A spatial planning analysis system based on data visualization

Abstract

This invention discloses a spatial planning and analysis system based on data visualization, belonging to the field of image processing technology. The system includes an extraction module for acquiring multiple target spatial planning images to be stitched, identifying the edges of planning regions in each target spatial planning image, and calculating the planning data offset of adjacent target planning regions at preset docking edges; a stitching region definition module for receiving the planning data offset and defining a stitching processing buffer; a generation module for generating a first stitching path for geometrically overlapping regions and a second stitching path for non-overlapping gap regions; and an output module for receiving the first and second stitching paths and generating the stitched spatial planning image. This invention achieves precise processing of complex stitching scenarios by generating first and second stitching paths differently for geometrically overlapping and non-overlapping gap regions through the generation module.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a spatial planning and analysis system based on data visualization. Background Technology

[0002] In regional coordinated development planning, it is often necessary to integrate planning maps compiled by different administrative units or at different times to form a unified planning base map for visualization analysis and decision-making across the entire region.

[0003] Existing spatial planning analysis systems based on data visualization have significant drawbacks when performing multi-map fusion: First, spatial planning is a dynamic process. Policy adjustments cause planning data in different regions to evolve independently over time. When planning content (such as land use boundaries and properties) at the edges of regions changes non-coordinatedly, direct splicing will lead to geometric misalignment and attribute conflicts at the junctions of adjacent regions. Second, during the fusion process, simple interpolation or forced alignment of biased junction areas introduces new classification errors. This results in classification results at the seams of the spliced ​​planning map that do not conform to the actual planning status of any of the original regions and disrupt the logical consistency of the overall planning. Ultimately, this leads to a discrepancy between the planning base map on which the visualization analysis is based and the actual situation, reducing the accuracy and decision-making value of the analysis.

[0004] Therefore, there is an urgent need to provide a spatial planning and analysis system based on data visualization to solve the above problems. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a spatial planning and analysis system based on data visualization.

[0006] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is to provide a spatial planning and analysis system based on data visualization, including an extraction module, a splicing region definition module, a generation module, and an output module; The extraction module is used to acquire multiple target planning space images to be stitched together, identify the planning area edges in each target planning space image, and calculate the planning data offset of adjacent target planning areas at a preset docking edge. The planning data offset includes geometric position offset value and planning attribute difference degree. The splicing area definition module receives the planning data offset and, based on the difference between the geometric position offset value and the planning attribute, defines a splicing processing buffer at the preset docking edge of the adjacent target planning area. The splicing processing buffer includes a geometrically overlapping area and a non-overlapping gap area. The generation module generates a first splicing path for the geometrically overlapping area and a second splicing path for the non-overlapping gap area. The output module receives the first stitching path and the second stitching path, and performs fusion and reconstruction of the planning data in the geometrically overlapping area according to the first stitching path. At the same time, it performs adaptive interpolation of the planning data in the non-overlapping gap area according to the second stitching path, merges the processed adjacent target planning areas, generates a stitched planning spatial image, and calculates the global stitching consistency evaluation index of the stitched planning spatial image.

[0007] The present invention is further configured such that: multiple target planning spatial images in the extraction module are retrieved and called from a preset geographic information system database or historical planning archive based on the timestamp and spatial range label specified by the user; The method for identifying the edge of the planning region in each of the target planning spatial images includes: performing grayscale and binarization preprocessing on the target planning spatial image, extracting the contours of connected regions representing different planning attributes in the target planning spatial image, and identifying the outer boundary of the connected region contour as the edge of the planning region; The method for obtaining the preset docking edge is as follows: S1. After completing the edge identification of the planning area, the system calculates the minimum spatial proximity between each two planning area edges based on the spatial positional relationship of all planning area edges, and determines the target planning areas corresponding to the two planning area edges whose minimum spatial proximity is less than a preset threshold as spatially adjacent areas, forming multiple sets of candidate adjacent area pairs. S2. For each pair of candidate adjacent regions, extract continuous edge segments in the two planning regions whose spatial distance is less than a preset distance. Define each pair of extracted continuous edge segments as a candidate docking edge segment and calculate the average spatial distance and average planning attribute similarity between each pair of candidate docking edge segments. S3. Based on the average spatial distance and average planning attribute similarity of each pair of candidate docking edge segments, construct a comprehensive proximity evaluation index, and determine the candidate docking edge segments whose comprehensive proximity evaluation index is higher than a preset screening threshold as preset docking edges.

[0008] The present invention is further configured such that the calculation steps for the planning data offset in the extraction module are as follows: S4. For each of the predetermined pre-defined docking edges, select multiple sampling points at predetermined intervals. For each sampling point, draw perpendicular lines from the normal direction of the pre-defined docking edge at the sampling point to two adjacent target planning areas, intersecting with the edges of their respective planning areas to obtain a pair of geometrically corresponding points. Calculate the Euclidean distance between each pair of geometrically corresponding points as the local geometric offset value of the sampling point. Statistically calculate the local geometric offset values ​​of all sampling points on the predetermined docking edge, calculate their average value and standard deviation, and use the weighted sum of the average value and standard deviation as the geometric position offset value of the predetermined docking edge. S5. For the same pair of adjacent target planning areas, in the preset neighboring areas on both sides of the preset docking edge, extract the attribute feature vectors that reflect the semantic information of the planned land use nature and development intensity, calculate the cosine similarity between the attribute feature vectors extracted from the two sides, and define the value obtained by subtracting the cosine similarity from 1 as the preliminary planning attribute difference at the preset docking edge. At the same time, considering the degree of deviation of the attribute evolution trend of the preset neighboring areas on both sides of the preset docking edge in the historical planning archive, the preliminary planning attribute difference is corrected to obtain the planning attribute difference. The planning attribute difference and the geometric position offset value together constitute the planning data offset.

[0009] The present invention is further configured such that: the geometrically overlapping area in the splicing area definition module refers to the area in which the pre-defined adjacent area planning elements on both sides of the pre-defined docking edge spatially intrude into each other, and the non-overlapping gap area refers to the blank area formed by the pre-defined adjacent area planning elements on both sides of the pre-defined docking edge not being covered; The method for defining the splicing processing buffer is as follows: Q1. Receive the geometric position offset value, multiply the geometric position offset value by a preset base width coefficient to obtain the preliminary single-sided base width of the buffer. Q2. Receive the planning attribute difference degree, compare the planning attribute difference degree with three preset threshold intervals (low, medium, and high), and determine the basic level coefficient based on the threshold interval where the planning attribute difference degree falls; adjust the basic level coefficient by combining the length of the preset docking edge and the frequency of historical planning changes to generate an attribute difference correction factor; multiply the initial buffer single-sided basic width by the attribute difference correction factor to obtain the corrected buffer single-sided actual width; using the buffer single-sided actual width as a reference, symmetrically expand outward along the preset docking edge to the adjacent target planning areas on both sides to directly generate the actual boundary of the splicing processing buffer. Q3. Within the generated splicing processing buffer, based on the spatial distribution data of planning elements extracted from the original target planning spatial image, perform spatial overlay analysis to identify sub-regions that are simultaneously covered by planning elements of the adjacent target planning regions on both sides, and mark them as geometrically overlapping regions. At the same time, identify sub-regions that are not covered by planning elements on either side, and mark them as non-overlapping gap regions, thus completing the internal structure division of the splicing processing buffer.

