[0199] The embodiment of the application further discloses a storage medium, which stores the program of the cityscape planning implementation effect planning control method of any one of the above.

[0200] Although the embodiments of the application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the application, and those skilled in the art can make changes, modifications, replacements, and variations to the above embodiments within the scope of the application.

Claims

1. A planning control method for evaluating the implementation effect of urban landscape planning, characterized in that, include: Real-time collection of various landscape data sources from planning drawing databases, on-site inspection records, high-definition image data, and building approval archives; extraction of core landscape control data including building height, facade material, color scheme, and street interface morphology; noise filtering and outlier removal of the core landscape control data; unification of data format and description standards; completion of preliminary standardization processing; and obtaining corresponding standardized landscape data. Based on the standardized landscape data, the management module is endowed with AI landscape assessment capabilities and AI dynamic monitoring capabilities through AI algorithms. The AI ​​landscape assessment capability is used to analyze the compliance of building and planning indicators and the coordination of landscape elements. The AI ​​dynamic monitoring capability is used to identify the deviation between the actual landscape and planning requirements in real time, and to set initial parameters including landscape compliance judgment standards and deviation warning thresholds. The system obtains the overall urban landscape management objectives. Based on these objectives and the current implementation status obtained from the AI ​​dynamic monitoring capabilities, the complex management tasks are broken down into sub-tasks: landscape compliance assessment, deviation problem identification, and rectification plan formulation. These sub-tasks are then assigned to the relevant implementation departments, including planning, urban management, and housing and construction. Each implementation department shares the corresponding landscape assessment report and the comparison results between planning and reality generated based on the AI ​​landscape assessment capabilities through the collaborative management platform. In conjunction with the responsibilities of each implementation department, a corresponding collaborative management strategy is formulated. Each implementing department executes control actions according to the aforementioned collaborative control strategy. These control actions include the planning department verifying planning indicators, the urban management department handling violations of urban appearance regulations, and the housing and construction department urging rectification. During the execution process, real-time data on the current urban appearance is collected and input into a control module with AI dynamic monitoring capabilities. The AI ​​analyzes and determines whether there is a deviation between the current urban appearance and the planning requirements. If there is a deviation, the control strategy is adjusted and synchronized to the collaborative management platform. After completing the phased management and control tasks, the AI ​​appearance assessment capability is invoked to comprehensively evaluate the implementation effect. The appearance data before and after the rectification in the phased management and control tasks are compared with the planning indicators to analyze the effectiveness of the management and control measures, obtain the corresponding effectiveness analysis results, and update the AI ​​appearance assessment model parameters, the task allocation mechanism of the implementing department, and the collaborative management and control strategy based on the effectiveness analysis results. When AI analysis determines that there are deviations between the current urban landscape and planning requirements, the method includes: establishing a graded response mechanism for deviations, classifying deviations into Level 1, Level 2, and Level 3 based on their impact on the urban landscape. Level 1 deviations involve damage to core landscape elements, including obstructing waterfront views and altering the main structure of historical buildings; Level 2 deviations involve violations of general control indicators, including building height exceeding the plan by 5% to 10% and color deviation exceeding 15%; Level 3 deviations involve discrepancies in local details, including damaged facade components and missing street-side greenery. For different levels of deviation, corresponding resources are automatically allocated. Level 1 deviations trigger a cross-departmental joint response, requiring planning and urban management departments to initiate collaborative handling within 2 hours; Level 2 deviations require the corresponding implementing department to formulate a rectification plan within 24 hours; and Level 3 deviations require the local implementing department to complete rectification within 72 hours.

2. The planning control method for the implementation effect of urban landscape planning according to claim 1, characterized in that, The method also includes the following steps in the preliminary standardization process of the core data for landscape control: Establish a semantic mapping rule for landscape attributes across data sources. For the same name or synonymous alternatives of building height and facade material in different data sources, generate unified semantic tags based on a preset landscape attribute dictionary. The semantic tags include labels that distinguish between absolute and relative building height, and labels that classify main and auxiliary materials of facade material. AI image segmentation is used to process the architectural features in high-definition image data to extract spatial feature data corresponding to the building outline and the proportion of doors and windows. The spatial feature data is then matched with the corresponding attributes in text data sources such as planning drawings and approval documents.

