Full-life-cycle dynamic monitoring system and method for global land comprehensive improvement

Abstract

The invention provides a full-life-cycle dynamic monitoring system and method for global land comprehensive improvement, and the method comprises the steps: firstly collecting and fusing multi-source heterogeneous data related to global land comprehensive improvement through a multi-source data fusion module, and constructing a data bottom plate with unified space-time reference and semantic association; then the intelligent monitoring module realizes land utilization type change identification, project progress tracking and index dynamic monitoring by utilizing a preset business rule and a machine learning model based on the data base plate, obtains monitoring data in real time and generates early warning information; and finally, the decision support module visually displays the monitoring data and the early warning information, and provides interaction service and decision support for the user at the same time. According to the invention, multi-source heterogeneous data can be efficiently integrated, whole-process dynamic monitoring, intelligent effect evaluation and risk early warning of a global land comprehensive improvement project are realized, and powerful support is provided for management decision.

Description

Technical Field

[0001] The present invention relates to the field of land resource management technology, and more specifically, to a full-life cycle dynamic monitoring system and method for comprehensive land management across the entire region. Background Art

[0002] Comprehensive land consolidation is a comprehensive project based on national land space planning, coordinating the consolidation of agricultural and construction land, and ecological restoration. It effectively improves land intensive utilization efficiency, while simultaneously restoring the ecology, improving the human living environment, promoting industrial integration, and facilitating rural revitalization and urban-rural integration, thereby achieving win-win economic, social, and ecological benefits. Therefore, efficient, accurate, and dynamic monitoring and supervision of comprehensive land consolidation projects throughout the entire process are crucial.

[0003] At present, the following problems exist in the process supervision of comprehensive land remediation in the whole region: 1) Data fragmentation and silos: Remediation involves multiple stages including planning, implementation, acceptance, and maintenance, and involves multiple departments such as natural resources, agriculture and rural areas, forestry, environmental protection, water conservancy, and housing and construction. Data sources are scattered and in different formats, making it difficult to effectively integrate and form a full-area, full-cycle view.

[0004] 2) Poor monitoring timeliness: Existing technologies mostly rely on a combination of text reports, tables, images, and manual on-site spot checks for data monitoring. This has a long cycle, high costs, and limited coverage, making it difficult to promptly detect deviations (such as illegal occupation, project delays, and substandard quality) and potential risks in the remediation process.

[0005] 3) The analysis and evaluation are highly subjective: the evaluation of the effectiveness of remediation (such as the increase in the amount of cultivated land and the improvement of land use efficiency) mostly relies on simple quantitative indicators, and lacks the support of objective, qualitative, and multi-dimensional intelligent analysis models for the mid- and late stages of project implementation.

[0006] 4) Insufficient early warning capabilities: It is difficult to proactively issue early warnings based on real-time or near-real-time data for issues such as delayed progress of remediation projects, abnormal use of funds, negative impacts on the ecological environment, and deviations from plan implementation.

[0007] 5) Weak decision support: There is a lack of effective tools to transform monitoring data into intuitive and actionable decision-making information (such as optimizing resource allocation, adjusting remediation strategies, and identifying demonstration areas). Summary of the Invention

[0008] In order to overcome the defect of the above-mentioned existing technologies that it is difficult to achieve efficient and high-precision supervision of the entire process of comprehensive land improvement in the entire region, the present invention provides a full-life cycle dynamic monitoring system and method for comprehensive land improvement in the entire region, which can efficiently integrate multi-source heterogeneous data, realize dynamic monitoring of the entire process of implementation of comprehensive land improvement projects in the entire region, intelligent effectiveness evaluation and risk warning, and provide strong support for management decisions.

[0009] In order to solve the above technical problems, the technical solutions of the present invention are as follows: A full-life cycle dynamic monitoring system for comprehensive land management in the entire region, comprising: a multi-source data fusion module, an intelligent monitoring module, and a decision support module connected in sequence; The multi-source data fusion module includes: a data access layer, a cleaning and management layer, a fusion processing layer, and an application interface layer connected in sequence; the multi-source data fusion module is used to collect and fuse multi-source heterogeneous data related to the comprehensive land improvement of the entire region, and build a data base with unified spatiotemporal references and semantic associations; The intelligent monitoring module includes: a characteristic index extraction layer, a dynamic change analysis layer, an evaluation and early warning layer, and a feedback optimization layer connected in sequence; the intelligent monitoring module is used to identify land use type changes, track project progress, and dynamically monitor indicators based on the data base, using preset business rules and machine learning models, to obtain monitoring data in real time and generate early warning information; The decision support module includes: a data link layer, a visualization rendering layer, an interactive service layer and a decision support layer connected in sequence; the data link layer is connected to the dynamic change analysis layer and the assessment and early warning layer respectively; the decision support module is used to visualize the monitoring data and early warning information, and provide interactive services and decision support to users.

[0010] Preferably, in the multi-source data fusion module, The data collected by the data access layer includes at least one or more of: satellite remote sensing images, drone aerial survey data, ground sensor data, business database and document table data, and any one or more of the publicly available data on the Internet; the data collected by the data access layer is classified, and the classified data is stored in multiple nodes using DSF / HDSF distributed storage technology; The cleaning management layer is used to clean the data collected by the data access layer based on predefined cleaning rules. The data cleaning includes at least one or more of: outlier marking and correction, missing value interpolation, and duplicate value removal; The fusion processing layer is used to sequentially perform spatiotemporal benchmark unification, semantic association modeling, and business scenario adaptation processing on the cleaned data to generate the data backplane; the spatiotemporal benchmark unification includes unifying and correcting the time and space coordinates of multi-source heterogeneous data; the semantic association modeling includes constructing a knowledge graph based on the spatiotemporal benchmark unified data and performing entity semantic association; the business scenario adaptation includes dynamically packaging the semantic association modeled data into several data sets adapted to different stages based on the data differences required for different stages of comprehensive land improvement in the entire region, and saving the data sets corresponding to all stages together as the data backplane; The application interface layer is used to output the data backplane to the intelligent monitoring module through a preset multi-element interface.

[0011] Preferably, in the intelligent monitoring module, The characteristic index extraction layer is used to extract a number of monitoring indicators from the data base and perform characteristic quantification processing based on the pre-built comprehensive land improvement monitoring indicator system for the entire region; The dynamic change analysis layer is used to automatically identify land parcels based on a preset machine learning model, further identify changes in land use types, and extract change data; at the same time, it implements project progress tracking and dynamic indicator monitoring based on preset business rules, and obtains monitoring data in real time; The evaluation and warning layer is used to compare the monitoring data of each indicator with the preset threshold value, implement hierarchical warning, and generate warning information; The feedback optimization layer is used to receive the actual rectification situation and compare it with the early warning information, and optimize the machine learning model and business rules based on the comparison results.

