Landform surveying method and system for territorial space planning

Abstract

The invention relates to the technical field of geographic information measurement, in particular to a landform surveying method and system for territorial space planning, and the method comprises the steps: 1, carrying out the fusion preprocessing of multi-source data; 2, constructing a planning constraint dynamic rule base; 3, space-time constraint coupling analysis: taking the digital elevation model as a reference, and correcting the space offset of the land cover map through a thin-plate spline function; comparing the planning threshold value with the current landform parameter in real time, and marking a conflict coordinate set of gradient overrun or settlement overspeed; 4, intelligently segmenting the landform units; and 5, planning suitability output: setting a dynamic buffer distance according to the fault activity grade and the rock-soil body type, and generating an early warning layer when a town unit exists in a buffer area. According to the method, remote sensing images, laser radar point clouds and geological survey data are fused, space-time dislocation among various data sources is eliminated, and the precision of space analysis is improved.

Description

Technical Field

[0001] This invention relates to the field of geographic information measurement technology, and in particular to a geomorphological surveying method and system for land spatial planning. Background Technology

[0002] As a guide for national spatial development, territorial spatial planning requires the precise integration of geomorphological survey data in decision-making processes such as the delineation of ecological protection red lines and the optimization of urban development boundaries. Current mainstream surveying methods suffer from the following technical deficiencies: 1. Inaccurate fusion of multi-source heterogeneous data leads to spatial misalignment; Due to inherent differences in acquisition principles, spatiotemporal references, and resolutions between remote sensing imagery, lidar point clouds, and geological survey data, traditional methods employ an independent processing mode. When optical images are used to generate land cover maps through supervised classification, their pixel coordinate system is not forcibly aligned with the coordinate system of the elevation model generated from the terrain point cloud; The overlay of geological fault line vector data and rasterized topographic data relies solely on simple spatial overlay analysis.

[0003] Technical Consequences: Steep slope areas (slope > 25°) were misclassified as urban construction land due to their similar image spectral characteristics to built-up areas, leading to incorrect delineation of ecological protection zones. The root cause of this problem lies in the differences in sensor physical characteristics and the disconnect in data processing workflows, resulting in an error rate exceeding 20% ​​in complex mountainous areas.

[0004] 2. The planning constraints are disconnected from the surveying logic; Existing surveying systems treat land and space planning texts as static auxiliary information: Key constraint parameters such as slope thresholds for ecological protection zones and land subsidence rates in urban construction areas need to be manually entered into the system and cannot be automatically updated with revisions to planning documents. Parameter comparison is only performed after the geomorphological analysis is completed, which means that areas with slopes exceeding the limit need to be repeatedly remeasured.

[0005] In a provincial land planning case, 17% of conflict areas went unidentified because the revised prohibition clause on steep slopes was not imported in a timely manner. This deficiency stemmed from the lack of an interaction mechanism between the semantics of the planning text and unstructured geomorphological data.

[0006] Therefore, there is an urgent need for a geomorphological surveying method and system for land spatial planning to solve the above problems. Summary of the Invention

[0007] To achieve the above objectives, the present invention provides a geomorphological survey method for land spatial planning, comprising the following steps: Step 1: Multi-source data fusion preprocessing: Simultaneously acquire remote sensing optical images, lidar point clouds, and historical geological survey data of the target area; Atmospheric correction is applied to remote sensing images to generate land cover maps, and building density in built-up areas and leaf area index in vegetation areas are dynamically identified. Ground filtering is applied to point cloud data to generate a digital elevation model and to extract slope, aspect, and topographic relief. Step 2: Constructing the dynamic rule base for planning constraints: Analyze the semantics of land and space planning texts to extract slope thresholds for ecological protection zones, subsidence rate thresholds for urban construction areas, and soil erosion thresholds for farmland areas; Convert text rules into machine-executable code and establish a mapping relationship between rule identifiers and geographic coordinates; Step 3: Spatiotemporal constraint coupling analysis: Based on the digital elevation model, the spatial offset of the land cover map is corrected by thin plate spline function; Real-time comparison of planning thresholds with current geomorphological parameters, marking conflict coordinate sets of slope exceeding limits or settlement exceeding speed; Step 4: Intelligent segmentation of terrain units: Initial units are generated based on the watershed watershed, with the conflict coordinates as the center. The consistency index of landform features within the calculation unit is generated by dynamically weighting the ratio of slope variability to the planned slope threshold, the deviation of vegetation cover variability from the planned cover threshold, and the density of potential geological hazards. When the index exceeds the limit, the boundary is redrawn along the ridgeline / valleyline; Step 5: Plan the adaptive output: Generate a 3D survey map that integrates a digital elevation model, conflict markers, and unit boundaries; A dynamic buffer distance is set based on the fault activity level and the type of soil and rock mass. When urban units are present in the buffer zone, an early warning layer is generated.

[0008] Preferably, the dynamic identification of building density in step 1 includes: Construct a convolutional neural network model, with the input being preprocessed remote sensing images and the output being the polygonal vector boundary of building roofs; Based on the spatial distribution ratio of urban development boundaries and ecological protection zones in the national land spatial planning, a differentiated grid division strategy is determined: a: Within the urban development boundary, a fine grid size that is positively correlated with the proportion of "high-density construction area" in the plan is adopted; b: Within the ecological protection zone, a sparse grid size that is negatively correlated with the area ratio of the "core protection zone" is adopted; The ratio of the total area of ​​the roof polygons within each grid to the grid area is used as the building density. When the grid spans different functional zones, the mixed density is calculated by area weighting.