3. The planning control method for the implementation effect of urban landscape planning according to claim 1, characterized in that, AI-powered appearance assessment capabilities are achieved through a multi-dimensional assessment indicator system, including: The multi-dimensional evaluation index system is divided into quantitative indicators and qualitative indicators. Among them, the quantitative indicators include building height deviation rate, line alignment rate, and color deviation degree. The building height deviation rate is the ratio of the difference between the actual height and the planned height to the planned height. The line alignment rate is the ratio of the length of the building's road-facing interface to the length of the road red line. The color deviation degree is the difference in HSV color value between the actual color and the planned color tone. Qualitative indicators include the matching degree of facade materials and the coordination of style elements. The matching degree of facade materials refers to the degree of conformity between the actual materials and the materials required by the plan. The coordination of style elements refers to the compatibility of the building form with the surrounding historical context and natural environment. Furthermore, the AI ​​landscape assessment capability also has a dynamic weight allocation function, which adjusts the weight of each indicator according to the different landscape types of the city. The weight of material matching degree is set to 30% in historical blocks and the weight of line coverage rate is set to 25% in commercial and business districts.

4. The planning control method for the implementation effect of urban landscape planning according to claim 1, characterized in that, The collaborative management platform also integrates a departmental user hierarchical confidentiality module and a smart contract module, including: The departmental user hierarchical confidentiality module is used to set differentiated permissions according to the responsibilities and authority levels of each implementing department. The planning department has the right to edit the landscape assessment report and the right to send it to all departments. The urban management department has the right to view the deviation problem record and the right to supplement and edit it. The housing and construction department has the right to view the rectification implementation record and the right to update the progress. The smart contract module pre-writes the responsibility list and task response time limit of each execution department. After the sub-task is decomposed, the smart contract automatically matches the execution department according to the task type and generates a task work order with a time limit. The appearance compliance assessment corresponds to the planning department, and the violation handling corresponds to the urban management department. If the execution department fails to provide feedback on the handling result within the time limit, the smart contract automatically triggers the reminder mechanism and records the delay information.

5. The planning control method for the implementation effect of urban landscape planning according to claim 1, characterized in that, In the process of comprehensively evaluating the implementation effect by utilizing the AI ​​landscape assessment capability, the method also includes: The system introduces a public experience evaluation dimension and a long-term impact evaluation dimension. The public experience evaluation dimension analyzes public feedback data through natural language processing technology. The public feedback data specifically includes messages on government platforms and questionnaires. Key information such as satisfaction and improvement suggestions are extracted and converted into a quantitative score of 0-100. The long-term impact assessment dimension compares the landscape data with regional economic data 6-12 months after the completion of the phased tasks. The regional economic data specifically includes business activity and cultural and tourism visit volume. The positive impact coefficient of landscape control on regional development is analyzed. The results of the comparison of indicators before and after rectification, public experience scores, and long-term impact coefficients are combined with a weighting of 4:3:3 to generate a comprehensive implementation effect score.

6. The planning control method for the implementation effect of urban landscape planning according to claim 1, characterized in that, When updating the AI ​​landscape assessment model parameters, the task allocation mechanism for implementing departments, and the collaborative management and control strategy based on the aforementioned effectiveness analysis results, a closed-loop iterative optimization mechanism is established. The method further includes: The sample data corresponding to the indicators with an overall implementation effect score of less than 80 points were included in the training set of the AI ​​landscape assessment model. The sample data included building cases with excessive height deviation rate and landscape area data with low public satisfaction. The model was retrained to improve the assessment accuracy of indicators with an overall implementation effect score of less than 80 points. Based on the task completion rate and delay rate of the executing departments recorded by the smart contract, the task allocation weight is adjusted. For departments with a completion rate of ≥90%, the proportion of similar tasks is increased, and for departments with a delay rate of >15%, the task load is reduced and optimization suggestions are pushed. The improvement suggestions frequently mentioned in the public experience assessment, which aim to improve the accessibility of waterfront walkways, will be incorporated into supplementary provisions of the collaborative management strategy.