[0012] Preferably, in the characteristic index extraction layer, a comprehensive land improvement monitoring index system is constructed based on the hierarchical analysis method, and the index system at least includes: a target layer, a criterion layer and an index layer; The target layer includes: basic bottom-line constraints, improvement of farmland ecosystem service functions, ecological protection and restoration and improvement of rural landscape, project and fund coordination and project supervision; At the criteria level, basic bottom-line constraints include: quantity control, quality control, and other controls; improving farmland ecosystem service functions includes: the degree of farmland concentration and contiguousness, the degree of intensive and economical use of construction land, and land transfer; ecological protection and restoration and rural landscape improvement include: ecological protection and restoration and rural landscape improvement; project and funding coordination includes: funding guarantees for improvement, project completion progress, and project quality targets; and project supervision includes: innovation and illegal monitoring. In the indicator layer, quantity control includes: the proportion of newly added cultivated land area, the proportion of newly added permanent basic farmland area and the surplus construction land index; quality control includes: improving the quality grade of cultivated land; other controls include: the area that conflicts with the ecological protection red line and the area of ​​historical context protection; the degree of farmland concentration and contiguousness includes: the area of ​​newly added high-standard farmland, the area of ​​developed and supplemented cultivated land, the area of ​​reclaimed paddy fields, the number of cultivated land small fields converted into large fields and the degree of increase in the concentration and contiguousness of permanent basic farmland; the degree of intensive and economical use of construction land includes: the proportion of increase and decrease linked indicators used for infrastructure construction, the scale of redevelopment of inefficient construction land, the scale of demolition and reclamation of rural construction land, the rate of decrease in the average household homestead area and the scale of reduction in village construction land; land transfer status includes: the area of ​​cultivated land for grain cultivation after remediation, the introduction of agricultural enterprises, the number of large grain-growing households and the proportion of land management rights transfer; ecological The protection and restoration situation includes: the area of ​​comprehensive mine management, the area of ​​comprehensive water environment management, the area of ​​forestland transformation, the area of ​​mangrove protection and restoration, the area of ​​coastal management, the number of kilometers of newly added ecological corridors and the soil and water conservation rate; the rural landscape improvement situation includes: the number of rural human settlement environment improvements, the coverage rate of sewage treatment facilities, the coverage rate of garbage treatment and the penetration rate of sanitary toilets; the improvement funding guarantee situation includes: the proportion of integrated agricultural funds, the completion rate of budget investment funds, the actual expenditure rate of funds and the proportion of social capital investment; the project completion progress includes: the completion rate of agricultural land improvement projects, the completion rate of construction land improvement projects, the completion rate of ecological protection and restoration projects, the completion rate of cultural protection and rural landscape improvement and other projects; the project quality target situation includes: the project quality qualification rate; innovation and illegal monitoring includes: the number of innovative systems and the area of ​​illegal land use.

[0013] Preferably, in the dynamic change analysis layer, the preset machine learning model is specifically an improved PiDiNet model, which is used for land parcel edge extraction; The structure of the improved PiDiNet model includes sequentially connected convolution blocks 1, 2, 3, and 4. Each convolution block outputs feature maps of different channels. After each feature map is uniformly converted into a channel, it is input into an attention layer to obtain the corresponding attention features. After the attention features corresponding to convolution blocks 2 and 3 are respectively subjected to feature difference, they are feature spliced ​​with the attention features corresponding to convolution block 1 to obtain spliced ​​features. The spliced ​​features are again subjected to channel conversion, and finally the Sigmoid function is used to obtain the detection result of the land edge. The convolution block 1 and the convolution block 2 have the same structure, both including a Gabor convolution layer, a batch normalization layer, a maximum pooling layer and an activation layer connected in sequence; the convolution block 3 and the convolution block 4 have the same structure, both including a convolution layer, a batch normalization layer, a maximum pooling layer and an activation layer connected in sequence.

[0014] Preferably, the Gabor convolution layer includes at least one Gabor filter and a learnable convolution kernel.

[0015] Preferably, in the decision support module, The data link layer is used to collect the monitoring data and early warning information in real time; The visualization rendering layer is used to achieve three-dimensional terrain visualization, two-dimensional thematic visualization, and timeline dynamic visualization based on monitoring data and early warning information. The three-dimensional terrain visualization includes: constructing a three-dimensional real-life model of the entire land comprehensive improvement project area, overlaying the monitoring data and early warning information of each plot to achieve three-dimensional information browsing; the two-dimensional thematic visualization uses an external GIS system to overlay and analyze monitoring data and early warning information to generate multi-thematic layers; the timeline dynamic visualization uses the timeline of the entire land comprehensive improvement project as the axis, combining monitoring data and early warning information to achieve dynamic playback and future deduction of the improvement process; The interactive service layer is used to provide interactive services for users, and the interactive services include general services and customized services; The decision support layer is used to provide decision support for users.

[0016] Preferably, in the visual rendering layer, a three-dimensional real-scene model of the entire land comprehensive improvement project area is constructed based on the WebGL or Unity 3D engine.

[0017] Preferably, in the interactive service layer, general services include: intelligent data annotation, spatial measurement, comparative analysis and intelligent search; The customized services include: providing managers with a cockpit view and a dynamic dashboard integrating core indicators; providing technical personnel with a professional analysis view, overlaying engineering drawings and sensor network topology diagrams for data interface debugging and model parameter adjustment; and providing grassroots users with a simplified interface to intuitively view the improvement effects of their own land through a three-dimensional real-life model, and providing online feedback functions.

[0018] The present invention also provides a full life cycle dynamic monitoring method for comprehensive land management in the entire region, based on the above system, comprising the following steps: S1: The multi-source data fusion module collects and fuses multi-source heterogeneous data related to comprehensive land management in the entire region to build a data base with unified spatiotemporal reference and semantic association; S2: Based on the data base, the intelligent monitoring module uses preset business rules and machine learning models to identify land use type changes, track project progress, and dynamically monitor indicators, obtain monitoring data in real time, and generate early warning information; S3: The decision support module visualizes the monitoring data and warning information, and provides interactive services and decision support to users.

[0019] The present invention provides a full-life cycle dynamic monitoring system and method for comprehensive land improvement across the entire region. First, a multi-source data fusion module is used to collect and fuse multi-source heterogeneous data related to comprehensive land improvement across the entire region, constructing a data base with unified spatiotemporal benchmarks and semantic associations. Then, based on the data base, an intelligent monitoring module uses preset business rules and machine learning models to identify land use type changes, track project progress, and dynamically monitor indicators, acquiring monitoring data in real time and generating early warning information. Finally, a decision support module visualizes the monitoring data and early warning information, while providing interactive services and decision support to users. Compared with the prior art, the beneficial effects of the technical solution of the present invention are: 1) Full-area and full-cycle coverage: Dynamic tracking of the entire process and all elements of comprehensive land improvement projects within a specific scope, from planning to management and maintenance.

