Multi-plot merging and cutting site selection algorithm

Abstract

The invention provides a multi-plot merging, cutting and site selection algorithm which is used for cutting of large plots and merging optimization of multiple plots in the field of land space planning. The problems of low utilization rate of large land parcels, space waste or scattered layout, low site selection efficiency, complex planning design and the like caused by difficulty in meeting specific area and shape requirements for large land parcel splitting, insufficient automatic processing for small land parcel merging and difficulty in dynamically adjusting the land parcel shapes and areas to meet different planning requirements in an existing algorithm are solved. The invention provides an algorithm for automatically splitting large-area land parcels and intelligently merging small-area land parcels, according to the algorithm, a multi-factor (area, shape, bitmap, purpose and the like) model is constructed, the priority of each land parcel is output and classified, and then according to a splitting and merging strategy, iteration and verification are repeated to obtain a reasonable result. According to the algorithm, flexible configuration and optimized layout of the site selection plot are realized, the utilization rate of space resources is improved, and efficient and operable technical support is provided for land space planning.

Description

Technical Field

[0001] The present invention belongs to the field of land space planning application technology, and in particular relates to a method for cutting a large plot of land and merging and optimizing multiple plots of land. Background Art

[0002] In land space planning and land use management, the rational division and merging of plots is of great significance for improving space utilization and optimizing layout. However, existing technologies often have difficulty meeting the requirements of specific area and shape when dealing with the precise division of large plots, which can easily lead to poor planning results. In addition, for the automatic merging of small plots, existing methods usually rely on manual operations or simple rule matching, resulting in low operational efficiency and difficulty in forming plots that meet planning standards. These limitations have led to uneven distribution and severe fragmentation of land resources, which is not conducive to the unified layout and functional zoning of space. At the same time, existing technologies lack flexible dynamic adjustment capabilities, making it difficult to meet the diverse needs of different planning projects, and therefore cannot provide decision makers with fast and effective plot optimization solutions. Summary of the Invention

[0003] This paper proposes a "multi-plot merging and cutting site selection algorithm" to address the difficult problem of plot cutting and merging in land spatial planning. By automatically splitting large plots and intelligently merging smaller plots, the algorithm enables flexible configuration and optimized layout of selected sites, improving the utilization of spatial resources and providing efficient and operational technical support for land spatial planning. The technical solution of this invention includes the following aspects:

[0004] Data preprocessing:

[0005] Before executing the plot cutting and merging operations, the land spatial planning data must be preprocessed to ensure the data is accurate and standardized. The preprocessing steps include:

[0006] Standardize spatial data and repair missing or abnormal data;

[0007] Standardize the geometric shape and boundaries of the plots for effective application in subsequent steps;

[0008] Ensure the consistency of coordinate systems, units, etc. of data from different sources.

[0009] Pre-selected multi-factor model construction:

[0010] This method uses multiple factors (such as plot area, shape, location, and land use) to construct a pre-selection model. By weighting the impact of each factor on site selection, the model outputs a priority for each plot, providing a basis for subsequent screening. This model leverages the multidimensional nature of spatial data to ensure that the impact of different factors on land use is considered.

[0011] Initial screening data set generation:

[0012] All plots are initially screened using a pre-selected multi-factor model. The screening criteria primarily consider the plot's area, shape, and its relationship to other plots, eliminating those that don't meet the criteria. This step effectively eliminates plots unsuitable for planning and improves algorithm efficiency.

[0013] Initial screening data set classification:

[0014] The initially filtered dataset is classified into groups based on attributes such as area, shape, and usage. These parcels are then assigned to different merging or splitting strategies based on their characteristics. This classification process ensures that each type of parcel receives the most appropriate treatment based on its characteristics.

