Building concession distance examination method based on CIM

Abstract

The invention relates to the field of building engineering planning, and provides a CIM-based building concession distance examination method, which aims at a building concession road red line single examination task in city information model CIM examination, calculates the shortest distance between a building and a road, compares the shortest distance with a theoretical minimum concession distance, and constructs a two-dimensional and three-dimensional collaborative examination mechanism to obtain a building concession distance. And automatic verification of the spatial relationship between the building contour and the road red line is realized. According to the method, the problems of accurate calculation and three-dimensional violation judgment of the building concession distance are mainly solved by taking a model passing the BIM examination as basic input and combining road red line data in a city information model CIM environment.

Description

Technical Field

[0001] This invention relates to the field of intelligent engineering planning technology, and more specifically, to a method for reviewing building setback distances based on CIM. Background Technology

[0002] In the construction project planning permit review system, Building Information Modeling (BIM) and City Information Modeling (CIM) technologies have formed a clear division of labor: BIM review mainly focuses on the geometric and attribute compliance verification of the internal components of the building model, including the self-consistency check of indicators such as building height, floor height, and building area, which falls under the scope of internal review of individual buildings; CIM review focuses on the interaction between the building model and the urban spatial environment, involving the verification of external spatial relationships such as building setbacks from road red lines and building spacing control.

[0003] As the legally mandated control line for urban planning (according to the "Guidelines for Review of Construction Project Design Schemes"), the setback distance of the road red line is a core control indicator for construction project planning permits. In actual engineering projects, building setback reviews face multiple technical challenges: modern building outlines are often irregular curves (such as arched facades and sawtooth roofs), increasing the difficulty of distance calculations; road red lines are mostly segmented linear or curved elements (such as intersections of urban arterial roads), requiring dynamic adaptation to local curvature; cantilevered components, canopies, and other three-dimensional elements may encroach on the red line area in the vertical direction, but their projected planar display must comply with regulations.

[0004] Currently, the following technical solutions are mainly used for reviewing building setbacks from road boundaries:

[0005] Two-dimensional planar review method: Export the BIM model as a two-dimensional outline (usually in DXF format), overlay it with the road boundary CAD layer, and determine the building's integrity by calculating the shortest Euclidean distance between the building outline polygon and the road boundary line segment. This method only verifies the projected plane and ignores the risk of intrusion in the building's height direction (e.g., the ground floor shop canopy exceeds the boundary but the projection is compliant). For non-convex polygon outlines, traditional algorithms only check vertex distances and cannot accurately locate the shortest distance point that may exist at the midpoint of the edge. When the road boundary is a segmented or curved feature, additional data preprocessing is required, and simplified processing leads to calculation errors.

[0006] The 3D spatial inspection method involves extending the road boundary line into a 3D buffer zone (e.g., a 5-meter-wide column) and performing full collision detection with the BIM model, outputting the intrusion volume or Boolean value result. This method has significant problems: the road boundary line is essentially a 2D linear element, and full 3D detection of the model results in over 90% of the space being invalid detection areas, leading to a serious waste of computational resources; the collision results (intrusion volume) are difficult to directly convert into the required planar setback distance (e.g., the "5-meter setback" requirement); and the detection of small-scale violations (e.g., an air conditioner unit protruding 0.3 meters) requires high-precision meshes, resulting in high computational costs and a high risk of false alarms. Summary of the Invention

[0007] This invention addresses the technical problems existing in the prior art by providing a CIM-based method for reviewing building setback distances. By constructing a two-dimensional and three-dimensional collaborative review mechanism, it achieves automated verification of the spatial relationship between building outlines and road red lines.

[0008] This invention provides a CIM-based method for reviewing building setback distances, comprising:

[0009] S1, acquire building foundation data and road boundary data;

[0010] S2, decompose the boundary of each plot to which each building belongs, obtain multiple line segments of the plot boundary, and calculate the unit direction vector and normal vector of each line segment;

[0011] S3, taking each building as the center, emits query rays in the direction of the normal vector of each line segment of the corresponding sub-plot boundary, and establishes the relationship between buildings and roads through spatial intersection analysis;

[0012] S4. Obtain the height of each building. Based on the height of each building and the associated road boundary attribute information, obtain the theoretical minimum setback distance of each building relative to the road by querying.

[0013] S5. For each building base outline and road red line, two point sets are generated using adaptive high-density interpolation. The two point sets are then used to accelerate the nearest neighbor search through spatial indexing. The shortest Euclidean distance between all point pairs in the two point sets is calculated, and the shortest distance point pairs are located.