[0020] 2) Deep integration of multi-source data: Break down data silos, form a unified and authoritative “data platform” for rectification, and improve information integrity.

[0021] 3) Real-time automated monitoring: significantly reduces manual reliance, improves monitoring frequency and efficiency, and enables timely detection and resolution of problems.

[0022] 4) Objective and intelligent evaluation: Model-driven quantitative evaluation reduces subjectivity and comprehensively and scientifically measures the effectiveness of rectification.

[0023] 5) Precise and proactive early warning: This invention can transform the traditional post-event handling plan into a pre-event early warning and in-event intervention plan, effectively preventing risks and improving management initiative.

[0024] 6) Scientific visualization of decision-making: Provides intuitive and comprehensive information views and auxiliary analysis tools to significantly improve the scientific nature and efficiency of user decision-making.

[0025] 7) Standardization and scalability: The system architecture and method of the present invention are universal and can provide a reference for comprehensive land improvement monitoring in different regions, with a wide range of adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a structural diagram of a full-life cycle dynamic monitoring system for comprehensive land management provided in Example 1.

[0027] Figure 2 This is the overall architecture diagram of a full-life cycle dynamic monitoring system for comprehensive land improvement in the entire region provided in Example 2.

[0028] Figure 3 This is a schematic diagram of the DSF / HDSF distributed storage technology provided in Example 2.

[0029] Figure 4 This is a schematic diagram of the construction of the comprehensive land improvement monitoring indicator system provided in Example 2.

[0030] Figure 5 This is a schematic diagram of the plot identification process provided in Example 2.

[0031] Figure 6 Schematic diagram of the Gabor convolution layer provided in Example 2.

[0032] Figure 7 This is a schematic diagram of the detection effect of the homogeneous mass provided in Example 2.

[0033] Figure 8 This is a schematic diagram of the detection effect of the non-uniform mass provided in Example 2.

[0034] Figure 9 This is a schematic diagram of the detection effect of dense plots provided in Example 2.

[0035] Figure 10 This is a flow chart of a full life cycle dynamic monitoring method for comprehensive land improvement in the entire region provided in Example 3. DETAILED DESCRIPTION

[0036] The accompanying drawings are for illustrative purposes only and are not to be construed as limiting the present application; In order to better illustrate this embodiment, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product size; It is understandable to those skilled in the art that some well-known structures and descriptions thereof may be omitted in the drawings.

[0037] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0038] Example 1 like Figure 1 As shown, this embodiment provides a full-life cycle dynamic monitoring system for comprehensive land management in the entire region, including: a multi-source data fusion module, an intelligent monitoring module, and a decision support module connected in sequence; The multi-source data fusion module includes: a data access layer, a cleaning and management layer, a fusion processing layer, and an application interface layer connected in sequence; the multi-source data fusion module is used to collect and fuse multi-source heterogeneous data related to the comprehensive land improvement of the entire region, and build a data base with unified spatiotemporal references and semantic associations; The intelligent monitoring module includes: a characteristic index extraction layer, a dynamic change analysis layer, an evaluation and early warning layer, and a feedback optimization layer connected in sequence; the intelligent monitoring module is used to identify land use type changes, track project progress, and dynamically monitor indicators based on the data base, using preset business rules and machine learning models, to obtain monitoring data in real time and generate early warning information; The decision support module includes: a data link layer, a visualization rendering layer, an interactive service layer and a decision support layer connected in sequence; the data link layer is connected to the dynamic change analysis layer and the assessment and early warning layer respectively; the decision support module is used to visualize the monitoring data and early warning information, and provide interactive services and decision support to users.

[0039] In the specific implementation process, this embodiment builds a dynamic monitoring and decision support platform for the entire life cycle of comprehensive land remediation, based on multi-source data fusion and centered on an intelligent analysis engine. The system architecture is as follows: (1) Multi-source data fusion module (data end); Comprehensive land remediation in the entire region is a comprehensive work with multiple objects, wide coverage, and complex "planning-engineering" attributes. It involves business data from multiple departments such as natural resources, agriculture and rural areas, ecological environment, and water conservancy, as well as multi-source heterogeneous data such as satellite remote sensing, drone inspections, ground sensors, and mobile terminal collection (such as spatiotemporal images, sensor time series data, business ledgers, planning drawings, policy texts, etc.). There are pain points such as heterogeneous formats (structured / semi-structured / unstructured), inconsistent spatiotemporal benchmarks, uneven quality, and weak semantic associations, which make it difficult to use data directly for intelligent assessment and early warning. Therefore, this module adopts a layered decoupling architecture, which is divided into data access layer, cleaning and governance layer, fusion processing layer, and application interface layer from the bottom layer to the application layer. Each layer works together to realize the "collection, management, integration, and use" closed loop of multi-source data.

[0040] 1) Data access layer: Through microservice architecture design, it is compatible with mainstream protocols such as MQTT, HTTP / HTTPS, FTP, OGC WMS / WFS, JDBC / ODBC, and supports fully compatible access to multi-source heterogeneous data, including satellite remote sensing images (optical / radar / SAR), drone aerial survey data, ground sensor networks, business databases (such as national land space planning, land use status, and cultivated land occupation and compensation balance systems), documents / tables (project ledgers, approval documents, policies and regulations), and Internet public data (such as meteorological disasters and regional population and economy).

[0041] 2) Cleaning and Governance Layer: To address the "dirty, disorganized, and fragmented" issues of multi-source data, this layer automatically corrects or marks outliers through structured data governance based on predefined business rules (such as consistency of land use classification codes and logical verification of area values). It also supports missing value interpolation and duplicate value removal. Furthermore, adaptive algorithms are used to optimize data quality for remote sensing imagery (radiometric correction and geometric registration) and documents (OCR recognition and extraction of key information).

[0042] 3) Fusion Processing Layer: Through the three-level integration of spatiotemporal benchmark unification, semantic association modeling, and business scenario adaptation, a "single map" data foundation for comprehensive land consolidation across the entire region is constructed. Spatiotemporal benchmark unification uses a unified coordinate system for the project area to calibrate the spatiotemporal coordinates of multi-source data. For data of varying resolutions, sub-pixel decomposition and spatial interpolation are used to achieve multi-scale fusion. Semantic association modeling, based on knowledge graph technology, constructs a land consolidation domain ontology (including core entities and relationships such as "project-plot-problem-measure-subject"). Distributed data is linked through entities to form "plot-attribute-behavior" triples, addressing data semantic fragmentation. Business scenario adaptation dynamically assembles fused datasets tailored to the business needs of the entire land consolidation cycle (planning, implementation, acceptance, and maintenance). For example, during the "implementation phase," remote sensing imagery, sensor data, and mobile device reporting are integrated to generate a multidimensional dynamic dataset covering "project progress, construction quality, and ecological impact."