[0009] Preferably, the extraction of the slope threshold in step 2 includes: Natural language processing technology was used to identify slope restriction clauses in planning texts, and the numerical value X in the keyword "development is prohibited if the slope is greater than X degrees" was extracted as the basic threshold. Link to historical geological disaster databases and compile a histogram of actual slope distribution at landslide event locations within the target area; The final threshold is the weighted average of the basic threshold and the median slope of the landslide concentration area. The weighting principle is as follows: a: When the number of landslide events exceeds the historical average, increase the weight of the median landslide slope; b: When modifiers such as "strict protection" appear in the planning text, increase the weight of the basic threshold.

[0010] Preferably, the implementation method of thin-plate spline function correction in step 3 includes: Adaptive control point layout: The density of control points is determined based on the topographic complexity index, which is calculated by multiplying the standard deviation of topographic relief of the digital elevation model by the slope change rate. Offset vector field construction: Using the deviation between the coordinates of the land cover map and the coordinates of the terrain raster at the control point as the observation value, a global offset vector field is generated by kriging interpolation; Minimize the energy function: Define the bending energy function of the thin plate spline function, add the coordinate weight factor of the planning conflict region, and use the conjugate gradient method to iteratively solve the spatial transformation parameters.

[0011] Preferably, the calculation of the geomorphic feature consistency index within the unit in step 4 includes: Calculation of the density of potential geological hazards: Taking the center of the unit as the center, the statistical range is determined according to the influence radius of historical landslides, and the number of potential hazards within the unit area is calculated; Weighting strategy: a: When the density of potential hazard points exceeds the threshold of the historical hazard level of the area, the slope variability weight will be increased to the set multiple of the vegetation cover variability weight. b: When the unit is located within the planned ecological protection red line, the vegetation cover variability weight is set to the set multiple of the slope variability weight; The set multiple has a piecewise linear relationship with the density of hidden danger points or the red line protection level.

[0012] Preferably, the rules for redrawing the boundary in step 4 include: When the index exceeds the limit, the priority of redrawing along the ridgeline is higher than that along the valleyline, and the priority ranking is based on: a: Spatial overlap between the calculation unit boundary and the ridgeline / valleyline; b: Assess the improvement rate of geomorphic feature consistency index of the new unit after rezoning; Candidate boundary lines are generated using the Deloni triangulation technique, and the boundary that maximizes the decrease in the new unit index is selected as the final dividing line. The iteration terminates when all unit indices do not exceed an adaptive threshold, which is dynamically set based on the statistical distribution characteristics of the initial segmentation unit indices.

[0013] Preferably, the setting of the dynamic buffer distance in step 5 includes: Basic buffer distance determined: a: According to the fault activity classification table, the basic distance of active faults is set as a multiple of the set distance of dormant faults; b: When the soil / rock type code indicates a loose sedimentary layer, increase the base distance by an extension distance that is positively correlated with the thickness of the sedimentary layer; Correction for groundwater rate of change: a: Obtain time-series data of groundwater levels from monitoring wells in the fault zone and calculate the standard deviation of the interannual variation rate; b: When the standard deviation exceeds the regional safe fluctuation range, the buffer distance is increased in a linear proportion to the rate of change and the base distance.

[0014] Preferably, the conditions for generating the warning layer include: Red alert trigger conditions: a: The center point of the urban construction unit falls within the dynamic buffer zone of the fault. b: The density of the unit building exceeds the upper limit value of this functional area in the planning constraint library; c: The proportion of the year-on-year decrease in groundwater depth within the unit exceeds the historical extreme value; Quantification of early warning levels: a: A yellow alert is triggered when two of the above conditions for triggering a red alert are met; b: A red alert is triggered when all three red alert triggering conditions are met and the unit is located within the influence range of the historical ground fissure zone.

[0015] Accordingly, embodiments of the present invention also provide a geomorphological survey system for land spatial planning, used to run any of the geomorphological survey methods for land spatial planning described in the embodiments of the present invention, including: The multi-source data acquisition and fusion module includes: The remote sensing image processing unit is configured to perform atmospheric correction on optical images and output a land cover type distribution map; The point cloud filtering and reconstruction unit is connected to the lidar sensor and configured to generate a digital elevation model and extract slope and aspect. The planning constraint dynamic rule base module includes: The semantic parsing unit takes land spatial planning text as input and outputs machine-executable code for slope threshold and settlement rate threshold. Spatial mapping unit stores a table of mapping relationships between rule identifiers and geographic coordinates; The spatiotemporal coupling analysis engine module, whose input is connected to the multi-source data acquisition and fusion module and the planning constraint dynamic rule base module, includes: The spatial correction unit receives the digital elevation model and land cover map, and corrects spatial offsets using thin-plate spline functions. The conflict marker unit compares the geomorphic parameters with the planning threshold in real time and outputs a set of conflict coordinates. The intelligent landform unit segmentation module, whose input is connected to the spatiotemporal coupling analysis engine module, includes: The watershed initialization unit generates watershed boundaries based on the conflict coordinate set; The index calculation unit is configured to dynamically weight the geomorphological feature consistency index by combining the density of geological hazard hazard points; Boundary optimization unit: When the index exceeds the limit, the unit boundary is redrawn along the ridgeline / valleyline. The planning and adaptation output module, whose input is connected to the intelligent landform unit segmentation module, includes: The 3D rendering unit integrates digital elevation models, conflict markers, and unit boundaries to generate survey maps. The dynamic buffer early warning unit sets the buffer distance and outputs an early warning layer based on the fault activity level and soil type.

[0016] Preferably, the dynamic buffer early warning unit further includes: Buffer distance setting sub-unit, with built-in fault activity classification table and soil type code library; It also includes an early warning response module, whose input is connected to a dynamic buffer early warning unit, including: The conditional decision sub-unit is configured to check whether each urban construction unit meets the following criteria: a. The cell center is located within the dynamic buffer; b. Building density exceeds the planned upper limit; c. The proportion of groundwater depth decreasing year-on-year beyond historical extremes; The hierarchical alarm subunit is configured as follows: a. When the above two conditions are met, a yellow alert is triggered and the yellow alarm device is activated; b. When all conditions are met and the unit is located in a historical ground fissure zone, the red alarm device and graphic display unit are activated.