7. A system for implementing a planning control method for assessing the effectiveness of urban landscape planning as described in any one of claims 1-6, characterized in that, include: The multi-source landscape data acquisition and standardization module collects various landscape data sources in real time from planning drawing databases, on-site inspection records, high-definition image data, and building approval archives. It extracts core landscape control data including building height, facade material, color tone, and street interface form. The module performs noise filtering and outlier removal on the core landscape control data, unifies the data format and description standards, completes the preliminary standardization process, and obtains the corresponding standardized landscape data. The AI-powered landscape management and parameter configuration module, based on the standardized landscape data, uses AI algorithms to endow the management module with AI landscape assessment capabilities and AI dynamic monitoring capabilities. The AI ​​landscape assessment capability is used to analyze the compliance of building and planning indicators and the coordination of landscape elements. The AI ​​dynamic monitoring capability is used to identify deviations between the actual landscape and planning requirements in real time, and to set initial parameters including landscape compliance judgment standards and deviation warning thresholds. The management task decomposition and collaboration strategy module obtains the corresponding overall urban landscape management objectives. Based on the overall urban landscape management objectives and the current implementation status obtained by the AI ​​dynamic monitoring capabilities, the module decomposes complex management tasks into sub-tasks such as landscape compliance assessment, deviation problem identification, and rectification plan formulation. These sub-tasks are then assigned to the corresponding implementation departments, namely planning, urban management, and housing and construction. Each implementation department shares the corresponding landscape assessment report and the comparison results between planning and reality generated based on the AI ​​landscape assessment capabilities through the collaborative management platform. The module also collaboratively formulates corresponding collaborative management strategies in conjunction with the responsibilities of each implementation department. The control execution and real-time deviation adjustment module allows each implementing department to execute control actions according to the collaborative control strategy. These control actions include the planning department verifying planning indicators, the urban management department handling violations of landscape regulations, and the housing and construction department urging rectification. During the execution process, real-time landscape data is collected and input into the control module with AI dynamic monitoring capabilities. The AI ​​analyzes and determines whether there is a deviation between the real landscape and the planning requirements. If a deviation exists, the control strategy is adjusted and synchronized to the collaborative management platform. The implementation effect evaluation and control optimization module, after completing the phased control tasks, calls the AI ​​appearance evaluation capability to comprehensively evaluate the implementation effect, compares the appearance data before and after the rectification in the phased control tasks with the planning indicators, analyzes the effectiveness of the control measures, obtains the corresponding effectiveness analysis results, and updates the AI ​​appearance evaluation model parameters, the task allocation mechanism of the implementing department and the collaborative control strategy based on the effectiveness analysis results; When AI analysis determines that there are deviations between the current urban landscape and planning requirements, the following measures are implemented: A graded response mechanism for deviations is established, classifying deviations into Level 1, Level 2, and Level 3 based on their impact on the urban landscape. Level 1 deviations involve damage to core landscape elements, including obstructing waterfront views and altering the main structure of historical buildings. Level 2 deviations violate general control indicators, including exceeding building height by 5%–10% and color deviation exceeding 15%. Level 3 deviations involve discrepancies in local details, including damaged facade components and missing street-side greenery. For each level of deviation, corresponding resources are automatically allocated. Level 1 deviations trigger a cross-departmental joint response, requiring planning and urban management departments to initiate coordinated action within 2 hours. Level 2 deviations require the corresponding implementing department to develop a rectification plan within 24 hours. Level 3 deviations require the local implementing department to complete rectification within 72 hours.

8. A planning control system for evaluating the implementation effect of urban landscape planning, characterized in that, Includes a processor, wherein the processor runs a program for a planning control method for the implementation effect of urban landscape planning as described in any one of claims 1-6.

9. A storage medium, characterized in that, A program storing a planning control method for assessing the implementation effect of urban landscape planning as described in any one of claims 1-6.

Images