[0043] 4) Application Interface Layer: This layer provides data services to upper-layer applications such as intelligent assessment and early warning monitoring, project management collaboration, and public services through multiple interfaces, including RESTful APIs, spatial data services (WMS / WFS), and message queues (Kafka). This layer provides a unified data catalog (e.g., "Project Basic Information," "Real-time Monitoring Data," and "Historical Remediation Case Library"), metadata descriptions (data source, accuracy, and update time), and access control. Furthermore, it can output labeled sample datasets for intelligent assessment model training, push real-time or near-real-time dynamic data streams (e.g., "Project progress is lagging") for early warning monitoring, and output lightweight data packages (e.g., satellite image thumbnails of the project area) for mobile applications.

[0044] (2) Intelligent monitoring module (monitoring terminal); To address the challenges faced by project managers in timely monitoring the ongoing changes in land use, project progress, and funding during comprehensive land remediation, as well as the "dynamic anomalies" of some projects, this module, centered on "full-factor perception, full-cycle tracking, and intelligent analysis," adopts a closed-loop "perception-analysis-warning-feedback" architecture. From underlying data processing to upper-level intelligent applications, this module implements the "capture-analysis-prediction-intervention" approach for dynamic changes layer by layer. The module's technical architecture consists of a feature index extraction layer, a dynamic change analysis layer, an assessment and warning decision-making layer, and a feedback optimization layer. These layers collaborate to support the dynamic monitoring needs of the remediation process.

[0045] 1) Feature Indicator Extraction Layer: Targeting core elements of the land consolidation process, such as "land type changes, project progress, ecological indicators, and problem risks," this layer converts project data from agricultural land consolidation, construction land consolidation, rural ecological protection and restoration, and rural landscape enhancement projects into analyzable dynamic features. Specifically, based on the nature and objectives of the monitoring indicators, indicator factors related to the objectives are clustered and combined at different levels according to their affiliation, thus forming a multi-objective, multi-level indicator system. In conjunction with the advice of experts from the land, agriculture, forestry, and water conservancy departments, a comprehensive regional consolidation monitoring indicator system is constructed from baseline, basic, and auxiliary evaluations. The selected feature indicators are then quantified using remote sensing image monitoring technology, three-dimensional auxiliary technology, close-range photography monitoring technology, GNSS monitoring technology, field survey technology, automatic land type identification technology, and survey statistical analysis techniques.

[0046] 2) Dynamic Change Analysis Layer: This layer is the core of the module. It primarily upgrades the remediation process from "phenomenon perception" to "causal analysis" through automated land classification change identification and embedded business rules. Automated land classification identification primarily uses a pre-configured deep learning network model to perform pixel-level classification of land types (such as cultivated land, forest land, and construction land). It then automatically identifies and extracts change characteristics for land type changes within the remediation project, combining a time series of pre-, mid-, and post-remediation data. Embedded business rule analysis utilizes a built-in library of land remediation policy standards (such as the "Land Remediation Project Management Measures" and "Cultivated Land Quality Grades") to transform business rules into analytical logic, enabling monitoring of the characteristic indicators extracted at the previous layer. For example, if it detects that "cultivated land restoration in a project area has not reached 80% of the planned area," it automatically correlates data such as "construction machinery input" and "capital investment" to pinpoint the cause of the progress delay.

[0047] 3) Assessment and Early Warning Decision-Making Layer: Combining dynamic analysis results with pre-set thresholds, a three-tiered early warning system (Yellow, Orange, and Red) is established to ensure early detection and rapid response to issues. Abnormality warnings (Yellow) primarily address short-term sudden changes, such as "a plot of land suddenly changes to construction land" or "sensor data jumps by more than 20%." These warnings are triggered through real-time calculations and sent to project managers, requiring a review within 48 hours. Risk warnings (Orange) primarily address deteriorating trends, such as "ecological benefit indicators fall 15% below expectations." Risk levels are assessed based on business rules and sent to management departments, requiring intervention plans such as adjusting irrigation methods and strengthening protective engineering. Major risk warnings (Red): For systemic issues (such as "the remediation area overlaps with permanent basic farmland by more than 5%" or "project progress is delayed by more than 30%, potentially impacting acceptance"), alternative plans are generated to accelerate construction progress through measures such as adjusting land use layout, and these plans are simultaneously sent to the relevant authorities.

[0048] 4) Feedback optimization layer: Continuously optimize monitoring capabilities through a data backflow mechanism, compare early warning results with actual rectification situations (e.g., “illegal buildings on a certain plot of land have been demolished”), evaluate early warning accuracy, and based on verification results, update AI model parameters (e.g., adjust the feature weights of the land classification model) or rule base (e.g., supplement the identification rules for “new types of violations”), and enter typical problems (e.g., “the recovery of cultivated land in a certain area is slow due to slope factors”) into the case library to provide experience reference for subsequent projects.

[0049] (3) Decision support module (management side); This module takes "data visualization, decision-making scenario, and operation collaboration" as its core, and builds a visualization display platform and intelligent decision-making support tool set covering the entire life cycle of "planning-implementation-acceptance-maintenance", converting abstract data into intuitive graphic language, and converting decision logic into actionable scenario guidance, providing project managers, technical teams, and regulatory departments with "visual, simulatable, and traceable" decision-making tools. It is the application hub for the system to achieve "scientific management and precise policy implementation". The module adopts a four-layer architecture of "data-rendering-interaction-decision-making". By integrating full-cycle data assets, integrating multi-dimensional visualization engines, and building a human-machine collaborative decision-making mechanism, it realizes dynamic visualization and management support for the entire life cycle of the remediation project. The module architecture is divided into a data link layer, a visualization rendering layer, an interactive service layer, and a decision support layer. Each layer collaborates to support visualization and management decision-making needs.

[0050] 1) Data Link Layer: Seamlessly connects standardized data assets output by the "Intelligent Monitoring Module" with externally extended data to generate "lightweight datasets" suitable for visualization (such as vector boundaries, raster imagery, time-series animations, and statistical charts). Furthermore, a data subscription service is provided through an API interface, enabling the visualization module to obtain data for specific ranges, time periods, and indicators on demand, enabling unified access to multi-source visualization data.

[0051] 2) Visualization Rendering Layer: Leveraging high-performance rendering technologies such as WebGL and Unity 3D, combined with GIS spatial analysis capabilities, a 3D visualization system consisting of "3D terrain + 2D thematic + timeline" is constructed. This system supports multi-granularity displays, from macro to micro, and from static to dynamic. Specifically, it includes 3D terrain visualization, 2D thematic visualization, and dynamic timeline visualization. 3D terrain visualization primarily uses oblique photogrammetry or laser point cloud data to generate a high-precision 3D real-world model of the project area. This is overlaid with information such as current land use and planning layout, enabling a "bird's-eye view" and "drill-down" experience. Clicking on a parcel reveals its current land type, remediation objectives, and project progress. 2D thematic visualization utilizes GIS overlay analysis to generate multiple thematic layers (such as a "cultivated land restoration potential map," "ecologically sensitive area distribution map," and "project progress Gantt chart"). Dynamic layer switching and overlay display are supported, such as simultaneous display of conflicting areas between "planned cultivated land" and "current construction land." The timeline dynamic visualization is based on the project timeline, integrating satellite image time series comparison, sensor time series data animation, and engineering progress animation (such as the full process simulation of "land leveling-irrigation facility construction-crop planting") to achieve "dynamic playback" and "future deduction" of the remediation process.