[0017] The beneficial effects of this invention are: 1. This invention employs multi-source data fusion preprocessing, utilizing atmospheric correction and ground filtering techniques to achieve precise spatial registration of remote sensing imagery and lidar point clouds. In particular, it corrects spatial offsets in land cover maps using thin-plate spline functions and leverages control point density and offset vector fields for spatial correction, effectively eliminating spatiotemporal misalignment between different data sources. This process ensures that all data sources are processed under the same spatial reference, thereby reducing misjudgments caused by data misalignment. Especially in complex mountainous areas, it minimizes errors due to spatial misalignment, ensuring the accuracy and reliability of geomorphological data.

[0018] 2. This invention constructs a dynamic rule base for planning constraints, automatically converting key constraints in land spatial planning texts into machine-executable code. It extracts constraint parameters such as slope and settlement rate from the text in real time and interfaces them with geomorphological data. By analyzing the planning text using natural language processing technology, necessary thresholds are extracted and the rule base is automatically updated, ensuring that the system can identify and apply new planning constraints in real time after each text revision. This method not only improves the flexibility and real-time nature of planning but also avoids the problem of constraint invalidation due to human input errors or update delays, ensuring the accuracy and timeliness of planning decisions.

[0019] 3. This invention utilizes spatiotemporal constraint coupling analysis to compare the consistency between planning constraints and current geomorphic parameters in real time. Potential conflict areas are identified in the early stages of data analysis, eliminating the need for secondary detection after geomorphic analysis is complete. In particular, the use of thin-plate spline functions for spatial correction of land cover maps and digital elevation models, along with real-time comparisons at each stage, effectively prevents geomorphic data from becoming disconnected from planning constraints. Once conflict areas are identified, the system can directly provide feedback and mark them, significantly reducing the time and resource waste caused by repeated measurements and improving surveying efficiency and accuracy.

[0020] 4. This invention introduces intelligent geomorphic unit segmentation technology. By calculating the geomorphic feature consistency index and combining multiple factors such as slope, vegetation cover, and geological hazard risks, geomorphic units are finely divided. Through a dynamic weighting strategy, the system automatically adjusts the segmentation rules under specific environmental conditions to ensure the consistency of geomorphic features of each unit with planning requirements. In particular, when the index exceeds the limit, the system automatically redraws the boundary along the ridgeline or valley line, ensuring that the natural features of the planned area are highly consistent with the actual geomorphic conditions, reducing the planning irrationality caused by changes in geomorphic features in traditional methods. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of the steps of the method of the present invention; Figure 2 This is a flowchart illustrating the implementation steps of the thin-plate spline function correction described in step 3 of the method of the present invention. Figure 3 This is a structural block diagram of the system of the present invention. Detailed Implementation

[0023] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0024] See Figure 1 Figure 3 This invention provides a geomorphological survey method for land spatial planning. In step 2, remote sensing technology, lidar technology, and historical geological survey data are used to acquire remote sensing optical images, lidar point cloud data, and historical geological survey data of the target area. The remote sensing images mainly provide land cover information, the lidar point cloud data is used to generate a digital elevation model, and the geological survey data provides underground and surface geological features.

[0025] Atmospheric correction algorithms (such as FLAASH or QUAC) are used to eliminate atmospheric effects and restore the surface reflectance of remotely sensed images to their true values. Furthermore, a land cover map is generated using supervised classification methods, and building density in built-up areas and leaf area index in vegetated areas are dynamically identified. This information is helpful for subsequent identification of ecological protection zones and urban construction areas.

[0026] Ground filtering techniques are used to remove non-ground points from lidar point cloud data, generating an accurate digital elevation model (DEM). Slope, aspect, and topographic relief are extracted from the DEM, and these topographic features can provide important references for planning decisions.

[0027] In step 2, the constraints in the land and space planning text are converted into executable rules by parsing the text: Using natural language processing (NLP) technology, semantic analysis was performed on land spatial planning texts to extract key constraint parameters such as slope threshold for ecological protection areas, subsidence rate threshold for urban construction areas, and soil erosion threshold for farmland areas.

[0028] The constraints extracted from the text are converted into machine-executable code, and a mapping relationship is established between geographic coordinates and rule identifiers. This ensures that the system can automatically compare the constraints with those in the planning text when performing topographic analysis.

[0029] Step 2 achieves precise analysis for planning decisions by comparing topographic data with planning constraints in real time: Thin-plate spline functions are used to spatially correct land cover maps, ensuring that remote sensing imagery and lidar point cloud data are accurately aligned in the same spatial coordinate system. This step resolves the spatial misalignment problem caused by different data sources.

[0030] By comparing parameters such as slope and aspect in the digital elevation model with planning thresholds extracted from the land spatial planning text in real time, areas with excessive slope or settlement rate are marked, forming a conflict coordinate set. This process can promptly identify potential planning conflict areas and avoid problems after the plan is implemented.

[0031] Step 4 refines the landform units using intelligent algorithms to improve the accuracy of landform analysis: Using conflict coordinates as the center, preliminary geomorphic units are generated based on the watershed method. The watershed method can divide a region into multiple river system units, each with unique topographic features.

[0032] The geomorphic feature consistency index for each unit is calculated, and dynamically weighted by combining slope variability, vegetation cover variability, and density of potential geological hazards. If the index exceeds the standard, it indicates that there is a potential risk in the planning of that unit, and the boundary needs to be redefined.