[0014] S6: Determine if the shortest distance between each building and the road is within the warning distance range. If yes, end the process; otherwise, generate a review report and proceed to S9.

[0015] S7. Based on the nearest point from each building to the road and the corresponding edge length, determine the linear range of the road red line along the road direction. Based on the linear range of the road red line, stretch it towards the building direction to generate a local three-dimensional road buffer model with the same height as the building and the same depth as the theoretical minimum setback distance.

[0016] S8. The generated local road 3D buffer model and building BIM model are compared using a bounding box-based fast collision detection algorithm to detect whether building components intrude into the road buffer zone.

[0017] S9 calculates and outputs the collision detection results, identifies the component ID, location coordinates, and intrusion depth value that intrude into the road buffer zone in the building BIM model, and outputs a 3D schematic diagram of the collision area.

[0018] This invention provides a CIM-based method for reviewing building setback distances. Addressing the single-item review task of building setbacks to road boundaries in CIM reviews, it achieves automated verification of the spatial relationship between building outlines and road boundaries by constructing a two-dimensional and three-dimensional collaborative review mechanism. This invention uses a model approved through BIM review as input, combined with road boundary data from the CIM environment, focusing on solving the problems of accurate calculation of building setback distances and three-dimensional violation determination. Attached Figure Description

[0019] Figure 1 The flowchart illustrates a CIM-based building setback distance review method according to one embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In addition, the technical features of the various embodiments or individual embodiments provided by the present invention can be arbitrarily combined with each other to form feasible technical solutions. Such combinations are not constrained by the order of steps and / or structural composition patterns, but must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0021] This invention provides a CIM-based method for reviewing building setback distances. Its core lies in rapidly screening global compliance through high-precision two-dimensional calculations and accurately triggering targeted three-dimensional local verifications, thus balancing accuracy and efficiency in the review process. The entire process organically combines two-dimensional planar distance calculations with three-dimensional spatial collision analysis, forming a complete automated review loop.

[0022] This method first processes data and establishes relationships in a two-dimensional plane, using a high-precision algorithm to calculate the shortest distance between the building foundation and the road boundary line, completing an initial screening. Then, it dynamically generates corresponding local three-dimensional road buffer models for the "suspected violations" identified in the two-dimensional screening, and performs precise spatial collision analysis with the building's BIM model. Finally, it reverse-maps the three-dimensional collision results back to the two-dimensional plane, generating a review report and visualization results that comply with planning regulations. This method effectively solves the core problems of missed three-dimensional violations in purely two-dimensional reviews, low computational efficiency in purely three-dimensional reviews, and the disconnect between review results and regulatory provisions.

[0023] Figure 1 A flowchart of a CIM-based building setback distance review method according to an embodiment of the present invention is shown, as follows: Figure 1 As shown, the method includes:

[0024] S1, acquire building foundation data and road boundary data.

[0025] Understandably, two key input data layers are obtained from the CIM platform: first, the building base layer from the building review drawings, which contains the projected planar outline of the building to be reviewed (usually a closed polygon); and second, the road layer, which contains information on all road boundary lines within the review area (usually linear features). Data is read through spatial database interfaces (such as OGCWFS services) or files (such as SHP, DWG) and converted into an internally unified coordinate reference system, laying the data foundation for subsequent spatial analysis.

[0026] S2 decomposes the boundary of each plot to which a building belongs, resulting in multiple line segments of the plot boundary. Calculate the unit direction vector and normal vector of each line segment.

[0027] Understandably, after obtaining the polygonal outline of each building base, all building base polygonal outlines are traversed. First, a geometric validity verification is performed (checking for issues such as self-intersection and duplicate vertices) to ensure the legality of the geometric entities. Subsequently, spatial relationship judgment (using spatial predicates such as ST_Within and ST_Contains) is used to verify whether each building base is completely contained within the boundary polygon of its respective land parcel, ensuring the spatial logic correctness of the "building-land parcel" relationship and eliminating basic errors such as buildings exceeding the land use boundary.

[0028] To facilitate directional setback analysis, the boundary polygon of each building's sub-plot is decomposed into a series of interconnected, independent line segments. For each segment, its direction vector from the starting point to the ending point is calculated, and its unit normal vector (indicating the "setback direction" perpendicular to the boundary) is further obtained through vector operations. This step transforms the planar spatial relationship problem into a discrete linear relationship problem, laying a precise spatial benchmark for subsequent directional-sensitive setback analysis and effectively solving the technical challenge of existing technologies that cannot distinguish between different setback requirements of different building boundaries facing the road.