[0052] 3) Interactive service layer: Based on the needs of different roles such as managers, technicians, and grassroots workers, general interactive functions and role customization functions are configured to improve the usability and practicality of the visualization system. General interactive tools include intelligent annotation (such as support for manual / automatic annotation of key points, adding text, pictures, and video notes), spatial measurement (such as providing distance, area, and volume measurement tools, and automatically linking the results to the database), comparative analysis (such as support for visual comparison of multiple schemes and multiple time phases (before and after remediation)), and intelligent search (such as support for natural language search, "find cultivated land with a slope greater than 15° in the project area, and the system automatically locates and highlights eligible plots). At the same time, this module also supports role customization functions. For managers, it provides a "cockpit" view, a dynamic dashboard integrating core indicators (such as "cultivated land restoration rate", "ecological compliance rate", and "fund utilization rate"), and supports one-click generation of report PPTs; for technical personnel, it provides a "professional analysis" view, overlaying engineering drawings (such as CAD design drawings) and sensor network topology diagrams, supporting data interface debugging and model parameter adjustment; for grassroots users, it provides a "simple version" mobile interface, which allows users to intuitively view the effects of their own land remediation through a three-dimensional real-life model and supports online feedback submission.

[0053] 4) Decision support layer: This layer deeply integrates visualization and intelligent analysis to build a closed loop of "visualization of analysis conclusions, visualization of decision paths, and visualization of implementation effects" to support full-cycle management decisions.

[0054] During the remediation project planning phase, the decision-support layer supports visualization of alternative plan comparison and compliance verification. This layer displays multiple planning alternatives (e.g., "Scheme A: Reclamation Area X + Ecological Zone Y," "Scheme B: Reclamation Area M + Ecological Zone N") in a three-dimensional overlay format, with indicators such as "cultivated land reserves," "area of ​​conflict with ecological red lines," and "cost budget" labeled for each plan. Compliance verification of the planning schemes is performed through spatial overlay analysis, automatically marking conflicting areas between the plans and the project boundary, helping to quickly identify compliance risks.

[0055] During the implementation phase of remediation projects, the decision-support layer supports visualization of progress tracking and problem resolution. Project plans (such as deadlines in the BIM model) are overlaid with actual progress (such as completed areas from drone aerial surveys), with color-coded processes distinguished by "completed (green), in progress (yellow), and delayed (red)" (e.g., "land leveling is 90% complete, irrigation facilities only 50% complete"). For alerts regarding "illegal occupation of arable land" and "ecological damage," the system automatically links to historical case libraries, annotates "similar problem locations" on a 3D map (e.g., "XX project was previously penalized for temporary construction in this area"), and pushes a resolution flow chart (e.g., "issue rectification notice, demolish within 3 days, and restore arable land").

[0056] During the project acceptance phase, the decision-support layer supports visualization of results and benefit assessments. Satellite imagery, sensor data, and villager feedback from before and after the remediation phase are integrated into an "electronic archive of project results," supporting 360-degree panoramic viewing (e.g., "click on a plot to view the entire process of wasteland before remediation, during construction, and paddy fields after remediation"). Dynamic charts (e.g., "cultivated land area growth trend chart," "eco-service value enhancement bar chart") and 3D models (e.g., "rice planting simulation on newly added cultivated land") can also be used to visually demonstrate key achievements such as "cultivated land recovery rate," "ecological restoration area," and "percentage of increased farmers' income."

[0057] During the maintenance phase of remediation projects, the decision-support layer supports visualization of risk warnings and maintenance strategies. High-risk areas (e.g., "a clogged irrigation channel causing drought in farmland") are marked on a 3D map. Interactive tools recommend maintenance plans (e.g., "50,000 yuan cost to repair a channel vs. 100,000 yuan loss from abandonment") and optimal maintenance options (e.g., prioritizing repair of high-value farmland facilities).

[0058] This system efficiently integrates multi-dimensional heterogeneous data from remote sensing, the Internet of Things, business management, social perception, and other sources to build a multi-source heterogeneous land consolidation data fusion solution for a full-domain, full-cycle thematic database. At the same time, this system introduces a real-time dynamic monitoring mechanism for comprehensive land consolidation projects across the entire region. By combining time-series remote sensing image analysis with real-time data from the Internet of Things, it enables automated and high-frequency monitoring of consolidation project progress, land use changes, and ecological and environmental elements. Furthermore, this system spatially and visually displays monitoring, evaluation, and early warning results on a unified platform, and provides specific decision-making support tools, significantly improving the scientific nature and efficiency of user decision-making.

[0059] Example 2 This embodiment provides a full-life cycle dynamic monitoring system for comprehensive land management in the entire region, including: a multi-source data fusion module, an intelligent monitoring module, and a decision support module connected in sequence; The multi-source data fusion module includes: a data access layer, a cleaning and management layer, a fusion processing layer, and an application interface layer connected in sequence; the multi-source data fusion module is used to collect and fuse multi-source heterogeneous data related to the comprehensive land improvement of the entire region, and build a data base with unified spatiotemporal references and semantic associations; The intelligent monitoring module includes: a characteristic index extraction layer, a dynamic change analysis layer, an evaluation and early warning layer, and a feedback optimization layer connected in sequence; the intelligent monitoring module is used to identify land use type changes, track project progress, and dynamically monitor indicators based on the data base, using preset business rules and machine learning models, to obtain monitoring data in real time and generate early warning information; The decision support module includes: a data link layer, a visualization rendering layer, an interactive service layer and a decision support layer connected in sequence; the data link layer is connected to the dynamic change analysis layer and the assessment and early warning layer respectively; the decision support module is used to visualize the monitoring data and early warning information, and provide interactive services and decision support to users.

[0060] In the specific implementation process, Figure 2 The figure shows the overall architecture of the system provided by this embodiment. This example is based on a pilot project for comprehensive land improvement in a certain town. According to the project characteristics and monitoring needs, the comprehensive land improvement project is supervised and monitored in three stages: before improvement, during implementation, and after improvement. At the same time, the automatic land recognition technology based on deep learning algorithms is introduced to solve the key and difficult technical problems in the dynamic monitoring of cultivated land changes, and provide good technical support for the real-scene three-dimensional visualization monitoring work of the business departments.