[0033] When the index exceeds the limit, the system automatically redraws the boundary along the ridgeline or valley line to ensure consistency between the landform features and planning requirements.

[0034] Step 5 generates comprehensive planning adaptability outputs to help decision-makers with spatial planning: By integrating digital elevation models, conflict markers, and geomorphic unit boundaries, an intuitive 3D survey map is generated. This map clearly displays the geomorphic features, planning constraints, and potential conflicts of the planning area, providing visual support for planning decisions.

[0035] Dynamic buffer zones are set based on fault activity levels and soil and rock types. When urban units are present within the buffer zone, an early warning layer is generated to alert planning decision-makers to potential geological hazard risks or construction safety issues.

[0036] In one possible implementation, a convolutional neural network (CNN) model is first constructed, taking preprocessed remote sensing image data as input. The CNN model can automatically extract features from the image, thereby identifying the rooftops of buildings within the image. By training the network, the output will be the polygonal vector boundary of each building's roof. To improve recognition accuracy, classic deep learning methods, such as U-Net or Mask R-CNN, can be used for semantic segmentation or instance segmentation to obtain the precise outline of each building.

[0037] Based on the urban construction area information in the planning document, a finer grid size that is positively correlated with the proportion of "high-density construction areas" is selected within the urban development area. This means that in the urban development area, areas with high building density will use a finer grid division to capture building distribution and density changes with higher spatial resolution.

[0038] Within the ecological protection zone, a sparse grid size negatively correlated with the area of ​​the "core protection zone" is adopted. Because ecological protection zones require the protection of the ecological environment and the reduction of human activities, the grid division in this area will be relatively lenient, with a larger grid size, to reduce unnecessary detail capture and meet regional protection requirements.

[0039] Within each grid, the total area of ​​the identified building roof polygons is calculated, and then the ratio is calculated to the grid area. This ratio is the building density.

[0040] For grids that span different functional zones, density calculations will be weighted according to the functional nature of each zone. For example, if a grid spans both an urban development zone and an ecological protection zone, the building density of the grid within each functional zone needs to be calculated separately, and then weighted according to the area ratio to obtain the final mixed density.

[0041] When a grid spans multiple different functional zones, building density is weighted according to the area proportion of each zone. For example, if a grid is partly located within the urban development boundary and partly within an ecological protection zone, the average density is calculated by weighting the proportion of each part. This weighted calculation ensures that the planning constraints and characteristics of different functional zones are reasonably reflected in the building density analysis.

[0042] By combining convolutional neural network technology with a differentiated grid partitioning strategy, an efficient and accurate method for identifying and calculating building density is provided, which not only meets the requirements of modern land spatial planning, but also provides strong data support for subsequent geomorphological analysis and planning decisions.

[0043] In one possible implementation, the land and space planning text (PDF, Word, TXT, etc.) is converted into a structured plain text format, which is then segmented into paragraphs and sentences to facilitate subsequent processing.

[0044] Using BERT or RoBERTa-like pre-trained Chinese language models, perform non-entity recognition (NER) on planning texts, paying particular attention to sentences containing keywords such as "slope", "greater than", "prohibited from development", and "not allowed to be built".

[0045] Use dependency parsing-based methods (such as LTP or HanLP dependency parsing trees) to extract the numerical value X from the sentence, locate and extract X in "slope greater than X degrees prohibits development".

[0046] The extracted value X is used as the basic slope threshold for the next weighted calculation.

[0047] Import historical geological disaster databases, including location information (latitude and longitude, time of occurrence, scale of landslides, etc.) for landslides and debris flows within the target area.

[0048] In a GIS environment (ArcGIS / QGIS / PyGDAL), historical landslide locations are overlaid with DEM data (5m / 10m resolution), and the slope value at the corresponding location of each landslide point is calculated.

[0049] Draw a histogram of slope distribution at the location of the landslide event.

[0050] Calculate the median slope value of a landslide event. As a typical value reflecting the slope range with high landslide incidence, it is used for safety threshold calibration.

[0051] Let: Base threshold (extracted from text): Median landslide slope: .

[0052] When the number of landslide events in the target area is higher than the historical average: Increase the median slope of the landslide The weights are assigned to reflect the priority of geological disaster risks. Increased to 0.6, =0.4.

[0053] When phrases such as "strictly protected" or "development strictly prohibited" appear in planning documents: Increase the basic threshold for text The weights are set to reflect the strict principle of planning. =0.7, =0.3.

[0054] Final calculation formula: .in: The final determined slope threshold is used to identify high-slope areas that are prohibited from development or require key protection during the geomorphological survey process for land spatial planning.

[0055] In one possible implementation, the target area is processed using digital elevation model (DEM) data to calculate the standard deviation of terrain relief (i.e., the degree of surface variation of the terrain) and the rate of change of slope (i.e., the rate at which the slope changes with space).

[0056] The standard deviation formula is used to statistically analyze the elevation differences in DEM data, reflecting the degree of elevation fluctuation within the region.

[0057] The slope of each grid cell in the DEM data is calculated (using methods such as trigonometric functions and gradient calculation), and the rate of change of slope between adjacent grid cells is further calculated.

[0058] Multiplying the two results in an index that reflects the complexity and variation of the landform. A higher index value indicates more complex topographic changes in the area, requiring more control points for correction.

[0059] The density of control points is automatically adjusted based on the geomorphic complexity index. More control points are deployed in areas of high geomorphic complexity, while fewer control points are deployed in areas of low complexity. This method optimizes computational resources and improves correction efficiency.

[0060] At the established control points, the deviation is calculated based on the known coordinates of the land cover map and the topographic raster coordinates. The land cover map coordinates are typically based on actual ground coordinates obtained from aerial photography, satellite remote sensing, etc., while the topographic raster coordinates are virtual coordinates obtained through DEM data processing. The deviation is defined as the difference between the land cover map coordinates and the raster coordinates.