[0010] The present invention is further configured such that the generation step of the first splicing path in the generation module is as follows: R1. For the geometrically overlapping region that needs to be processed, the geometrically overlapping region is subdivided into multiple geometrically continuous overlapping sub-regions with relatively consistent internal attributes. The geometric center point, area, and planning attribute codes of the adjacent target planning regions on both sides covered by each overlapping sub-region are extracted. A description unit containing geometric shape and attribute information is established for each overlapping sub-region. R2. Based on the description unit, evaluate the compatibility level of the planning attributes of the adjacent target planning areas on both sides. The compatibility level is determined by comparing the relationship between the planning attribute codes on both sides in a preset planning attribute compatibility lookup table. Based on the evaluated compatibility level, assign a preset fusion rule to each overlapping sub-region. R3. Based on the fusion rules assigned to each overlapping sub-region and the geometric features of the overlapping sub-region, calculate the first final planning attribute value of each first location point to be determined within each overlapping sub-region, generate a first spatial interpolation path for each overlapping sub-region according to the first final planning attribute value, connect and smooth the first spatial interpolation paths of all overlapping sub-regions according to their spatial adjacency relationship, and integrate them to form a first splicing path. The method for calculating the first final planning attribute value in step R3 is as follows: R301. Obtain the fusion rule assigned to each of the overlapping sub-regions. The fusion rule pre-sets the weight calculation function and attribute interpolation model applicable to the overlapping sub-regions. According to the fusion rule, for each first location point to be determined inside the overlapping sub-region, calculate the normalized distance from the first location point to the geometric boundaries of the adjacent target planning regions on both sides covered by the overlapping sub-region, and generate distance weight pairs. R302. Based on the distance weight pair, and combined with the planning attribute codes of the two adjacent target planning regions obtained from the description unit, a preset attribute compatibility matrix is ​​queried to obtain the dynamic similarity coefficient between the planning attribute codes of the two adjacent target planning regions. The distance weight pair is corrected using the dynamic similarity coefficient to generate a fusion weight pair. The standard attribute values ​​corresponding to the planning attribute codes of the two adjacent target planning regions are multiplied by the corresponding weights in the fusion weight pair and then summed to obtain the first final planning attribute value of the first location point. The calculation process traverses all the first location points in the overlapping sub-region.

[0011] The present invention is further configured such that the generation step of the second splicing path in the generation module is as follows: R4. For the non-overlapping gap area, analyze the spatial geometric characteristics of the non-overlapping gap area and the effective areas with clearly defined planning attributes adjacent to it within a preset range, and extract the planning attribute codes and spatial distribution patterns of the effective areas as usable planning attribute information for filling the non-overlapping gap area. R5. Based on the available planning attribute information, calculate the attribute influence weight from each second location point at the boundary of the non-overlapping gap area to different effective areas, and evaluate it according to the preset planning logic consistency rule to determine the dominant attribute source or generate a new transition attribute code for different locations inside the non-overlapping gap area, thus forming a gap filling strategy. R6. Based on the gap filling strategy, calculate the second final planning attribute value for each second location point that needs to be filled in the non-overlapping gap area. Connect the calculation process of the second final planning attribute value according to the spatial proximity relationship of the second location points to form a continuous second spatial interpolation path covering the entire non-overlapping gap area. Define the second spatial interpolation path as the second splicing path.

[0012] The present invention is further configured such that: the step of fusing and reconstructing the planning data in the geometrically overlapping region according to the first splicing path, and simultaneously adaptively interpolating the planning data in the non-overlapping gap region according to the second splicing path in the output module is as follows: T1. Receive the first splicing path transmitted by the generation module, and according to the spatial interpolation rules of each overlapping sub-region defined in the first splicing path, assign the calculated first final planning attribute value to each first position point in the geometric overlapping region, complete the planning attribute reconstruction of all pixels or elements in the entire geometric overlapping region, and generate the processed geometric overlapping region data. T2. Receive the second splicing path, and according to the gap filling strategy defined in the second splicing path and the second spatial interpolation path, assign the calculated second final planning attribute value to each second location point in the non-overlapping gap area, complete the interpolation and filling of the planning attributes in the entire non-overlapping gap area, and generate the processed non-overlapping gap area data.

[0013] The present invention is further configured such that the output module merges the processed adjacent target planning regions by: geometrically aligning and seamlessly stitching the processed geometrically overlapping region data with the processed non-overlapping gap region data, while integrating the internal region data in the original target planning region that is not affected by the stitching processing buffer, to generate a complete stitched planning space image.

[0014] The present invention is further configured such that the calculation method for the global splicing consistency evaluation index in the output module is as follows: T3. In the generated stitched planning space image, several evaluation sample points are arranged along the center line of the stitching processing buffer at a preset density. For each evaluation sample point, the planning attribute code corresponding to the first position point or the second position point is extracted from the original target planning space image adjacent to its left and right sides, and the difference value between the two planning attribute codes is calculated as the local attribute difference measure of the first position point or the second position point. T4. Based on the local attribute difference measure of all the evaluation sample points, calculate its mean and standard deviation, and combine the geometric position offset value and the planning attribute difference degree to generate a global splicing consistency evaluation index through a weighted fusion formula.

[0015] The present invention is further configured such that: the output module further includes rendering the stitched planned spatial image and the global stitching consistency evaluation index together onto a preset visualization interface to generate an interactive visualization analysis report containing an evaluation result layer and a planned image layer.