[0015] Split-merge strategy construction:

[0016] This paper designs flexible splitting and merging strategies based on the characteristics of the classified plots. The splitting strategy is suitable for cutting large plots, while the merging strategy is mainly for merging small plots. The strategy construction process includes:

[0017] Split dataset construction: Based on spatial planning requirements, large plots are divided into several small plots to meet specific spatial layouts;

[0018] Split direction selection: select the optimal cutting direction based on the spatial position and shape of the plot;

[0019] Split step size control: By adjusting the split step size, we can avoid the waste of space resources caused by unreasonable subdivision;

[0020] Split iteration and verification: Through repeated iteration and verification, the cutting plan is continuously optimized to ensure that it meets the site selection requirements;

[0021] Merged dataset construction: Based on the adjacent relationship, small plots are merged into larger and more reasonable units to improve space utilization;

[0022] Adjacent area identification: During the merging process, adjacent areas are identified and spatially closely adjacent plots are prioritized for merging;

[0023] Prioritization: sorting the merge operations based on the relative position and priority of the plots to ensure an optimized layout;

[0024] Merger operation and verification: During the merging process, the merged plots are verified to ensure that they meet planning requirements and avoid unreasonable overlap.

[0025] Result saving and visualization:

[0026] The cut and merged parcel data is saved as a new spatial dataset and a detailed attribute column is generated. The saved dataset contains the geometric information, area, location, land use, etc. of each parcel.

[0027] Through the above technical solutions, the present invention can efficiently solve the problems of plot cutting and merging in land space planning, provide flexible site selection solutions, optimize the utilization of spatial resources, improve land utilization rate, and promote the sustainable development of land space planning.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. Improved spatial resource utilization: This invention utilizes an intelligent multi-plot merging and cutting strategy to optimize plot layout, effectively improving land utilization. Especially for large plots, rational segmentation can avoid wasting spatial resources. Small plots can be merged to form more efficient land use units, avoiding resource fragmentation. 2. Flexible adaptation to diverse plot characteristics: This invention constructs flexible segmentation and merging strategies based on multi-dimensional characteristics such as plot area, shape, and purpose, allowing for the development of optimal segmentation or merging schemes tailored to the specific circumstances of each plot. This flexibility enables the algorithm to adapt to diverse planning needs, offering advantages over traditional fixed-rule segmentation. 3. Optimized spatial planning decision-making: This invention utilizes a multi-factor model to pre-select and screen plots, preemptively filtering out plots that do not meet planning requirements, reducing unnecessary computation and analysis and improving spatial planning decision-making efficiency. Classification and prioritization further enhance the accuracy of land resource allocation. 4. High levels of automation and intelligence: This invention utilizes an automated segmentation and merging process, enabling the processing of large-scale plots in a short period of time, reducing the need for manual intervention and lowering labor costs. At the same time, the algorithm incorporates dynamic adjustment and feedback mechanisms during execution, allowing it to adjust strategies based on real-time needs, ensuring flexibility and accuracy. 5. Strong adaptability and sustainable optimization: The dynamic adjustment and feedback mechanisms of the present invention enable real-time adjustments based on changing planning requirements, ensuring the algorithm's adaptability and continuous optimization. This ensures excellent adaptability over long periods of time, enabling continuous adjustments to meet changing planning objectives and demonstrating strong sustainable development capabilities.

[0030] Through the above advantages, the present invention not only solves the problems of insufficient land resource utilization and low efficiency of plot cutting and merging in the existing technology, but also provides an efficient, flexible and intelligent solution that can significantly improve the effect and efficiency of land space planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. The following is a brief introduction to the drawings required for use in the embodiments.

[0032] Figure 1 This is a flow chart of a method for intelligent land space planning and site selection according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0034] like Figure 1 FIG. 1 is a flowchart of a multi-plot merging and cutting site selection algorithm provided by an embodiment of the present invention, including:

[0035] Data preprocessing:

[0036] Data collection and import: Obtain the spatial and attribute data of the plots related to the site selection area. Spatial data can be in common Geographic Information System (GIS) formats, such as Well-Known Text (WKT), GeoJSON, or Shapefile (SHP). The Shapefile format is particularly suitable for storing the geometric data and related attribute information of multiple plots. The Shapefile format is composed of multiple files (such as .shp, .shx, .dbf, etc.), where .shp files store geometric shapes, .dbf files store attribute data, and .shx files are used for indexing. By reading the Shapefile file, the geometric boundaries and attribute data of the plots are extracted.