[0061] The calculated control point deviations were spatially interpolated using the Kriging interpolation method. Kriging is a spatial statistical method that estimates the distribution of deviations in areas where no control points were established based on the spatial correlation between deviation values.

[0062] Generate a global offset vector field by interpolating the offset information (e.g., offset amount, direction) of all regions to obtain a continuous vector field with uniform energy distribution.

[0063] Thin plate splines are a type of smooth function commonly used for spatial transformations. They can approximate the spatial transformation relationship between control points by minimizing the bending energy of the surface.

[0064] Its energy function is defined as: ; in, Let be the transformation function to be solved. These are weighting coefficients. These are the target coordinates. It is a regularization factor.

[0065] During the calculation of thin-plate spline transformations, coordinate weighting factors are added for areas with planning conflicts (such as prohibited development zones and protected areas). This ensures that changes in these areas are given priority during the correction process, avoiding discrepancies between the planned area and the actual terrain.

[0066] The conjugate gradient method is used for iterative solution to minimize the energy of the thin-plate spline function. The conjugate gradient method is a commonly used optimization algorithm that iteratively updates the transformation parameters, ensuring each update follows the optimization path until the energy function converges to its minimum, thus obtaining the optimal spatial transformation parameters.

[0067] In one possible implementation, within the territorial spatial planning area, the center point of each calculation unit is first determined, which is typically located at the geometric center of the planning unit.

[0068] The statistical range is set with the center of the unit as the center, based on the historical landslide impact radius. The landslide impact radius is the range of impact calculated based on historical landslide event data within the region, which may take into account factors such as the magnitude of the landslide impact and historical disaster data.

[0069] Within the defined statistical scope, all known potential geological hazards (such as landslides, debris flows, collapses, etc.) are collected and marked, and confirmed through remote sensing data, field surveys, or historical records.

[0070] The number of potential hazards is counted, and the density of potential hazards per unit area (unit: hazard points / square kilometer) is calculated, that is, the number of potential hazards per square kilometer within a specified range.

[0071] If the calculated density of potential hazards exceeds a certain historical hazard level threshold (for example, a high density of potential hazards may be considered a high-risk area), the geomorphological consistency calculation for that area will take into account the more significant impact of slope variability.

[0072] At this point, the weight of slope variability is increased to a multiple of the weight of vegetation cover variability. This strategy emphasizes the impact of topography (slope) on geological hazard risk, reflecting the potential for a higher risk of geological hazards in the area.

[0073] If the unit is located within a planned ecological protection red line area, the importance of vegetation cover variability increases. Protection red lines are typically established to maintain the ecological environment and prevent destructive development; therefore, special attention needs to be paid to the protection and restoration of vegetation within these areas.

[0074] In this case, the weight of vegetation cover variability will be set to a multiple of the weight of slope variability, thereby increasing the importance of vegetation characteristics in the consistency assessment.

[0075] The change in the set multiplier has a piecewise linear relationship based on the density of potential hazards or the red line protection level. Specifically, the weight can be adjusted in segments according to different potential hazard density ranges (such as low, medium, and high density ranges) or different red line protection levels (such as level one, level two, and level three protection).

[0076] For example, when the density of potential hazard points is in the low-density range, the weight of slope variability may only be 1 times that of vegetation cover variability; while when the density enters the high-density range, the weight of slope variability may increase to 3 times that of vegetation cover variability. Similarly, the weight of the protection level of the red line protection area is also increased or decreased in stages according to the protection requirements.

[0077] In one possible implementation, the type of boundary to be prioritized for redrawing is determined by calculating the degree of overlap between the boundary of the planning unit and the ridgeline or valleyline.

[0078] Ridge lines typically represent the highest points in the terrain, signifying watersheds; river valley lines, on the other hand, mark the low-lying areas along rivers. Because topography significantly impacts landform stability, these features are given priority consideration during rezoning.

[0079] Methods for calculating spatial overlap include using GIS technology and spatial analysis algorithms to calculate the intersection area between the unit boundary and the ridgeline or valley line. A high overlap indicates a strong correlation between the boundary and the ridgeline or valley line, which may mean that the boundary area needs to be prioritized for rezoning.

[0080] For each candidate redrawn boundary, the geomorphic feature consistency index of the redrawn new unit is first evaluated. This index measures the overall consistency of the geomorphic features of the new unit, that is, the comprehensive performance of features such as geological hazard risks, vegetation coverage, and slope changes.

[0081] The improvement rate refers to the degree of improvement in the geomorphic feature consistency index of the new unit after rezoning compared to before rezoning. The calculation method is as follows: ; This indicator is used to measure whether rezoning can effectively improve the consistency of regional geomorphological features, thereby enhancing the rationality and scientific nature of the planning.

[0082] Deloni triangulation is a spatial partitioning method commonly used in Geographic Information Systems (GIS). It generates a series of triangles from a set of points and ensures that no other point lies inside the circumcircle of any triangle, thus guaranteeing the "optimality" of the partition.

[0083] In this method, the technique is used to generate candidate new boundary lines on the boundaries of computational cells. Each boundary line is selected based on its maximum effect on improving the geomorphic feature consistency index.

[0084] After generating candidate boundary lines, the boundary that maximizes the reduction in the geomorphic feature consistency index of the unit is selected as the final dividing line. This selection is based on the principle of "optimality," aiming to ensure that the geomorphic feature consistency of the new unit is improved to the greatest extent and to avoid unreasonable division in spatial planning.

[0085] The rezoning process is iterated until the geomorphic feature consistency index of all units does not exceed a dynamically set adaptive threshold.