[0016] The beneficial effects of this invention are as follows: 1. This invention accurately identifies the edges of the planning area and calculates the geometric and attribute offsets by the extraction module, and delineates the buffer zone to distinguish overlapping and gap areas by the delineation module. This effectively solves the problem of edge geometric misalignment and attribute conflict caused by dynamic planning, thereby significantly improving the geometric and semantic accuracy of multi-source planning spatial image stitching. 2. This invention generates first and second splicing paths differently for geometrically overlapping and non-overlapping gap areas through a generation module, and performs fusion reconstruction and adaptive interpolation based on dynamic weights and attribute compatibility, thereby achieving accurate processing of complex splicing situations and avoiding new errors introduced by traditional simple interpolation or forced alignment methods. 3. This invention calculates the global stitching consistency evaluation index through the output module and renders the index and the final image together into an interactive visualization report, which intuitively presents the stitching quality and potential problems, providing a reliable and quantifiable data basis for planning analysis and decision-making, and greatly enhancing the decision support value of visualization analysis. Attached Figure Description

[0017] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a flowchart of the method for defining the splicing processing buffer of the present invention. Detailed Implementation

[0018] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0019] Please see Figures 1-2 A spatial planning and analysis system based on data visualization includes an extraction module, a splicing region definition module, a generation module, and an output module; The extraction module is used to acquire multiple target planning space images to be stitched, identify the planning area edges in each target planning space image, and calculate the planning data offset of adjacent target planning areas at the preset docking edge. The planning data offset includes the geometric position offset value and the planning attribute difference degree. The splicing area definition module receives the planning data offset and, based on the difference between the geometric position offset value and the planning attribute, defines a splicing processing buffer at the preset docking edge of the adjacent target planning area. The splicing processing buffer includes geometrically overlapping areas and non-overlapping gap areas. The generation module generates a first stitching path for geometrically overlapping areas and a second stitching path for non-overlapping gap areas. The output module receives the first stitching path and the second stitching path, and performs fusion and reconstruction of the planning data in the geometrically overlapping area according to the first stitching path. At the same time, it performs adaptive interpolation of the planning data in the non-overlapping gap area according to the second stitching path, merges the processed adjacent target planning areas, generates a stitched planning spatial image, and calculates the global stitching consistency evaluation index of the stitched planning spatial image.

[0020] This system lays a data foundation for stitching by accurately quantifying the geometric position offset and planning attribute differences of the edges through the extraction module; intelligently delineating buffer zones containing geometrically overlapping areas and non-overlapping gap areas through the definition module; generating first and second stitching paths for the two types of areas respectively through the generation module, achieving differentiated and accurate fusion and interpolation; and finally, generating a seamless stitched image and providing a global stitching consistency evaluation index through the output module. This system systematically solves the problems of misalignment and conflict in dynamic planning data stitching, providing a reliable data base for high-quality spatial planning visualization analysis.

[0021] In this module, multiple target planning spatial images are retrieved and accessed from a preset geographic information system database or historical planning archive based on the timestamps and spatial range labels specified by the user. The pre-defined geographic information system database refers to a standardized spatial data warehouse that integrates multi-source, multi-temporal basic geographic information and thematic planning data. The standardized spatial data warehouse organizes and stores vector maps, remote sensing images, digital elevation models and various planning thematic layers according to a unified coordinate reference system, data model and metadata specifications.

[0022] The Historical Planning Archive is a structured electronic archive management system. The system archives and stores the results maps, text reports, approval documents and related policy documents of each planning project in chronological order. It also establishes an index based on planning project number, version number and timestamp, supporting rapid retrieval and version tracing based on spatiotemporal dimensions.

[0023] The method for identifying the edge of the planning region in each target planning spatial image includes: performing grayscale and binarization preprocessing on the target planning spatial image, extracting the contours of connected regions that represent different planning attributes in the target planning spatial image, and identifying the outer boundary of the connected region contour as the edge of the planning region; The default method for obtaining the docking edge is as follows: S1. After completing the edge identification of the planning area, the system calculates the minimum spatial proximity between each two planning area edges based on the spatial positional relationship of all planning area edges. The target planning areas corresponding to the two planning area edges with a minimum spatial proximity less than a preset threshold are identified as spatially adjacent areas, forming multiple pairs of candidate adjacent areas. The method for calculating minimum spatial proximity is as follows: S101. For any two planning area edges, calculate the Euclidean distance from each point on the first edge to all points on the second edge, and record the minimum value as the shortest distance from that point to the second edge. S102. Traverse all points on the first edge to obtain a set of nearest distance values. Take the minimum value in this set of nearest distance values ​​and define it as the minimum spatial proximity between the two planning area edges.

[0024] Preset threshold: Used to determine whether two planning areas are spatially adjacent. Its optimal value is usually set according to the spatial resolution of the target planning spatial image and the scale of the typical planning area. For example, for a planning map with a scale of 1:10000, the optimal value of the preset threshold can be set to the actual distance corresponding to 10 image pixels, or 50 meters based on experience.

[0025] S2. For each pair of candidate adjacent regions, extract continuous edge segments in the two planning regions whose spatial distance is less than a preset distance. Define each pair of extracted continuous edge segments as a candidate docking edge segment and calculate the average spatial distance and average planning attribute similarity between each pair of candidate docking edge segments. Preset distance: Used to filter out possible local edges in candidate adjacent region pairs. Its optimal value is usually slightly larger than the preset threshold for judging spatial adjacency mentioned above. For example, it can be set to 1.5 to 2 times the preset threshold to cover close edge segments caused by irregular edge undulations.

[0026] The steps for extracting continuous edge segments are as follows: On the edges of the two planning regions that are determined to be candidate adjacent regions, the edge point sequences are traversed respectively, and all continuous point sequences that satisfy "the distance between two adjacent points in the sequence is less than the preset step size" and "the shortest distance from the whole line segment to the other edge is always less than the preset distance" are identified. Each such continuous point sequence is defined as a continuous edge segment.

[0027] The method for calculating the average spatial distance and the average planning attribute similarity is as follows: S201. For the extracted pair of candidate docking edge segments, calculate the shortest distance from each sampling point on the first edge segment to the second edge segment, and calculate the arithmetic mean of the shortest distances of all sampling points to obtain the average spatial distance of the pair of candidate docking edge segments. S202. Extract the dominant planning attribute codes of the planning areas to which this pair of candidate docking edge segments are attached within their respective adjacent ranges. S203. Query the preset attribute similarity comparison table, obtain an attribute similarity score between 0 and 1 based on the matching relationship of the two dominant planning attribute codes, and directly use the attribute similarity score as the average planning attribute similarity of the two candidate docking edge segments.

[0028] S3. Based on the average spatial distance and average planning attribute similarity of each pair of candidate docking edge segments, construct a comprehensive proximity evaluation index, and determine the candidate docking edge segments whose comprehensive proximity evaluation index is higher than the preset screening threshold as the preset docking edge.

[0029] The steps for constructing the comprehensive proximity evaluation index are as follows: normalize the calculated average spatial distance and convert it into a proximity score between 0 and 1. At the same time, use the average planning attribute similarity as another score between 0 and 1. Assign preset geometric weights and attribute weights to these two scores respectively. Add the two weighted scores to obtain the comprehensive proximity evaluation index between 0 and 2.

[0030] Preset screening threshold: Used to select the most suitable docking edge from the candidate docking edge segments as the final processing object. Its optimal value needs to be determined based on a large number of experiments. It is usually set to a high standard, such as 1.5, to ensure that the selected edge segments have high docking feasibility in terms of geometry and properties.