[0037] When importing data, you can use the GDAL library or GeoPandas library to load Shapefile data into a geographic database for unified management, ensuring the structuring and standardization of plot data to support subsequent spatial query and analysis operations.

[0038] Coordinate system conversion: To ensure the uniformity of the spatial data of the plot, its coordinate system needs to be uniformly converted to a universal geodetic coordinate system (such as WGS84 or a local projection coordinate system). This step can use the following coordinate conversion formula:

[0039] x′=xcosθ-ysinθ

[0040] y′=xcosθ+ycosθ

[0041] Where θ is the desired rotation angle, x and y are the original coordinates, and x′ and y′ are the transformed coordinates. If the coordinate systems of different data sources are inconsistent, unify them to the same standard coordinate system to ensure the accuracy of subsequent spatial operations.

[0042] Parcel data cleaning: Parcel data is cleaned for topological consistency. Redundant and overlapping parcel boundaries are removed to avoid duplicate calculations. Incomplete polygons are closed using boundary repair algorithms (such as the shortest path algorithm). The data topology is checked to ensure that it meets conditions such as closure, non-overlap, and the absence of holes to improve the geometric validity of the parcels. After cleaning, the parcels must conform to GIS topological rules to ensure accuracy in spatial operations.

[0043] Land parcel attribute standardization: Standardize the attribute data of the land parcel. Unify the area into square meters or hectares, and use the following area conversion formula:

[0044] Standard area = original area × conversion factor

[0045] For example, when converting acres to square meters, the conversion factor is 4046.86. Unified attribute field naming and missing data filling ensure that all land parcel data has the same attribute dimensions and data quality, meeting the needs of multi-factor screening and classification.

[0046] Data stratification and classification: Based on the land parcel's area, purpose, land use type and other attribute characteristics, use rules or machine learning classification algorithms to stratify and classify the land parcels. The commonly used attribute weight calculation formula is:

[0047]

[0048] Preselected multi-model factor construction:

[0049] The construction of a multi-factor model will quantitatively analyze and screen various influencing factors to ensure that the land parcel meets the suitability and planning requirements of various indicators. The construction of a multi-factor model includes the following key steps and methods:

[0050] Slope planning category matching: According to land planning requirements, the use of the plot needs to match the planning category. Assume that the planning category of the plot is L and the target planning category is L target (such as "residential land").

[0051] When the planning category of the plot satisfies L=L target When the land parcel meets the planning requirements, it becomes a candidate land parcel. This step is matched and screened based on the land parcel attribute information, without the need for complex calculations.

[0052] Ownership type analysis: Ownership type factor is used to identify the ownership type of the land parcel to determine its suitability for development. Assume that the ownership type of the land parcel is O, and set the ownership type with development priority to O. priority (such as "state-owned").

[0053] If yes, then the land parcel meets the ownership conditions. This step filters the land parcel attribute information and no further calculation is required.

[0054] Slope factor screening: The slope factor is used to determine whether the slope of the plot is suitable for development and construction. Assuming the slope value is θ, set the slope range suitable for development [θ min ,θ max ], for example, [0°, 25°].

[0055] Calculate the slope value of each grid cell on the plot and average it to obtain the plot slope index [θ mean ]. If the conditions are met:

[0056] θ min ≤θ mean ≤θ max

[0057] The plot is then determined to meet the slope requirements and marked as a candidate plot.

[0058] Elevation factor screening: The elevation factor is used to exclude plots that pose a risk of natural disasters due to their altitude. The elevation value is recorded as h, and the elevation suitable interval [h min ,h max ], for example, 10 meters to 100 meters.

[0059] Calculate the average elevation h of the plot mean , if:

[0060] [h min ≤h mean ≤h max ]

[0061] The plot passes the elevation screening and is marked as meeting the conditions.