[0086] The threshold is set based on the statistical distribution characteristics of the geomorphic feature consistency index of the initial segmentation units, and is dynamically determined through data analysis. This threshold setting ensures the flexibility and accuracy of the re-segmentation process, while also avoiding over-segmentation or under-segmentation.

[0087] By combining various methods such as spatial overlap, improvement rate of geomorphic feature consistency index, and Deloni triangulation technology, this paper provides an efficient, flexible and scientific method for redrawing boundaries in territorial spatial planning. This method can not only optimize the geomorphic feature consistency of planning units, but also enhance the ability to prevent geological disasters and protect the ecology.

[0088] In one possible implementation, during geological surveys, faults are classified into active faults and dormant faults. Active faults are those that have experienced seismic activity in the past, while dormant faults are those that have not experienced seismic activity for a long period of time.

[0089] According to the fault activity classification table, active faults typically require a larger buffer distance to prevent earthquakes or other fault activity from impacting the surrounding area. The basic buffer distance is set based on the fault's activity level; the buffer distance for active faults is usually larger than that for dormant faults. The basic buffer distance for dormant faults is set as a multiple of that for active faults, with the specific multiple determined based on local geological conditions and engineering requirements.

[0090] Sedimentary layers are accumulations formed by the action of wind, rain, glaciers, and other forces. Loose sedimentary layers, especially thick ones, have low strength and bearing capacity, making them susceptible to external disturbances, particularly during seismic activity, which can exacerbate vibrations and surface deformation.

[0091] When the soil type code indicates that the soil layer is a loose sedimentary layer, the buffer distance needs to be increased according to the thickness of the sedimentary layer. The thicker the sedimentary layer, the greater the increase in the buffer distance. The purpose of this is to increase protection against geological hazards that may be caused by the sedimentary layer, such as foundation settlement and liquefaction.

[0092] Changes in groundwater levels are closely related to geological activities, such as fault activity. Time-series data on groundwater levels are periodically acquired through monitoring wells installed in fault zones. This data reflects long-term changes in groundwater levels, providing a basis for determining whether adjustments to buffer distances are necessary.

[0093] By analyzing the interannual variation rate of groundwater level, its standard deviation is calculated. The standard deviation reflects the degree of fluctuation in groundwater level changes. If the interannual variation is large, it may mean that the groundwater system is greatly affected by external factors (such as fault activity, precipitation, etc.), thus affecting geological stability.

[0094] If the standard deviation exceeds the set regional safe fluctuation range, it indicates increased instability of the groundwater system, which may affect geomorphic stability. In this case, the foundation buffer distance needs to be increased linearly proportionally to the magnitude of the groundwater level change rate. This can increase protection against abnormal groundwater changes and prevent groundwater fluctuations from negatively impacting the stability of the fault zone.

[0095] By finely setting the basic buffer distance and dynamically adjusting for changes in groundwater, a comprehensive buffer scheme for national spatial planning is provided. This scheme can effectively improve the safety and stability of the planned area and cope with potential risks brought about by changes in geological conditions and the environment. This not only has a positive significance for improving the scientific and rational nature of spatial planning, but also reduces engineering accidents and disasters caused by geological activity, thus protecting the safety of people's lives and property.

[0096] In one possible implementation, in land spatial planning, an urban construction unit refers to an area designated for urban functions in the plan. Each unit has its specific geographical scope.

[0097] The dynamic buffer zone is determined by the aforementioned geomorphological survey methods. This buffer zone is dynamically adjusted based on factors such as geological activity, groundwater changes, and sedimentary layers.

[0098] Using Geographic Information System (GIS) technology, it is determined whether the center point of an urban construction unit is located within the dynamic buffer zone. If the center point falls within the buffer zone, it indicates that the unit is located within a potential geological hazard risk area, requiring an early warning.

[0099] Each functional zone has its own planning constraints, including maximum building density, population density, etc., and these data are stored in a planning constraint database.

[0100] Using GIS systems or urban planning software, building density within urban construction units is calculated. If the density exceeds the maximum allowable value for that functional area, an early warning is triggered. Excessive building density can lead to problems such as excessive resource consumption, increased pressure on infrastructure, and increased risk of geological disasters.

[0101] By setting up groundwater monitoring wells within urban construction units, groundwater depth data can be collected periodically.

[0102] The year-on-year decrease is calculated by comparing the current data with historical data. If the current decrease exceeds the proportion of historical extreme values, it indicates an abnormal change in the groundwater system of the area, which may be related to factors such as geological activity and fault zones.

[0103] Drastic changes in groundwater depth may affect geological stability and increase the risk of landslides, foundation settlement, and other geological hazards.

[0104] A yellow alert is a warning of potential risks, usually indicating a higher risk level and the need to take certain preventative measures.

[0105] If a town or urban development unit meets two of the three red alert trigger conditions mentioned above, a yellow alert will be triggered. For example, if the building density of the unit is too high and the year-on-year decrease in groundwater depth exceeds the standard, it may mean that the area is at higher risk and requires enhanced monitoring.

[0106] A red alert indicates an extremely high risk, usually meaning that the security situation in the area is very serious and a major disaster may occur.

[0107] A red alert is triggered when a town or urban development unit simultaneously meets all three red alert triggering conditions and is located within the influence zone of a known historical ground fissure zone. The influence zone of a historical ground fissure zone refers to areas where ground fissures have historically occurred. Such areas are prone to future geological activity leading to fissures or land subsidence, increasing the risk of disasters.