[0031] The calculation steps for the planning data offset in the extraction module are as follows: S4. For each predetermined pre-defined docking edge, select multiple sampling points at predetermined intervals. For each sampling point, draw perpendicular lines from the normal direction of the pre-defined docking edge at the sampling point to two adjacent target planning areas, intersecting with the edges of their respective planning areas to obtain a pair of geometrically corresponding points. Calculate the Euclidean distance between each pair of geometrically corresponding points as the local geometric offset value of the sampling point. Statistically calculate the local geometric offset values ​​of all sampling points on the pre-defined docking edge, calculate their average value and standard deviation, and use the weighted sum of the average value and standard deviation as the geometric position offset value of the pre-defined docking edge. Preset interval: Used to control the density of sampling points. Its value is usually dynamically adjusted according to the total length and complexity of the preset docking edge.

[0032] The weighted sum of the mean and standard deviation refers to multiplying the average of the calculated local geometric offset values ​​by a first weighting coefficient, and then adding the standard deviation by a second weighting coefficient. The sum of the first and second weighting coefficients is 1. By adjusting these two weighting coefficients, the sensitivity of the final geometric position offset value to the overall deviation can be controlled.

[0033] S5. For the same pair of adjacent target planning areas, in the preset neighboring areas on both sides of the preset docking edge, extract the attribute feature vectors that reflect the semantic information of the planned land use nature and development intensity. Calculate the cosine similarity between the attribute feature vectors extracted from the two sides. Subtract the cosine similarity from 1 and define the value as the preliminary planning attribute difference at the preset docking edge. At the same time, consider the degree of deviation of the attribute evolution trend of the preset neighboring areas on both sides of the preset docking edge in the historical planning archive, and correct the preliminary planning attribute difference to obtain the planning attribute difference. The planning attribute difference and the geometric position offset value together constitute the planning data offset.

[0034] The preset adjacent area refers to the spatial range set on both sides of the preset docking edge for extracting attribute feature vectors. Its optimal value is usually set according to the typical influence radius of the planning element. For example, in land use planning, this range can be set to extend 100 meters from the docking edge to both sides.

[0035] Considering the degree of deviation in attribute evolution trends between the pre-defined adjacent areas on both sides of the pre-defined docking edge and the historical planning archive, the preliminary planning attribute difference is corrected, and the specific steps to obtain the planning attribute difference are as follows: S501. Extract the dominant attribute coding sequence of the pre-set adjacent areas on both sides of the pre-set docking edge in the past several planning cycles from the historical planning archive. S502. Calculate the difference between two attribute coding sequences at the same point in time, and count the proportion of their evolution direction (e.g., both becoming commercial or one becoming residential or one becoming ecological) being the same. Use this proportion as a trend consistency coefficient. Subtract this trend consistency coefficient from 1 to obtain a trend deviation correction amount. Add the initial planning attribute difference degree to this trend deviation correction amount to obtain the final planning attribute difference degree.

[0036] Among them, the geometrically overlapping area in the splicing area definition module refers to the area where the pre-set adjacent area planning elements on both sides of the pre-set docking edge spatially intrude into each other, and the non-overlapping gap area refers to the blank area formed when the pre-set adjacent area planning elements on both sides of the pre-set docking edge are not covered. The method for defining the splicing processing buffer is as follows: Q1. Receive the geometric position offset value, multiply the geometric position offset value by the preset base width coefficient to obtain the preliminary single-sided base width of the buffer. This preliminary single-sided base width of the buffer is used as the core initial parameter for defining the splicing processing buffer in subsequent steps. The preset base width coefficient is used to convert the dimensionless relative quantity of geometric position offset into a base value of buffer width with actual distance significance. Its optimal value needs to be combined with the spatial scale of the planning data. For example, in urban scale planning, it can be set to 10, which means that each unit of geometric position offset corresponds to a base buffer width of 10 meters.

[0037] Q2. Receive the planning attribute difference degree, compare the planning attribute difference degree with the preset low, medium and high threshold ranges, and determine the basic level coefficient according to the threshold range in which the planning attribute difference degree falls; combine the preset docking edge length and the frequency of historical planning changes, adjust the basic level coefficient, and generate the attribute difference correction factor; multiply the initial buffer single-sided basic width with the attribute difference correction factor to obtain the corrected buffer single-sided actual width; based on the buffer single-sided actual width, symmetrically expand outward along the preset docking edge to the adjacent target planning areas on both sides to directly generate the actual boundary of the splicing processing buffer; The preset three threshold ranges of low, medium and high are used to classify the difference of planning attributes in order to determine the basic level coefficient. For example, the low difference range can be set to [0, 0.3), the medium difference range to [0.3, 0.7), and the high difference range to [0.7, 1.0], which correspond to basic level coefficients of 1.0, 1.5 and 2.0 respectively.

[0038] The specific steps for adjusting the basic level coefficient and generating the attribute difference correction factor, based on the preset docking edge length and the frequency of historical planning changes, are as follows: Q201. Obtain the length value of the preset docking edge, and the number of attribute changes in the area near the edge in the last five planning versions as statistically analyzed from the historical planning archive. Normalize the length value and normalize the number of changes. Q202. Take the weighted average of the normalized length value and the number of changes to obtain an adjustment coefficient. Multiply this adjustment coefficient by the base level coefficient determined in step Q2 to obtain the final attribute difference correction factor.

[0039] Q3. Within the generated splicing processing buffer, based on the spatial distribution data of planning elements extracted from the original target planning spatial image, perform spatial overlay analysis to identify sub-regions that are simultaneously covered by planning elements of adjacent target planning regions on both sides, and mark them as geometrically overlapping regions. At the same time, identify sub-regions that are not covered by planning elements on either side, and mark them as non-overlapping gap regions, thus completing the internal structure division of the splicing processing buffer.

[0040] The method for extracting spatial distribution data of planning elements is as follows: object-oriented image segmentation is performed on the original target planning spatial image, adjacent pixels with similar spectral and texture features are merged into homogeneous objects, and the geometric shape, area, location, and planning element category code associated from image metadata or attribute table are extracted from each object to form a spatial distribution dataset of planning elements.

[0041] The steps for generating the first concatenated path in the generation module are as follows: R1. For the geometrically overlapping regions that need to be processed, the geometrically overlapping regions are subdivided into multiple geometrically continuous overlapping sub-regions with relatively consistent internal attributes. The geometric center point, area, and planning attribute codes of the adjacent target planning regions on both sides covered by each overlapping sub-region are extracted. A description unit containing geometric shape and attribute information is established for each overlapping sub-region. The geometric center point of each overlapping sub-region is obtained by calculating the average coordinates of all boundary points of that sub-region, and the area is obtained by calculating the geometric area of ​​the polygon of that sub-region.