[0061] This system adopts a hybrid database architecture optimized for land consolidation management scenarios, fully utilizing the advantages of different storage technologies, and classifies and summarizes the multi-dimensional data collected from the national land space basic information platform, remote sensing images, IoT sensors, drone aerial photography, business management systems, mobile patrol apps, Internet public opinion, etc. according to data characteristics and application scenarios. Then, it adopts DSF / HDSF distributed storage technology to disperse and store the classified data in multiple nodes, and takes advantage of its high scalability, fault tolerance and read-write performance to achieve efficient data management and rapid retrieval; Figure 3 The figure shows a schematic diagram of DSF / HDSF distributed storage technology.

[0062] This system constructs a thematic model library for comprehensive land remediation in a structured manner, including a basic library for entity-based basic data, an indicator library for entity feature indicators, a rule library for regularized operation rules, and an intelligent model for cognitive reasoning. It dynamically integrates and logically associates multi-source data to form a "one-picture" data base with "remediation projects / plots" as the core, which is used to store core business data throughout its life cycle that has been standardized, associated, and integrated.

[0063] like Figure 4 As shown in the figure, during dynamic monitoring, this system is based on literature research method, human-land coupling theory, field investigation, and expert consultation method. It focuses on agricultural land consolidation, construction land consolidation, rural ecological protection and restoration, and rural landscape improvement in land comprehensive consolidation projects and improves the monitoring efficiency and effectiveness of the entire life cycle of land comprehensive consolidation projects. It constructs a scientific, reasonable, objective and comprehensive regional land comprehensive consolidation monitoring indicator system, focusing on monitoring infrastructure conditions, natural resource conditions and land type distribution in the consolidation area, consolidation area scope, project construction area, project task completion, indicator status, and analysis of planning target achievement status.

[0064] According to the nature and objectives of the monitoring indicators, the indicator factors related to the objectives are clustered and combined at different levels according to their affiliation, thus forming a multi-objective, multi-level indicator system. This system selects 5 categories with a total of 41 indicators to construct a comprehensive improvement monitoring indicator system for the entire region. Through remote sensing image monitoring technology, 3D auxiliary technology, close-range photography monitoring technology, GNSS monitoring technology, field survey technology, automatic identification of cultivated land plots, survey statistical analysis technology, ArcGIS database technology and Fragstats analysis technology, etc., real-time monitoring of each indicator throughout the entire life cycle is carried out throughout the entire process of comprehensive land improvement project monitoring in the entire region, effectively controlling the changes in project implementation progress, time and space, quality, economic, social and ecological benefits, etc., in order to comprehensively and systematically reflect, analyze and evaluate the content of land improvement indicators, timely discover deviations and correct the development direction, so as to improve the effective implementation rate of the plan and promote the realization of monitoring goals. The comprehensive improvement monitoring indicator system for the entire region is shown in Table 1: Table 1 Summary of the indicator and technical system for comprehensive land improvement in the entire region

[0065] In order to realize the analysis of land type changes at different time points in any time dimension, this system combines the advantages of deep convolutional neural networks and uses enhanced boundary extraction methods to realize edge detection of land type plots. It can achieve efficient and accurate detection of land type plot edges and improve the ability to automatically identify different land types.

[0066] like Figure 5 As shown in the figure, in order to ensure the detection effect of small plots and eliminate the interference of irrelevant data such as roads, residential areas, and woodlands, the paddy field patches are first superimposed as input, and then the large plot data is input into the deep learning edge detection network based on pixel interpolation to predict the edges of small plots inside the large plots. Finally, according to the prediction results, image processing and computer vision algorithms are applied to gradually obtain the fused edge map, edge binary map, edge expansion map, edge refinement map and closed area map, and finally the complete small plot area and number are obtained.

[0067] In this embodiment, an improved PiDiNet model is used to implement land edge extraction; the structure of the improved PiDiNet model includes sequentially connected convolution blocks 1, 2, 3, and 4, each of which outputs feature maps of different channels, and each feature map is subjected to a unified channel conversion, and then input into an attention layer to obtain the corresponding attention features; the attention features corresponding to convolution blocks 2 to 3 are respectively subjected to feature difference, and then feature splicing is performed with the attention features corresponding to convolution block 1 to obtain splicing features; the splicing features are again subjected to channel conversion, and finally the Sigmoid function is used to obtain the detection result of the land edge; In this embodiment, in order to solve the problem of complex and diverse edge directions of land parcels, Gabor convolution layers are introduced in convolution blocks 1 and 2 of the improved PiDiNet model. This structure mainly combines Gabor filters and learnable convolution layers. Its structure is as follows: Figure 6 As shown in the figure. Gabor filters have the ability to achieve local optimization in both the spatial and frequency domains. Gabor convolutional layers efficiently and clearly extract multi-level features at different scales and directions, significantly enhancing the ability to perceive complex and diverse edges. Based on the Gabor filter's ability to extract edge features and the advantage of parameter updates during backpropagation, this model can extract significant texture features at land edges and the geometric relationship features constructed between different edges. By defining different convolution channels for feature extraction, high-dimensional features are reduced and encoded, resulting in more discernible edge features.

[0068] After obtaining the edge of the plot, the thresholding algorithm is used to make choices and decisions on the edge pixels of the plot, directly eliminating some pixels below or above a certain value; then, morphological algorithms, such as dilation and erosion algorithms, are used to connect the discontinuous edge lines to obtain better recognition results; finally, skeleton extraction and contour assembly are performed. The above-mentioned edge detection algorithm based on deep learning can detect the pixels of the contour boundary based on the differences between pixels, but it does not treat the contour as a whole, so these edge pixels need to be assembled into a contour.

[0069] In addition, in order to verify the beneficial effects of the improved PiDiNet model in this system, this embodiment also carried out a verification experiment. First, cultivated land samples were collected for the experimental data, and samples were drawn using ArcGIS software. A total of 100 image slices of 512×512 pixels were obtained, and then data enhancement operations such as flipping and rotation were performed to expand the number to 500; 300 images were randomly selected as training sets, 100 images as validation sets, and 100 images as test sets. During the training stage, each high-resolution image has two labels: cultivated land texture and cultivated land edge. The texture label is a binary image, 0 represents background, 1 represents cultivated land, and the cultivated land edge label is a single-pixel wide line. The optimized algorithm can identify most plots in the plot detection task, and the edges are roughly accurate. The geometry of the identified plots basically conforms to the original image. The detection effect is good in uniform plots and non-uniform plots, and the recognition ability is also strong in dense plots, such as Figures 7-9 shown.