[0108] Accordingly, embodiments of the present invention also provide a geomorphological survey system for land spatial planning, used to run any of the geomorphological survey methods for land spatial planning described in the embodiments of the present invention, including: The multi-source data acquisition and fusion module includes: The remote sensing image processing unit is configured to perform atmospheric correction on optical images and output a land cover type distribution map; The point cloud filtering and reconstruction unit is connected to the lidar sensor and configured to generate a digital elevation model and extract slope and aspect. The planning constraint dynamic rule base module includes: The semantic parsing unit takes land spatial planning text as input and outputs machine-executable code for slope threshold and settlement rate threshold. Spatial mapping unit stores a table of mapping relationships between rule identifiers and geographic coordinates; The spatiotemporal coupling analysis engine module, whose input is connected to the multi-source data acquisition and fusion module and the planning constraint dynamic rule base module, includes: The spatial correction unit receives the digital elevation model and land cover map, and corrects spatial offsets using thin-plate spline functions. The conflict marker unit compares the geomorphic parameters with the planning threshold in real time and outputs a set of conflict coordinates. The intelligent landform unit segmentation module, whose input is connected to the spatiotemporal coupling analysis engine module, includes: The watershed initialization unit generates watershed boundaries based on the conflict coordinate set; The index calculation unit is configured to dynamically weight the geomorphological feature consistency index by combining the density of geological hazard hazard points; Boundary optimization unit: When the index exceeds the limit, the unit boundary is redrawn along the ridgeline / valleyline. The planning and adaptation output module, whose input is connected to the intelligent landform unit segmentation module, includes: The 3D rendering unit integrates digital elevation models, conflict markers, and unit boundaries to generate survey maps. The dynamic buffer early warning unit sets the buffer distance and outputs an early warning layer based on the fault activity level and soil type.

[0109] Preferably, the dynamic buffer early warning unit further includes: Buffer distance setting sub-unit, with built-in fault activity classification table and soil type code library; It also includes an early warning response module, whose input is connected to a dynamic buffer early warning unit, including: The conditional decision sub-unit is configured to check whether each urban construction unit meets the following criteria: a. The cell center is located within the dynamic buffer; b. Building density exceeds the planned upper limit; c. The proportion of groundwater depth decreasing year-on-year beyond historical extremes; The hierarchical alarm subunit is configured as follows: a. When the above two conditions are met, a yellow alert is triggered and the yellow alarm device is activated; b. When all conditions are met and the unit is located in a historical ground fissure zone, the red alarm device and graphic display unit are activated.

[0110] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0111] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A geomorphological survey method for land spatial planning, characterized in that, Includes the following steps: Step 1: Multi-source data fusion preprocessing: Simultaneously acquire remote sensing optical images, lidar point clouds, and historical geological survey data of the target area; Atmospheric correction is applied to remote sensing images to generate land cover maps, and building density in built-up areas and leaf area index in vegetation areas are dynamically identified. Ground filtering is applied to point cloud data to generate a digital elevation model and to extract slope, aspect, and topographic relief. Step 2: Construction of the dynamic rule base for planning constraints: Analyze the semantics of land and space planning texts to extract slope thresholds for ecological protection zones, subsidence rate thresholds for urban construction areas, and soil erosion thresholds for farmland areas; Convert text rules into machine-executable code and establish a mapping relationship between rule identifiers and geographic coordinates; Step 3: Spatiotemporal constraint coupling analysis: Based on the digital elevation model, the spatial offset of the land cover map is corrected by thin plate spline function; Real-time comparison of planning thresholds with current geomorphological parameters, marking conflict coordinate sets of slope exceeding limits or settlement exceeding speed; Step 4: Intelligent segmentation of terrain units: Initial units are generated based on the watershed watershed, with the conflict coordinates as the center. The consistency index of landform features within the calculation unit is generated by dynamically weighting the ratio of slope variability to the planned slope threshold, the deviation of vegetation cover variability from the planned cover threshold, and the density of potential geological hazard points. When the index exceeds the limit, the boundary is redrawn along the ridgeline / valleyline; Step 5: Plan the adaptive output: Generate a 3D survey map that integrates a digital elevation model, conflict markers, and unit boundaries; A dynamic buffer distance is set based on the fault activity level and the type of soil and rock mass. When urban units are present in the buffer zone, an early warning layer is generated.

2. The geomorphological survey method for land spatial planning according to claim 1, characterized in that, The dynamic identification of building density in step 1 includes: Construct a convolutional neural network model, with the input being preprocessed remote sensing images and the output being the polygonal vector boundary of building roofs; Based on the spatial distribution ratio of urban development boundaries and ecological protection zones in the national land spatial planning, a differentiated grid division strategy is determined: a: Within the urban development boundary, a fine grid size that is positively correlated with the proportion of "high-density construction area" in the plan is adopted; b: Within the ecological protection zone, a sparse grid size that is negatively correlated with the area ratio of the "core protection zone" is adopted; The ratio of the total area of ​​the roof polygons within each grid to the grid area is used as the building density. When the grid spans different functional zones, the mixed density is calculated by area weighting.

3. The geomorphological survey method for land spatial planning according to claim 1, characterized in that, Step 2, the extraction of the slope threshold, includes: Natural language processing technology was used to identify slope restriction clauses in planning texts, and the numerical value X in the keyword "development is prohibited if the slope is greater than X degrees" was extracted as the basic threshold. Link to historical geological disaster databases and compile a histogram of actual slope distribution at landslide event locations within the target area; The final threshold is the weighted average of the basic threshold and the median slope of the landslide concentration area. The weighting principle is as follows: a: When the number of landslide events exceeds the historical average, increase the weight of the median landslide slope; b: When modifiers such as "strict protection" appear in the planning text, increase the weight of the basic threshold.