[0042] R2. Based on the description unit, evaluate the compatibility level of the planning attributes of the adjacent target planning areas on both sides. The compatibility level is determined by comparing the relationship between the planning attribute codes on both sides in the preset planning attribute compatibility lookup table. Based on the evaluated compatibility level, assign preset fusion rules to each overlapping sub-region. The fusion rules define the attribute inheritance priority to be adopted when the attributes on both sides are compatible, and the attribute negotiation or creation logic to be activated when the attributes on both sides conflict. Specifically: when the compatibility level is "highly compatible" (such as residential land and ancillary commercial land), the fusion rule is set to "primary attribute inheritance + secondary attribute supplementation", that is, the attribute code of the planning area with a larger area proportion is retained as the basic attribute, while the attribute code of the other side is embedded as a supplementary attribute in the attribute description field; when the compatibility level is "moderately compatible" (such as industrial land and warehousing land), the fusion rule adopts a "weighted fusion" strategy, based on the attributes on both sides. The area proportion of overlapping sub-regions is dynamically weighted, and the quantitative indicators (such as plot ratio and building density) in the planning attribute code are calculated by weighted average. For qualitative indicators (such as land use nature), they are converted into intermediate transitional attributes through a preset attribute mapping table. When the compatibility level is "low compatibility" (such as ecological protection land and commercial development land), the fusion rule activates the "conflict marking + manual intervention" mechanism. The system automatically marks the overlapping sub-region as an attribute conflict area and highlights it in the visualization interface. At the same time, the original attribute codes and historical planning basis of both sides are extracted to generate a conflict analysis report for planners to make manual decisions.

[0043] The preset planning attribute compatibility lookup table is a two-dimensional relational matrix. The rows and columns of this matrix are all possible planning attribute codes. Each cell in the matrix stores the compatibility level between the corresponding two attribute codes. This compatibility level is predefined by planning experts based on relevant regulations and planning principles.

[0044] R3. Based on the fusion rules assigned to each overlapping sub-region and the geometric features of the overlapping sub-region, calculate the first final planning attribute value of each first location point to be determined within each overlapping sub-region. Generate a first spatial interpolation path for each overlapping sub-region to guide the fusion of its internal attributes and spatial allocation according to the first final planning attribute value. Connect and smooth the transition of the first spatial interpolation paths of all overlapping sub-regions according to their spatial adjacency relationship and integrate them to form a first splicing path. The steps for generating the first spatial interpolation path are as follows: R301. Based on the calculated first final planning attribute value of each first location point, construct an attribute value spatial distribution field covering the entire overlapping sub-region. R302. Calculate the direction and magnitude of the attribute gradient in this attribute value space distribution field; R303. Starting from one side boundary of the overlapping sub-region, trace along a path orthogonal to the attribute gradient direction. The attribute value changes most gently along this path. Use the best path obtained by tracing as the first spatial interpolation path for the overlapping sub-region.

[0045] The calculation method for the first final planning attribute value in step R3 is as follows: R301. Obtain the fusion rule assigned to each overlapping sub-region. The fusion rule pre-sets the weight calculation function and attribute interpolation model applicable to the overlapping sub-region. According to the fusion rule, for each first position point to be determined inside the overlapping sub-region, calculate the normalized distance from the first position point to the geometric boundaries of the two adjacent target planning regions covered by the overlapping sub-region, and generate distance weight pairs. A weighting function is a mathematical function that maps distance to weights, such as the inverse distance weighting function. An attribute interpolation model is an algorithm that predicts the attribute values ​​of unknown points based on spatial proximity and attribute similarity, such as the Kriging interpolation model.

[0046] The first location point specifically refers to the spatial discrete point in the geometrically overlapping area that needs to be used for attribute fusion calculation. The density of these points determines the precision of the fusion reconstruction result.

[0047] R302. Based on the distance weight pair, combined with the planning attribute codes of the two adjacent target planning areas obtained from the description unit, the preset attribute compatibility matrix is ​​queried to obtain the dynamic similarity coefficient between the planning attribute codes of the two adjacent target planning areas. The distance weight pair is corrected using the dynamic similarity coefficient to generate the fusion weight pair. The standard attribute values ​​corresponding to the planning attribute codes of the two adjacent target planning areas are multiplied by the corresponding weights in the fusion weight pair and then summed to obtain the first final planning attribute value of the first position point. The calculation process traverses all first position points in the overlapping sub-region.

[0048] The preset attribute compatibility matrix is ​​a two-dimensional relationship table commonly used in existing technologies to quantitatively assess the compatibility strength between different land use or functional types. For details, please refer to the compatibility provisions provided in relevant standards or specifications for urban and rural planning land use compatibility.

[0049] The steps for generating the second concatenation path in the generation module are as follows: R4. For non-overlapping gap areas, analyze the spatial geometric characteristics of the non-overlapping gap areas and the effective areas with clearly defined planning attributes adjacent to them within a preset range. Extract the planning attribute codes and spatial distribution patterns of the effective areas as usable planning attribute information to fill the non-overlapping gap areas. The preset range refers to the spatial radius for searching for valid attribute information in the surrounding area when filling non-overlapping gap areas. Its optimal value is usually set to 1 to 2 times the length of the diagonal of the smallest outer rectangle of the gap area.

[0050] R5. Based on available planning attribute information, calculate the attribute influence weight of each second location point at the boundary of the non-overlapping gap area to different effective areas, and evaluate it according to the preset planning logic consistency rule. Determine the dominant attribute source or generate new transition attribute codes for different locations within the non-overlapping gap area to form a gap filling strategy. Pre-defined rules for consistency in planning logic include the rationality of land use transitions and the continuity of infrastructure.

[0051] The second location point specifically refers to a spatial discrete point within a non-overlapping gap region where attribute interpolation calculations are required.

[0052] The method for calculating the attribute influence weight is as follows: calculate the Euclidean distance from the second location point to the geometric boundary of each effective region, take the reciprocal of all distance values ​​and normalize them so that the sum of the weights from all effective regions is 1. This normalized reciprocal is used as the attribute influence weight of each effective region on the second location point.

[0053] The methods for forming gap-filling strategies include: selecting the effective region with the largest weight as the dominant attribute source based on the calculated attribute influence weight; or when the weights of multiple effective regions are close and their attributes are different, generating a new transition attribute code that integrates multiple attributes according to the planning logic consistency rule.

[0054] R6. Based on the gap filling strategy, calculate the second final planning attribute value for each second location point that needs to be filled in the non-overlapping gap area. Connect the calculation process of the second final planning attribute value according to the spatial proximity relationship of the second location points to form a continuous second spatial interpolation path covering the entire non-overlapping gap area. Define the second spatial interpolation path as the second splicing path.