[0062] Municipal facilities distance analysis: The distance factor of municipal facilities (such as roads, water supply, power supply facilities, etc.) is used to ensure the infrastructure convenience of the plot. Assume that the distance from the plot to the nearest municipal facility is d infra , set the appropriate distance threshold d for municipal facilities infra_max (e.g. 1000 meters). If:

[0063] d infra ≤d infra_max

[0064] The plot is then judged to be convenient for municipal facilities and marked as a candidate plot. This distance can be calculated using the Euclidean distance formula.

[0065] Analysis of distance to public service facilities: The distance factor of public service facilities (such as schools, hospitals, supermarkets, etc.) is used to evaluate the convenience of living facilities on the plot. Assume that the distance from the plot to the nearest public service facility is d service , set the appropriate distance threshold d for public service facilities service_max (e.g. 2000 meters). If:

[0066] d service ≤d service_max

[0067] Then the plot is judged to be convenient for public service facilities and marked as a candidate plot.

[0068] Restricted plots exclusion: To ensure ecological protection and the integrity of agricultural land, some specific plots (such as basic farmland and scenic spots) are designated as restricted plots and are prohibited from being used for other purposes. Let the geometric boundaries of these restricted plots be Geom restricted , the geometric boundary of the candidate plot is Geom candidate If:

[0069] Geom candidate ΙGeom restricted =θ

[0070] That is, if the two do not overlap, the candidate plot passes the restrictive plot exclusion screening and is allowed to be used for development.

[0071] Initial screening dataset generation:

[0072] Reading and formatting parcel data: First, extract global parcel data from the system (such as .shp files or GeoJSON format) and convert it to a format suitable for analysis (such as raster format or table format). This step ensures that the data can be correctly read and processed in subsequent calculations.

[0073] Attribute screening condition definition: According to the basic requirements of land space planning, set the screening conditions of the plot, including the following:

[0074] ·Plot area: exclude plots with an area less than the minimum value A min to ensure that the land parcels have sufficient area for development.

[0075] Land use type: Based on the planning attributes of the land parcel (such as agricultural land, industrial land), filter out types that do not meet the requirements.

[0076] Ownership: Only plots with clear ownership and available for development (such as state-owned or collectively owned land) are retained. Spatial Condition Screening: Based on the geographical location and spatial relationship of the plots, further screening is conducted to identify plots that meet the spatial conditions, including the following steps:

[0077] Slope screening: Calculate the average slope value θ for each plot mean Eliminate slopes exceeding the set slope threshold θ max to ensure that the land is suitable for development.

[0078] Elevation screening: calculate the average elevation value h of the plot mean , remove the elevations below or above the appropriate range [h min ,h max ] to avoid potential flooding or high-altitude development challenges.

[0079] Elimination of ecological protection areas: Land parcels located in ecological protection areas, basic farmland, and scenic spots will be eliminated to ensure compliance with legal provisions on ecological protection and agricultural land use.

[0080] Distance screening: Consider the distance from the plot to major municipal facilities and public service facilities to assess its convenience for living and production. This includes the following two aspects:

[0081] Municipal facility distance screening: Calculate the distance d from the plot to the nearest municipal facility infra , eliminating plots that are beyond the service range of municipal facilities.

[0082] Public service facility distance screening: Calculate the distance d from the plot to the nearest public service facility (such as school, hospital) service , eliminating plots that are beyond reasonable living range

[0083] Initial screening plot marking and export: Plots that pass the above screening criteria are marked as "initial candidate plots" to generate an initial candidate dataset. This dataset contains plots that meet the area, ownership, slope, elevation, space, and distance criteria, providing input for the subsequent multi-factor comprehensive model construction.

[0084] Initial screening dataset classification:

[0085] The main purpose is to establish classification marks for candidate plots to ensure that plots with different characteristics can be split, merged, or retained in a targeted manner in subsequent operations. The specific classification steps are as follows:

[0086] Classification basis setting: Based on factors such as the area, shape, ownership and surrounding facilities of the land parcel, set the initial screening and classification criteria. The main basis includes:

[0087] Area size: The plots are divided into “small plots”, “suitable plots” and “large plots”, with the area threshold A min and A max For the dividing line.