4. The geomorphological survey method for land spatial planning according to claim 1, characterized in that, The implementation methods for thin-plate spline function correction described in step 3 include: Adaptive control point layout: The density of control points is determined based on the topographic complexity index, which is calculated by multiplying the standard deviation of topographic relief of the digital elevation model by the slope change rate. Offset vector field construction: Using the deviation between the coordinates of the land cover map and the coordinates of the terrain raster at the control point as the observation value, a global offset vector field is generated by kriging interpolation; Minimize the energy function: Define the bending energy function of the thin plate spline function, add the coordinate weight factor of the planning conflict region, and use the conjugate gradient method to iteratively solve the spatial transformation parameters.

5. A geomorphological survey method for land spatial planning according to claim 1, characterized in that, Step 4 involves calculating the geomorphic feature consistency index within a unit, including: Calculation of the density of potential geological hazards: Taking the center of the unit as the center, the statistical range is determined according to the influence radius of historical landslides, and the number of potential hazards within the unit area is calculated; Weighting strategy: a: When the density of potential hazard points exceeds the threshold of the historical hazard level of the area, the slope variability weight will be increased to the set multiple of the vegetation cover variability weight. b: When the unit is located within the planned ecological protection red line, the vegetation cover variability weight is set to the set multiple of the slope variability weight; The set multiple has a piecewise linear relationship with the density of hidden danger points or the red line protection level.

6. A geomorphological survey method for land spatial planning according to claim 1, characterized in that, The rules for redrawing the boundaries in step 4 include: When the index exceeds the limit, the priority of redrawing along the ridgeline is higher than that along the valleyline, and the priority ranking is based on: a: Spatial overlap between the calculation unit boundary and the ridgeline / valleyline; b: Assess the improvement rate of geomorphic feature consistency index of the new unit after rezoning; Candidate boundary lines are generated using the Deloni triangulation technique, and the boundary that maximizes the decrease in the new unit index is selected as the final dividing line. The iteration terminates when all unit indices do not exceed an adaptive threshold, which is dynamically set based on the statistical distribution characteristics of the initial segmentation unit indices.

7. A geomorphological survey method for land spatial planning according to claim 1, characterized in that, The dynamic buffer distance setting in step 5 includes: Basic buffer distance determined: a: According to the fault activity classification table, the basic distance of active faults is set as a multiple of the set distance of dormant faults; b: When the soil / rock type code indicates a loose sedimentary layer, increase the base distance by an extension distance that is positively correlated with the thickness of the sedimentary layer; Correction for groundwater change rate: a: Obtain time-series data of groundwater levels from monitoring wells in the fault zone and calculate the standard deviation of the interannual variation rate; b: When the standard deviation exceeds the regional safe fluctuation range, the buffer distance is increased in a linear proportion to the rate of change and the base distance.

8. A geomorphological survey method for land spatial planning according to claim 1, characterized in that, The conditions for generating the warning layer include: Red alert trigger conditions: a: The center point of the urban construction unit falls within the dynamic buffer zone of the fault. b: The density of the unit building exceeds the upper limit value of this functional area in the planning constraint library; c: The proportion of the year-on-year decrease in groundwater depth within the unit exceeds the historical extreme value; Quantification of early warning levels: a: A yellow alert is triggered when two of the above conditions for triggering a red alert are met; b: A red alert is triggered when all three red alert triggering conditions are met and the unit is located within the influence range of the historical ground fissure zone.

9. A geomorphological surveying system for land spatial planning, used to run any one of the geomorphological surveying methods for land spatial planning as described in claims 1-8, characterized in that, include: The multi-source data acquisition and fusion module includes: The remote sensing image processing unit is configured to perform atmospheric correction on optical images and output a land cover type distribution map; The point cloud filtering and reconstruction unit is connected to the lidar sensor and configured to generate a digital elevation model and extract slope and aspect. The planning constraint dynamic rule base module includes: The semantic parsing unit takes land spatial planning text as input and outputs machine-executable code for slope threshold and settlement rate threshold. Spatial mapping unit stores a table of mapping relationships between rule identifiers and geographic coordinates; The spatiotemporal coupling analysis engine module, whose input is connected to the multi-source data acquisition and fusion module and the planning constraint dynamic rule base module, includes: The spatial correction unit receives the digital elevation model and land cover map, and corrects spatial offsets using thin-plate spline functions. The conflict marker unit compares the geomorphic parameters with the planning threshold in real time and outputs a set of conflict coordinates. The intelligent landform unit segmentation module, whose input is connected to the spatiotemporal coupling analysis engine module, includes: The watershed initialization unit generates watershed boundaries based on the conflict coordinate set; The index calculation unit is configured to dynamically weight the geomorphological feature consistency index by combining the density of geological hazard hazard points; Boundary optimization unit: When the index exceeds the limit, the unit boundary is redrawn along the ridgeline / valleyline. The planning and adaptation output module, whose input is connected to the intelligent landform unit segmentation module, includes: The 3D rendering unit integrates digital elevation models, conflict markers, and unit boundaries to generate survey maps. The dynamic buffer early warning unit sets the buffer distance and outputs an early warning layer based on the fault activity level and soil type.

10. A geomorphological survey system for land spatial planning according to claim 9, characterized in that, The dynamic buffer early warning unit further includes: Buffer distance setting sub-unit, with built-in fault activity classification table and soil type code library; It also includes an early warning response module, whose input is connected to a dynamic buffer early warning unit, including: The conditional decision sub-unit is configured to check whether each urban construction unit meets the following criteria: a. The cell center is located within the dynamic buffer; b. Building density exceeds the planned upper limit; c. The proportion of groundwater depth decreasing year-on-year beyond historical extremes; The hierarchical alarm subunit is configured as follows: a. When the above two conditions are met, a yellow alert is triggered and the yellow alarm device is activated; b. When all conditions are met and the unit is located in a historical ground fissure zone, the red alarm device and graphic display unit are activated.

Images