[0055] The steps for calculating the second final planning attribute value are as follows: R601. According to the gap filling strategy, if the dominant attribute source is adopted, the standard attribute value corresponding to the planning attribute code of the dominant effective area is directly assigned to the second location point. R602. If it is necessary to generate transitional attributes, the standard attribute values ​​corresponding to the planning attribute codes of all valid areas shall be weighted and merged according to their attribute influence weights, and the calculation result shall be used as the transitional attribute value. R603. The value obtained by directly assigning or weighting the above values ​​shall be used as the second final planning attribute value of the second location point.

[0056] The method for generating the second spatial interpolation path is as follows: R604. Calculate the second final planning attribute values ​​of all second location points within the non-overlapping gap region to form an attribute distribution field; R605. In an attribute distribution field, identify the ridges or valleys of attribute value changes. R606. Along the direction of the most significant attribute change, connect a series of second position points to form a core path that guides the diffusion trend of the interpolated attribute space, namely the second space interpolation path.

[0057] The steps in the output module, namely, fusing and reconstructing the planning data within the geometrically overlapping area according to the first splicing path, and adaptively interpolating the planning data within the non-overlapping gap area according to the second splicing path, are as follows: T1. Receive the first stitching path transmitted by the generation module, and according to the spatial interpolation rules of each overlapping sub-region defined in the first stitching path, assign the calculated first final planning attribute value to each first location point in the geometric overlapping region, complete the reconstruction of the planning attributes of all pixels or features in the entire geometric overlapping region, and generate the processed geometric overlapping region data. T2. Receive the second stitching path, and according to the gap filling strategy and the second spatial interpolation path defined in the second stitching path, assign the calculated second final planning attribute value to each second location point in the non-overlapping gap area, complete the interpolation and filling of the planning attributes in the entire non-overlapping gap area, and generate the processed non-overlapping gap area data.

[0058] Specifically, the output module merges the processed adjacent target planning areas by performing geometric alignment and seamless stitching of the processed geometrically overlapping area data and the processed non-overlapping gap area data, while integrating the internal area data of the original target planning area that is not affected by the stitching processing buffer, to generate a complete stitched planning space image.

[0059] The calculation method for the global splicing consistency evaluation index in the output module is as follows: T3. In the generated stitched planning space image, several evaluation sample points are arranged along the center line of the stitching processing buffer at a preset density. For each evaluation sample point, the planning attribute code corresponding to the first position point or the second position point is extracted from the original target planning space image adjacent to its left and right sides, and the difference value between the two planning attribute codes is calculated as the local attribute difference measure of the first position point or the second position point. The preset density refers to the spacing between evaluation sample points arranged along the center line of the splicing processing buffer. Its optimal value should be able to effectively capture attribute changes within the buffer. For example, it can be set to one-tenth of the actual width of one side of the buffer.

[0060] The method for calculating the difference between two planning attribute codes is as follows: query the preset attribute difference matrix, which defines the semantic difference between any two planning attribute codes, with a value range of 0 to 1, and directly read the cell value of the corresponding two codes in the matrix as their difference value.

[0061] T4. Based on the local attribute difference measure of all evaluation sample points, calculate its mean and standard deviation, and combine the geometric position offset value and the planning attribute difference degree to generate a global stitching consistency evaluation index through a weighted fusion formula; the lower the global stitching consistency evaluation index, the smoother the transition of planning attributes and the more consistent the logic of the stitched planning spatial image in the stitching processing buffer and its surrounding area.

[0062] The weighted fusion formula is as follows: Global splicing consistency evaluation index = (average value of local attribute difference measurement * weight W1) + (standard deviation of local attribute difference measurement * weight W2) + (geometric position offset value * weight W3) + (planning attribute difference degree * weight W4), where W1, W2, W3, and W4 are preset weight coefficients, and their sum is 1.

[0063] The output module also includes rendering the stitched planned spatial image and the global stitching consistency evaluation index together onto a preset visualization interface, generating an interactive visualization analysis report containing the evaluation result layer and the planned image layer, for users to conduct spatial planning analysis and decision-making.

[0064] The preset visualization interface is a web-based or desktop-based graphical user interaction platform that integrates map display, layer control, attribute query, spatial measurement, and report generation functions to host and display interactive visualization analysis reports.

[0065] Example In a new town planning project in a certain city, the system extraction module retrieved two target planning spatial images to be stitched from the preset geographic information system database and historical planning archive, based on the timestamps 2020 and 2023 and the spatial range labels "Area A" and "Area B".

[0066] Each image is preprocessed by grayscale and binarization, and the contours of connected regions are extracted and their outer boundaries are identified as the edges of the planned region.

[0067] The system calculates the minimum spatial proximity between the edges of every two planning areas, and determines the target planning areas corresponding to the two edges that are less than 50 meters apart as spatially adjacent areas, forming candidate adjacent area pairs.

[0068] For the candidate adjacent region pair, continuous edge segments with a spatial distance of less than 100 meters are extracted and defined as candidate docking edge segments. The average spatial distance is calculated to be 15 meters, and the average planning attribute similarity is obtained by querying the preset attribute similarity table, which is 0.8. Based on this, a comprehensive proximity evaluation index of 1.7 is constructed, which is higher than the preset screening threshold of 1.5. Therefore, the candidate docking edge segment is determined as the preset docking edge.

[0069] Sampling points were selected at 5-meter intervals on the preset docking edge, and the calculated geometric position offset value was 8.2 meters. Attribute feature vectors were extracted from the preset neighboring areas within 100 meters on both sides of the preset docking edge, and the cosine similarity was calculated to obtain a preliminary planning attribute difference of 0.25. Combining the attribute evolution trend extracted from the historical planning archive, a trend deviation correction of 0.1 was calculated, resulting in a planning attribute difference of 0.35.

[0070] The splicing area definition module multiplies the geometric position offset value of 8.2 with the preset foundation width coefficient of 10 to obtain an initial buffer zone foundation width of 82 meters on one side. Since the planning attribute difference degree of 0.35 falls within the preset medium difference degree range, the foundation level coefficient is determined to be 1.5. Combining the docking edge length and the frequency of historical changes, an attribute difference correction factor of 1.2 is generated, and the actual width of the buffer zone on one side is calculated to be 98.4 meters. The splicing processing buffer zone is then symmetrically expanded outward based on this.

[0071] Within this stitching processing buffer, spatial overlay analysis is performed based on the spatial distribution data of planning elements extracted from the original image to identify geometrically overlapping areas and non-overlapping gap areas. The generation module further subdivides the geometrically overlapping areas into multiple overlapping sub-regions, extracting the geometric center point, area, and planning attribute codes on both sides of each overlapping sub-region to establish a description unit.