[0088] Plot shape: Based on aspect ratio and shape regularity, plots with more regular shapes are prioritized as "suitable plots", while plots with irregular shapes enter further processing.

[0089] Ownership consistency: Ensure that the ownership of adjacent plots is consistent when they are merged; plots with inconsistent ownership will not be included in subsequent mergers.

[0090] Impact on surrounding facilities: Based on the distance of the plot from municipal facilities and public services, more convenient locations will be prioritized as suitable plots.

[0091] Initial screening classification mark of plots: Generate classification marks of the initial screening data set according to the above classification standards:

[0092] Small plot: The area is smaller than A min The plots are marked as "small-area plots" and serve as candidates for merging.

[0093] Large plots: larger than A max The plots are marked as “large area plots” and are candidates for splitting.

[0094] Suitable plot: Area A min to A max The plots that are between the two areas and have regular shapes and meet the planning conditions are marked as "suitable plots" and are directly retained.

[0095] Verification of classification results: Verify the initially classified plots to ensure that each type of plot meets the classification conditions. Specifically including:

[0096] Area and shape check: Ensure that "small area plots" and "large area plots" meet the area and shape marking conditions.

[0097] Ownership consistency check: Ensure that small plots that meet the merging conditions and adjacent plots suitable for merging have consistent ownership information.

[0098] Spatial distribution rationality check: Ensure that the distribution of plots meets the adjacency requirements. Plots that do not meet the adjacency conditions will not be merged.

[0099] Output classification results: Finally, the marked plots are divided into categories to generate three types of result datasets: suitable plots, small-area plots, and large-area plots, and exported for subsequent further processing operations.

[0100] Split the dataset:

[0101] The main purpose is to reasonably split large plots that exceed the appropriate area range to better meet planning requirements. The following is a detailed description and steps for splitting the dataset.

[0102] Split dataset construction: Generate split datasets from large area plots for gradual splitting and processing.

[0103] Here’s how:

[0104] For the plots marked as “large area plots”, according to the initial area threshold A max Construct a split dataset containing all data with an area greater than A max plot of land.

[0105] In the construction of the split data set, the area, shape, ownership and other information of each plot are recorded for the splitting algorithm to call.

[0106] Split direction selection: Determine a reasonable split direction for each large area parcel in the split dataset.

[0107] Here’s how:

[0108] Calculate the aspect ratio of each plot and analyze its main axis direction (major or minor axis direction).

[0109] If the aspect ratio of a plot is greater than a certain threshold, it is preferentially split along the long axis to make the shape of the split plot more regular.

[0110] For plots that are close to square in shape, they are preferentially split along the direction parallel to the boundaries of the surrounding plots to ensure that the boundaries of the split plots match.

[0111] The split direction is calculated based on the aspect ratio:

[0112]

[0113] where R threshold is the aspect ratio threshold.

[0114] Split step control: Control the step size during splitting to ensure that the split plots meet the area and shape requirements.

[0115] Here’s how:

[0116] Determine the split step size d as the split length, usually based on the expected target area A target and the main axis length L of the plot main calculate:

[0117]

[0118] According to the value of the step size d, the plot is gradually split along the determined splitting direction.

[0119] The area of ​​the split plot should be as close as possible to A target , to avoid plots that are too large or too small.

[0120] rule:

[0121] The step size d is chosen to be moderate to avoid generating too many small blocks or a single large block.

[0122] If the area deviation of the split plot exceeds the set tolerance range, the step size will be adjusted and the plot will be split again.

[0123] Split iteration and verification: Through iterative splitting and verification, ensure that the generated plots meet planning requirements.

[0124] Here’s how:

[0125] Iterative splitting: After each split, the generated sub-plots are verified to see if they meet the area and shape requirements. For plots that do not meet the requirements, the splitting process is repeated.