[0070] All accuracy metrics for boundary detection results are calculated based on the confusion matrix. The confusion matrix represents the classification results of a statistical classification model. The vertical and horizontal columns of the matrix represent the number of pixels corresponding to the true value category and the number of pixels corresponding to the predicted result category. In the segmentation results, for land parcel classification problems, the categories are divided into positive examples (Positive) and negative examples (Negative). Correct predictions by the classifier are recorded as True (T), and incorrect predictions are recorded as Negative (F). These four basic combinations constitute the four basic elements of the confusion matrix. The recognition accuracy calculation process is as follows:

[0071]

[0072] Among them, True Positive (TP) means that the model predicts a positive example and the true value is a positive example, indicating the correctly identified boundary; False Positive (FP) means that the model predicts a positive example and the true value is a negative example, indicating the wrong part of the detection result; False Negative (FN) means that the model predicts a negative example and the true value is a negative example, indicating the unrecognized part of the real boundary; True Negative (TN) means that the model predicts a positive example and the true value is a negative example; F-score is used as a global boundary accuracy indicator; Furthermore, the calculation process of ODS (optimal dataset scale) is as follows: a fixed threshold β is selected and applied to all images to maximize the F-score of the entire dataset. At this time, β is the ODS; OIS (optimal image scale) means that a different β is selected for each image to maximize the F-score of the image. At this time, β is the OIS.

[0073] In this system, the improved algorithm showed significant improvements in the four indicators. In particular, the average accuracy of the representative boundary increased by 1.7 percentage points, and the accuracy of the plot recognition increased to 0.856, achieving accurate detection of plots in most cultivated land, as shown in Table 2: Table 2 Test accuracy comparison table

[0074] Based on this machine learning algorithm, it is possible to quickly detect changes in natural resource categories within a specific area over a specific timeframe, assisting in the monitoring and evaluation of comprehensive land remediation efforts across the entire region and enabling regular monitoring of land use changes before and after remediation. For example, by extracting changes in cultivated land within the remediation area and controlling automated camera patrols within the area for verification, the system then labels "non-grain" plots identified during monitoring and analysis (including conversion of cultivated land to forest, cultivated land to grassland, cultivated land to construction land, and cultivated land to unused land). This results in dynamic monitoring of cultivated land for grain production, providing an important reference for evaluating the effectiveness of comprehensive land remediation projects across the entire region.

[0075] This system also compares the monitoring data of each indicator with the preset threshold to achieve graded early warning and generate early warning information; after the rectification is completed, the actual rectification situation is received and compared with the early warning information, and the above-mentioned machine learning model is optimized based on the comparison results.

[0076] This system also has a decision support module, in which real-time display and monitoring of comprehensive situation, project management, dynamic monitoring, results display, system management and other functions are realized; by using the various functional modules of this system, scientific and reasonable decision support can be provided to users.

[0077] Example 3 like Figure 10 As shown, this embodiment provides a full life cycle dynamic monitoring method for comprehensive land management in the entire region, based on the system in embodiment 1 or 2, including the following steps: S1: The multi-source data fusion module collects and fuses multi-source heterogeneous data related to comprehensive land management in the entire region to build a data base with unified spatiotemporal reference and semantic association; S2: Based on the data base, the intelligent monitoring module uses preset business rules and machine learning models to identify land use type changes, track project progress, and dynamically monitor indicators, obtain monitoring data in real time, and generate early warning information; S3: The decision support module visualizes the monitoring data and warning information, and provides interactive services and decision support to users.

[0078] During the specific implementation process, the multi-source data fusion module is first used to collect and fuse multi-source heterogeneous data related to the comprehensive land improvement in the entire region, and build a data base with unified temporal and spatial benchmarks and semantic associations; then the intelligent monitoring module, based on the data base, uses preset business rules and machine learning models to realize land use type change identification, project progress tracking and dynamic monitoring of indicators, obtain monitoring data in real time, and generate early warning information; finally, the decision support module visualizes the monitoring data and early warning information, and provides interactive services and decision support to users.

[0079] This method can efficiently integrate multi-source heterogeneous data, realize dynamic monitoring of the entire process of comprehensive land improvement projects in the entire region, intelligent effectiveness evaluation and risk warning, and provide strong support for management decisions.

[0080] The same or similar reference numerals correspond to the same or similar components; The terms used in the drawings to describe positional relationships are for illustrative purposes only and are not to be construed as limiting the present application. Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the claims of the present invention.

Claims

1. A full life cycle dynamic monitoring system for comprehensive land management in the entire region, characterized by: It includes: a multi-source data fusion module, an intelligent monitoring module and a decision support module connected in sequence; The multi-source data fusion module includes: a data access layer, a cleaning and management layer, a fusion processing layer, and an application interface layer connected in sequence; the multi-source data fusion module is used to collect and fuse multi-source heterogeneous data related to the comprehensive land improvement of the entire region, and build a data base with unified spatiotemporal references and semantic associations; The intelligent monitoring module includes: a characteristic index extraction layer, a dynamic change analysis layer, an evaluation and early warning layer, and a feedback optimization layer connected in sequence; the intelligent monitoring module is used to identify land use type changes, track project progress, and dynamically monitor indicators based on the data base, using preset business rules and machine learning models, to obtain monitoring data in real time and generate early warning information; The decision support module includes: a data link layer, a visualization rendering layer, an interactive service layer and a decision support layer connected in sequence; the data link layer is connected to the dynamic change analysis layer and the assessment and warning layer respectively; The decision support module is used to visualize the monitoring data and early warning information, and provide interactive services and decision support to users.

2. A full life cycle dynamic monitoring system for comprehensive land management according to claim 1, characterized in that: In the multi-source data fusion module, The data collected by the data access layer includes at least one or more of: satellite remote sensing images, drone aerial survey data, ground sensor data, business database and document table data, and any one or more of the publicly available data on the Internet; the data collected by the data access layer is classified, and the classified data is stored in multiple nodes using DSF / HDSF distributed storage technology; The cleaning management layer is used to clean the data collected by the data access layer based on predefined cleaning rules. The data cleaning includes at least one or more of: outlier marking and correction, missing value interpolation, and duplicate value removal; The fusion processing layer is used to sequentially perform spatiotemporal benchmark unification, semantic association modeling, and business scenario adaptation processing on the cleaned data to generate the data backplane; the spatiotemporal benchmark unification includes unifying and correcting the time and space coordinates of multi-source heterogeneous data; the semantic association modeling includes constructing a knowledge graph based on the spatiotemporal benchmark unified data and performing entity semantic association; the business scenario adaptation includes dynamically packaging the semantic association modeled data into several data sets adapted to different stages based on the data differences required for different stages of comprehensive land improvement in the entire region, and saving the data sets corresponding to all stages together as the data backplane; The application interface layer is used to output the data backplane to the intelligent monitoring module through a preset multi-element interface.

3. The full life cycle dynamic monitoring system for comprehensive land management according to claim 1 is characterized in that: In the intelligent monitoring module, The characteristic index extraction layer is used to extract a number of monitoring indicators from the data base and perform characteristic quantification processing based on the pre-built comprehensive land improvement monitoring indicator system for the entire region; The dynamic change analysis layer is used to automatically identify land parcels based on a preset machine learning model, further identify changes in land use types, and extract change data; at the same time, it implements project progress tracking and dynamic indicator monitoring based on preset business rules, and obtains monitoring data in real time; The evaluation and warning layer is used to compare the monitoring data of each indicator with the preset threshold value, implement hierarchical warning, and generate warning information; The feedback optimization layer is used to receive the actual rectification situation and compare it with the early warning information, and optimize the machine learning model and business rules based on the comparison results.