[0072] Based on a pre-defined planning attribute compatibility comparison table, the attribute compatibility level on both sides is assessed as "moderately compatible," and a "weighted fusion" rule is assigned to the overlapping sub-regions. According to this rule, normalized distances to the two boundary lines are calculated for each first location point within the overlapping sub-region, generating distance weight pairs. A dynamic similarity coefficient of 0.7 is obtained by querying the pre-defined attribute compatibility matrix, and the distance weight pairs are adjusted to generate fusion weight pairs. The standard attribute values ​​on both sides are weighted and summed to obtain the first final planning attribute value for each first location point. Based on this, a first spatial interpolation path is generated and integrated to form the first stitching path. For non-overlapping gap areas, the effective area within a pre-defined 200-meter radius is analyzed, and usable planning attribute information is extracted.

[0073] The attribute influence weights from each second location point at the boundary of the gap region to each effective region are calculated. A gap-filling strategy is formed based on the planning logic consistency rule. Accordingly, a second final planning attribute value is calculated for each second location point, forming a second spatial interpolation path, i.e., a second stitching path. The output module reconstructs the geometrically overlapping region data according to the first stitching path, generating processed geometrically overlapping region data. Simultaneously, it performs attribute interpolation on the non-overlapping gap regions according to the second stitching path, generating processed non-overlapping gap region data. These two data are then merged with the internal region data unaffected by the buffer to generate a stitched planned spatial image.

[0074] Evaluation sample points are laid out along the centerline of the splicing buffer at a preset density of 9.8 meters. The local attribute difference measure of each sample point is calculated using a preset attribute difference matrix, with a mean of 0.12 and a standard deviation of 0.05. Combined with the geometric position offset value of 8.2 and the planning attribute difference of 0.35, a global splicing consistency evaluation index of 1.85 is calculated using a weighted fusion formula. Finally, the spliced ​​planned spatial image and this index are rendered together onto a preset visualization interface to generate an interactive visualization analysis report.

[0075] This embodiment demonstrates that the system intelligently delineates buffer zones through quantitative geometry and attribute offset, and generates differentiated stitching paths for overlapping and gap regions, achieving high-precision and logically consistent seamless integration of dynamically changing planning data. The generated stitched images and quantitative evaluation indicators provide an accurate and reliable visual data base for planning analysis, significantly improving the scientific nature and efficiency of decision-making.

[0076] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A data visualization based spatial planning analysis system, characterized in that: It includes an extraction module, a splicing region definition module, a generation module, and an output module; The extraction module is used to acquire multiple target planning space images to be stitched together, identify the planning area edges in each target planning space image, and calculate the planning data offset of adjacent target planning areas at a preset docking edge. The planning data offset includes geometric position offset value and planning attribute difference degree. The splicing area definition module receives the planning data offset and, based on the difference between the geometric position offset value and the planning attribute, defines a splicing processing buffer at the preset docking edge of the adjacent target planning area. The splicing processing buffer includes a geometrically overlapping area and a non-overlapping gap area. The generation module generates a first splicing path for the geometrically overlapping area and a second splicing path for the non-overlapping gap area. The output module receives the first stitching path and the second stitching path, and performs fusion and reconstruction of the planning data in the geometrically overlapping area according to the first stitching path. At the same time, it performs adaptive interpolation of the planning data in the non-overlapping gap area according to the second stitching path, merges the processed adjacent target planning areas, generates a stitched planning spatial image, and calculates the global stitching consistency evaluation index of the stitched planning spatial image.

2. The spatial planning analysis system based on data visualization of claim 1, wherein: The extraction module retrieves multiple target planning spatial images by searching and calling them from a preset geographic information system database or historical planning archive based on the user-specified timestamp and spatial range label. The method for identifying the edge of the planning region in each of the target planning spatial images includes: performing grayscale and binarization preprocessing on the target planning spatial image, extracting the contours of connected regions representing different planning attributes in the target planning spatial image, and identifying the outer boundary of the connected region contour as the edge of the planning region; The method for obtaining the preset docking edge is as follows: S1. After completing the edge identification of the planning area, the system calculates the minimum spatial proximity between each two planning area edges based on the spatial positional relationship of all planning area edges, and determines the target planning areas corresponding to the two planning area edges whose minimum spatial proximity is less than a preset threshold as spatially adjacent areas, forming multiple sets of candidate adjacent area pairs. S2. For each pair of candidate adjacent regions, extract continuous edge segments in the two planning regions whose spatial distance is less than a preset distance. Define each pair of extracted continuous edge segments as a candidate docking edge segment and calculate the average spatial distance and average planning attribute similarity between each pair of candidate docking edge segments. S3. Based on the average spatial distance and average planning attribute similarity of each pair of candidate docking edge segments, construct a comprehensive proximity evaluation index, and determine the candidate docking edge segments whose comprehensive proximity evaluation index is higher than a preset screening threshold as preset docking edges.

3. The spatial planning analysis system based on data visualization of claim 2, wherein: The calculation steps for the planning data offset in the extraction module are as follows: S4. For each of the predetermined docking edges, select multiple sampling points at predetermined intervals. For each sampling point, draw perpendicular lines from the normal direction of the predetermined docking edge at the sampling point to two adjacent target planning areas, intersecting with the edges of their respective planning areas to obtain a pair of geometrically corresponding points. Calculate the Euclidean distance between each pair of geometrically corresponding points as the local geometric offset value of the sampling point. The local geometric offset values ​​of all sampling points on the preset docking edge are statistically analyzed, and their average value and standard deviation are calculated. The weighted sum of the average value and the standard deviation is used as the geometric position offset value of the preset docking edge. S5. For the same pair of adjacent target planning areas, in the preset neighboring areas on both sides of the preset docking edge, extract the attribute feature vectors that reflect the semantic information of the planned land use nature and development intensity, calculate the cosine similarity between the attribute feature vectors extracted from the two sides, and define the value obtained by subtracting the cosine similarity from 1 as the preliminary planning attribute difference at the preset docking edge. At the same time, considering the degree of deviation of the attribute evolution trend of the preset neighboring areas on both sides of the preset docking edge in the historical planning archive, the preliminary planning attribute difference is corrected to obtain the planning attribute difference. The planning attribute difference and the geometric position offset value together constitute the planning data offset.