[0126] Area verification: Calculate the area of ​​each split plot. If the area is not within the allowable range [A main ,A max ], it is further split.

[0127] Shape Verification: Calculate the aspect ratio of the split plot to ensure its shape is regular. If the shape does not meet the standard (e.g. the aspect ratio is too large), adjust the split direction or step size and split again.

[0128] Peripheral consistency verification: Check the boundaries of the split plots to ensure that they are consistent with the boundaries of adjacent plots to avoid generating irregular boundaries.

[0129] The area verification and shape verification formulas are as follows:

[0130]

[0131] After the above steps, the large-area plots are reasonably split to ensure that the area, shape and boundaries of the split plots meet the planning requirements, and suitable plots that meet the area range are generated for subsequent further processing.

[0132] Merge datasets:

[0133] The primary goal is to rationally merge small or irregular plots into larger, more regular plots, thereby improving land resource utilization efficiency and making the plots more suitable for planning needs. The following is a detailed description and steps for merging datasets.

[0134] Merged dataset construction: Filter out small plots that do not meet the requirements in terms of area or shape, and construct a merged dataset.

[0135] Here’s how:

[0136] Based on the initial area lower limit A min and shape standards, screen all plots and select those with an area smaller than A min or plots whose aspect ratio exceeds a set threshold.

[0137] The screened plots were included in the merged dataset, and their location, shape and adjacent plot information were recorded.

[0138] rule:

[0139] Small plots that do not meet the requirements in area or shape are added to the merged dataset first for subsequent merging operations.

[0140] Be sure to include neighboring parcel information in the merged dataset so that the merge can be performed efficiently.

[0141] Adjacent area identification: Identify the adjacent areas of each plot to be merged in order to select the merging target.

[0142] Here’s how:

[0143] Identify adjacent parcels by analyzing parcel boundaries. Use a nearest neighbor algorithm or buffer zone analysis to identify parcels that are adjacent to the boundaries of the parcel to be merged.

[0144] Calculate the distance d between adjacent plots. If the distance is less than the given merge distance threshold D merge , then the land parcel is considered as a mergible target.

[0145] Adjacent distance judgment conditions:

[0146] d ij <D merge

[0147] where d ij represents the distance between plot i and plot j, D merge is the merge distance threshold.

[0148] Prioritization: Prioritize the mergeable plots to ensure that favorable plots are processed first in the merge operation.

[0149] Here’s how:

[0150] Sort by merging priority rules, which include factors such as area, number of adjacent plots, and shape.

[0151] Smaller plots will be merged first to ensure that small plots can be incorporated into larger areas as quickly as possible.

[0152] Priority will be given to plots with more adjacent plots in order to expand the merged area and improve the regularity of the plots.

[0153] For plots with irregular shapes (such as those with a large aspect ratio), merging operations are performed first to optimize the plot shape.

[0154] Priority formula:

[0155]

[0156] Where A is the plot area, N adjacent is the number of adjacent plots, R shape is the aspect ratio of the plot, ω1, ω2, ω3 are weight coefficients.

[0157] Merge operation and verification: Merge the plots to be merged step by step according to priority, and verify whether the merged plots meet the requirements of area, shape, etc.

[0158] Here’s how:

[0159] Merge operation: Select the parcels to be merged in order of priority, merging them first with the adjacent parcel with the highest priority. The merging method can be direct boundary merging or boundary smoothing through buffer operation.

[0160] Area verification: The area of ​​the merged plot must be within the specified range [A min ,A max ] to ensure that the merger effect meets the planning requirements.

[0161] Shape verification: The aspect ratio of the merged plots should meet the shape requirements. If the aspect ratio is too large, try to adjust the merge boundary or merge it with other plots until the shape standards are met.

[0162] Boundary smoothing: Smooth the boundaries of the merged plots to eliminate irregular boundaries and ensure that the merged plots have a regular shape.

[0163] Combined verification conditions:

[0164]

[0165] Among them A merged is the area of ​​the merged plot, R threshold is the shape ratio threshold.