4. A full life cycle dynamic monitoring system for comprehensive land management according to claim 3, characterized in that: In the characteristic index extraction layer, a comprehensive land improvement monitoring index system is constructed based on the hierarchical analysis method. The index system includes at least: a target layer, a criterion layer and an index layer; The target layer includes: basic bottom-line constraints, improvement of farmland ecosystem service functions, ecological protection and restoration and improvement of rural landscape, project and fund coordination and project supervision; At the criteria level, basic bottom-line constraints include: quantity control, quality control, and other controls; improving farmland ecosystem service functions includes: the degree of farmland concentration and contiguousness, the degree of intensive and economical use of construction land, and land transfer; ecological protection and restoration and rural landscape improvement include: ecological protection and restoration and rural landscape improvement; project and funding coordination includes: funding guarantees for improvement, project completion progress, and project quality targets; and project supervision includes: innovation and illegal monitoring. In the indicator layer, quantity control includes: the proportion of newly added cultivated land area, the proportion of newly added permanent basic farmland area and the surplus construction land index; quality control includes: improving the quality grade of cultivated land; other controls include: the area that conflicts with the ecological protection red line and the area of ​​historical context protection; the degree of farmland concentration and contiguousness includes: the area of ​​newly added high-standard farmland, the area of ​​developed and supplemented cultivated land, the area of ​​reclaimed paddy fields, the number of cultivated land small fields converted into large fields and the degree of increase in the concentration and contiguousness of permanent basic farmland; the degree of intensive and economical use of construction land includes: the proportion of increase and decrease linked indicators used for infrastructure construction, the scale of redevelopment of inefficient construction land, the scale of demolition and reclamation of rural construction land, the rate of decrease in the average household homestead area and the scale of reduction in village construction land; land transfer status includes: the area of ​​cultivated land for grain cultivation after remediation, the introduction of agricultural enterprises, the number of large grain-growing households and the proportion of land management rights transfer; ecological The protection and restoration situation includes: the area of ​​comprehensive mine management, the area of ​​comprehensive water environment management, the area of ​​forestland transformation, the area of ​​mangrove protection and restoration, the area of ​​coastal management, the number of kilometers of newly added ecological corridors and the soil and water conservation rate; the rural landscape improvement situation includes: the number of rural human settlement environment improvements, the coverage rate of sewage treatment facilities, the coverage rate of garbage treatment and the penetration rate of sanitary toilets; the improvement funding guarantee situation includes: the proportion of integrated agricultural funds, the completion rate of budget investment funds, the actual expenditure rate of funds and the proportion of social capital investment; the project completion progress includes: the completion rate of agricultural land improvement projects, the completion rate of construction land improvement projects, the completion rate of ecological protection and restoration projects, the completion rate of cultural protection and rural landscape improvement and other projects; the project quality target situation includes: the project quality qualification rate; innovation and illegal monitoring includes: the number of innovative systems and the area of ​​illegal land use.

5. The full life cycle dynamic monitoring system for comprehensive land management according to claim 3 is characterized in that: In the dynamic change analysis layer, the preset machine learning model is specifically an improved PiDiNet model, which is used for land edge extraction; The structure of the improved PiDiNet model includes sequentially connected convolution blocks 1, 2, 3, and 4. Each convolution block outputs feature maps of different channels. After each feature map is uniformly converted into a channel, it is input into an attention layer to obtain the corresponding attention features. After the attention features corresponding to convolution blocks 2 and 3 are respectively subjected to feature difference, they are feature spliced ​​with the attention features corresponding to convolution block 1 to obtain spliced ​​features. The spliced ​​features are again subjected to channel conversion, and finally the Sigmoid function is used to obtain the detection result of the land edge. The convolution block 1 and the convolution block 2 have the same structure, both including a Gabor convolution layer, a batch normalization layer, a maximum pooling layer and an activation layer connected in sequence; the convolution block 3 and the convolution block 4 have the same structure, both including a convolution layer, a batch normalization layer, a maximum pooling layer and an activation layer connected in sequence.

6. A full life cycle dynamic monitoring system for comprehensive land management in the entire region according to claim 5, characterized in that: The Gabor convolution layer includes at least one Gabor filter and a learnable convolution kernel.

7. The full life cycle dynamic monitoring system for comprehensive land management according to claim 1 is characterized in that: In the decision support module, The data link layer is used to collect the monitoring data and early warning information in real time; The visualization rendering layer is used to realize three-dimensional terrain visualization, two-dimensional thematic visualization and timeline dynamic visualization based on monitoring data and warning information; The three-dimensional terrain visualization includes: constructing a three-dimensional real-life model of the entire land comprehensive improvement project area, overlaying monitoring data and early warning information for each plot to enable three-dimensional information browsing; the two-dimensional thematic visualization overlays and analyzes monitoring data and early warning information through an external GIS system to generate multi-thematic layers; the timeline dynamic visualization uses the timeline of the entire land comprehensive improvement project as the axis, combining monitoring data and early warning information to enable dynamic playback and future deduction of the improvement process; The interactive service layer is used to provide interactive services for users, and the interactive services include general services and customized services; The decision support layer is used to provide decision support for users.

8. The full life cycle dynamic monitoring system for comprehensive land management according to claim 7 is characterized in that: In the visualization rendering layer, a three-dimensional real-scene model of the entire land comprehensive improvement project area is constructed based on the WebGL or Unity 3D engine.

9. The full life cycle dynamic monitoring system for comprehensive land management according to claim 7 is characterized in that: In the interactive service layer, general services include: data intelligent annotation, spatial measurement, comparative analysis and intelligent search; The customized services include: providing managers with a cockpit view and a dynamic dashboard integrating core indicators; providing technical personnel with a professional analysis view, overlaying engineering drawings and sensor network topology diagrams for data interface debugging and model parameter adjustment; and providing grassroots users with a simplified interface to intuitively view the improvement effects of their own land through a three-dimensional real-life model, and providing online feedback functions.

10. A full life cycle dynamic monitoring method for comprehensive land management in an entire region, based on the system described in any one of claims 1 to 9, characterized in that: The following steps are involved: S1: The multi-source data fusion module collects and fuses multi-source heterogeneous data related to comprehensive land management in the entire region to build a data base with unified spatiotemporal reference and semantic association; S2: Based on the data base, the intelligent monitoring module uses preset business rules and machine learning models to identify land use type changes, track project progress, and dynamically monitor indicators, obtain monitoring data in real time, and generate early warning information; S3: The decision support module visualizes the monitoring data and warning information, and provides interactive services and decision support to users.

Images