4. The spatial planning analysis system based on data visualization of claim 3, wherein: In the splicing area definition module, the geometrically overlapping area refers to the area where the pre-defined adjacent area planning elements on both sides of the pre-defined docking edge spatially intrude into each other, and the non-overlapping gap area refers to the blank area formed when the pre-defined adjacent area planning elements on both sides of the pre-defined docking edge are not covered. The method for defining the splicing processing buffer is as follows: Q1. Receive the geometric position offset value, multiply the geometric position offset value by a preset base width coefficient to obtain the preliminary single-sided base width of the buffer. Q2. Receive the planning attribute difference degree, compare the planning attribute difference degree with three preset threshold intervals (low, medium, and high), and determine the basic level coefficient based on the threshold interval where the planning attribute difference degree falls; adjust the basic level coefficient by combining the length of the preset docking edge and the frequency of historical planning changes to generate an attribute difference correction factor; multiply the initial buffer single-sided basic width by the attribute difference correction factor to obtain the corrected buffer single-sided actual width; using the buffer single-sided actual width as a reference, symmetrically expand outward along the preset docking edge to the adjacent target planning areas on both sides to directly generate the actual boundary of the splicing processing buffer. Q3. Within the generated splicing processing buffer, based on the spatial distribution data of planning elements extracted from the original target planning spatial image, perform spatial overlay analysis to identify sub-regions that are simultaneously covered by planning elements of the adjacent target planning regions on both sides, and mark them as geometrically overlapping regions. At the same time, identify sub-regions that are not covered by planning elements on either side, and mark them as non-overlapping gap regions, thus completing the internal structure division of the splicing processing buffer.

5. The spatial planning analysis system based on data visualization of claim 4, wherein: The steps for generating the first spliced ​​path in the generation module are as follows: R1. For the geometrically overlapping region that needs to be processed, the geometrically overlapping region is subdivided into multiple geometrically continuous overlapping sub-regions with relatively consistent internal attributes. The geometric center point, area, and planning attribute codes of the adjacent target planning regions on both sides covered by each overlapping sub-region are extracted. A description unit containing geometric shape and attribute information is established for each overlapping sub-region. R2. Based on the description unit, evaluate the compatibility level of the planning attributes of the adjacent target planning areas on both sides. The compatibility level is determined by comparing the relationship between the planning attribute codes on both sides in a preset planning attribute compatibility lookup table. Based on the evaluated compatibility level, assign a preset fusion rule to each overlapping sub-region. R3. Based on the fusion rules assigned to each overlapping sub-region and the geometric features of the overlapping sub-region, calculate the first final planning attribute value of each first location point to be determined within each overlapping sub-region, generate a first spatial interpolation path for each overlapping sub-region according to the first final planning attribute value, connect and smooth the first spatial interpolation paths of all overlapping sub-regions according to their spatial adjacency relationship, and integrate them to form a first splicing path. The method for calculating the first final planning attribute value in step R3 is as follows: R301. Obtain the fusion rule assigned to each of the overlapping sub-regions. The fusion rule pre-sets the weight calculation function and attribute interpolation model applicable to the overlapping sub-regions. According to the fusion rule, for each first location point to be determined inside the overlapping sub-region, calculate the normalized distance from the first location point to the geometric boundaries of the adjacent target planning regions on both sides covered by the overlapping sub-region, and generate distance weight pairs. R302. Based on the distance weight pair, and combined with the planning attribute codes of the two adjacent target planning regions obtained from the description unit, a preset attribute compatibility matrix is ​​queried to obtain the dynamic similarity coefficient between the planning attribute codes of the two adjacent target planning regions. The distance weight pair is corrected using the dynamic similarity coefficient to generate a fusion weight pair. The standard attribute values ​​corresponding to the planning attribute codes of the two adjacent target planning regions are multiplied by the corresponding weights in the fusion weight pair and then summed to obtain the first final planning attribute value of the first location point. The calculation process traverses all the first location points in the overlapping sub-region.

6. The spatial planning analysis system based on data visualization of claim 5, wherein: The steps for generating the second splicing path in the generation module are as follows: R4. For the non-overlapping gap area, analyze the spatial geometric characteristics of the non-overlapping gap area and the effective areas with clearly defined planning attributes adjacent to it within a preset range, and extract the planning attribute codes and spatial distribution patterns of the effective areas as usable planning attribute information for filling the non-overlapping gap area. R5. Based on the available planning attribute information, calculate the attribute influence weight from each second location point at the boundary of the non-overlapping gap area to different effective areas, and evaluate it according to the preset planning logic consistency rule to determine the dominant attribute source or generate a new transition attribute code for different locations inside the non-overlapping gap area, thus forming a gap filling strategy. R6. Based on the gap filling strategy, calculate the second final planning attribute value for each second location point that needs to be filled in the non-overlapping gap area. Connect the calculation process of the second final planning attribute value according to the spatial proximity relationship of the second location points to form a continuous second spatial interpolation path covering the entire non-overlapping gap area. Define the second spatial interpolation path as the second splicing path.

7. The spatial planning analysis system based on data visualization of claim 6, wherein: The steps in the output module to fuse and reconstruct the planning data within the geometrically overlapping region according to the first splicing path, and to adaptively interpolate the planning data within the non-overlapping gap region according to the second splicing path, are as follows: T1. Receive the first splicing path transmitted by the generation module, and according to the spatial interpolation rules of each overlapping sub-region defined in the first splicing path, assign the calculated first final planning attribute value to each first position point in the geometric overlapping region, complete the planning attribute reconstruction of all pixels or elements in the entire geometric overlapping region, and generate the processed geometric overlapping region data. T2. Receive the second splicing path, and according to the gap filling strategy defined in the second splicing path and the second spatial interpolation path, assign the calculated second final planning attribute value to each second location point in the non-overlapping gap area, complete the interpolation and filling of the planning attributes in the entire non-overlapping gap area, and generate the processed non-overlapping gap area data.

8. The spatial planning analysis system based on data visualization of claim 7, wherein: The specific content of the output module merging the processed adjacent target planning areas is as follows: the processed geometrically overlapping area data and the processed non-overlapping gap area data are geometrically aligned and seamlessly stitched together, while integrating the internal area data of the original target planning area that is not affected by the stitching processing buffer, to generate a complete stitched planning space image.

9. The spatial planning analysis system based on data visualization of claim 8, wherein: The calculation method for the global splicing consistency evaluation index in the output module is as follows: T3. In the generated stitched planning space image, several evaluation sample points are arranged along the center line of the stitching processing buffer at a preset density. For each evaluation sample point, the planning attribute code corresponding to the first location point or the second location point is extracted from the original target planning space image adjacent to its left and right sides, and the difference value between the two planning attribute codes is calculated as the local attribute difference measure of the first location point or the second location point. T4. Based on the local attribute difference measure of all the evaluation sample points, calculate its mean and standard deviation, and combine the geometric position offset value and the planning attribute difference degree to generate a global splicing consistency evaluation index through a weighted fusion formula.

10. The spatial planning analysis system based on data visualization of claim 9, wherein: The output module also includes rendering the stitched planned spatial image and the global stitching consistency evaluation index together onto a preset visualization interface to generate an interactive visualization analysis report containing an evaluation result layer and a planned image layer.

Images