[0166] After the above steps, the merger of small plots is completed, ensuring that the merged plots meet planning requirements in terms of area, shape and boundaries, and improving the consistency and practicality of land use.

[0167] The present invention effectively adjusts the area distribution of the plots by splitting and merging the candidate plots, so that the plots meet the site selection requirements of the spatial planning.

Claims

1. A multi-plot merging and cutting site selection algorithm in land space planning, characterized by: The algorithm processes the given plot data through multiple steps, including data preprocessing, pre-selection of multi-factor models, generation of preliminary screening datasets, classification of preliminary screening datasets, construction of split-merge strategies, and result storage. The algorithm includes: Data preprocessing step: clean, remove noise and standardize the input plot data to ensure data accuracy and consistency; Pre-selection multi-factor model construction steps: construct a pre-selection model through multiple factors (such as geographical information, environmental factors, etc.) to provide support for subsequent screening; The steps of generating the preliminary screening data set are as follows: based on the pre-selected model, a preliminary test data set that meets the requirements is selected from the processed data; Initial screening data set classification step: Classify the initial screening data set to generate data sets of multiple categories; The steps of constructing the split-merge strategy include: constructing a split dataset, selecting a split direction, controlling the split step size, split iteration and verification, as well as constructing a merge dataset, identifying adjacent areas, prioritizing, merging operations and verification, ultimately forming a multi-plot merging plan suitable for site selection; Result saving step: Save the final merged data set to a database or file system to ensure that it can be directly used by subsequent applications.

2. The multi-plot merging and cutting site selection algorithm in land space planning according to claim 1 is characterized in that: The split data set construction in the split-merge strategy construction step includes: According to the input plot data and target area, select the appropriate splitting method to divide the plot into multiple smaller units for further processing; Set the split direction and step size to ensure that each unit after splitting meets the requirements of subsequent analysis and site selection.

3. The multi-plot merging and cutting site selection algorithm in land space planning according to claim 1 is characterized in that: The split direction selection in the split-merge strategy construction step includes: According to the geographical characteristics of the target area and the spatial distribution of the data, the optimal splitting direction is automatically selected to achieve a more efficient splitting operation.

4. The multi-plot merging and cutting site selection algorithm in land space planning according to claim 1 is characterized in that: The split step size control in the split-merge strategy construction step includes: According to the area of ​​the split region and the accuracy requirements of the split, the step size of each split is dynamically adjusted to ensure that the split effect meets the expectations.

5. The multi-plot merging and cutting site selection algorithm in land space planning according to claim 1 is characterized in that: The adjacent region identification in the step of constructing the merged data set includes: Perform spatial analysis on each split data unit to identify adjacent areas and merge them based on the adjacent relationship.

6. The multi-plot merging and cutting site selection algorithm in land space planning according to claim 1 is characterized in that: The result saving step includes: The processed merged dataset is saved in a predetermined format and supports subsequent query and analysis, ensuring that the results can be easily used in actual land spatial planning applications.

7. The multi-plot merging and cutting site selection algorithm in land space planning according to claim 1 is characterized in that: The step of generating the preliminary screening data set comprises: The data were screened based on the pre-selected multi-factor model to obtain a data set that preliminarily met the site selection requirements, providing a basis for subsequent classification and merging.

8. The multi-plot merging and cutting site selection algorithm in land space planning according to claim 1 is characterized in that: The priority sorting in the split-merge strategy construction step includes: Prioritize the data units involved in the merge operation to ensure that regions with higher merge value are processed first during merging, thereby optimizing the final result.

9. The application of the multi-plot merging and cutting site selection algorithm in land space planning according to claim 1, wherein the application at least comprises: Merging and cutting multiple plots, that is, merging multiple plots into new plots that meet planning requirements based on the site selection algorithm; The merged dataset is stored and queried, providing effective spatial data query and subsequent analysis capabilities; Analytical tools for land spatial planning support further analysis, simulation and planning decisions for merged plots.

Images