[0029] The decomposition algorithm for the polygon boundaries of the plots includes the following steps: (1) Extraction of polygon vertex sequences:

[0030] Read the WKT or GeoJSON format data of the plot boundary polygons and extract the vertex coordinate sequence of the plot boundary polygons. ,in (Closed polygon). Verify the topological correctness of the vertex sequence, ensuring there are no self-intersecting or duplicate vertices.

[0031] (2) Generation of line segment units:

[0032] Construct a set of line segment units in vertex order:

[0033] Each line segment From the starting point and the end point definition, From 0 to ,when At that time, the destination was Assign a unique identifier SID = [plot ID]_[segment number] to each line segment, preserve the topological relationship information of the original polygon, and record the predecessor and successor line segments of each line segment.

[0034] (3) Handling of special circumstances:

[0035] For complex polygons containing islands and holes, only the outer boundary is processed; the inner islands and holes are not included in the setback analysis. For self-intersecting polygons, topology repair is performed before decomposition. For duplicate vertices, they are automatically merged into a single vertex to avoid generating zero-length line segments.

[0036] Calculate the direction vector and unit normal vector for each line segment, specifically including:

[0037] (1) Calculation of direction vector:

[0038] For line segments The starting coordinates are The endpoint coordinates are ;

[0039] Calculate the direction vector: ;

[0040] Calculate the length of the line segment: .

[0041] (2) Calculation of unit normal vector:

[0042] Calculate the normal vector: or It depends on the direction of retreat required.

[0043] This invention adopts the principle of "building setback" to determine that the direction of the normal vector points outward from the building, and calculates the unit normal vector: Verify the length of the unit normal vector: .

[0044] (3) Confirmation of retreat direction:

[0045] To verify whether the direction of the normal vector is correct, start from the midpoint of the line segment. Take a small distance along the direction of the normal vector checkpoint Whether it is located outside the building base polygon; if the point is located inside, take the opposite direction: .

[0046] For road segments near the boundaries of sub-plots, identify all road segments at the corners of the plot boundaries. Use an angle threshold filter (typically set between 60° and 160°) to calculate the angle between the direction vectors of the two road segments at the corner. If the angle falls within this range, the corner is considered close to a right angle and belongs to a normal road turn. Since the setback calculation has already been covered by the two adjacent road segments, the road segment at that corner is filtered out from the candidate set for this calculation to avoid duplicate calculations and result interference.

[0047] S3, with each building as the center, emits query rays in the direction of the normal vector of each line segment of the corresponding sub-plot boundary, and establishes the mapping relationship between buildings and roads through spatial intersection analysis.

[0048] Understandably, for each building base polygon, a query ray is cast from its geometric center or a specific feature point as the origin, radiating towards the normal direction of the sub-plot boundary obtained in step S2. Using the ray intersection function of the spatial database, the nearest intersection point between these rays and the road redline layer is calculated. Roads with intersection points have spatial relationships with buildings, and each valid intersection point establishes a spatial association from the building to a certain road, recording the setback direction and the target road.

[0049] This step ensures the validity and uniqueness of the setback direction. All ray paths are analyzed. If a ray originating from building A is occluded by the base polygon of building B before reaching the target road (i.e., the ray intersects with the polygon of building B), then building B is determined to be closer to the road in that direction, and the association of this ray from building A is invalid. All such occluded rays are removed, ensuring that each building retains only the nearest setback direction that allows unobstructed direct access to the road.

[0050] Furthermore, since road boundary data may have logical segments or be discontinuous, the same road segment may be simultaneously associated with rays from multiple buildings. To address this, spatial and attribute deduplication of the "building-road" mapping relationship is performed. Duplicate references to the same road segment are merged based on road ID and spatial location (e.g., using spatial clustering algorithms). This ensures that each road segment is processed only once in subsequent calculations, improving computational efficiency and generating a unique theoretical minimum setback reference value.

[0051] S4: Obtain the height of each building. Based on the height of each building and the associated road boundary attribute information, obtain the theoretical minimum setback distance of each building relative to the road by querying.

[0052] Understandably, S3 identifies multiple roads associated with each building space, and through predefined unique building identifiers (such as UUIDs), it performs a correlation query with the BIM review results database to accurately obtain the "building height" attribute value of each building.

[0053] Subsequently, based on the input road boundary attribute information (such as road grade and design width) and building height, the built-in "Setback Rule Configuration Table" is queried to automatically match and output the theoretical minimum setback distance value that the building must meet on the side adjacent to this road segment. This process transforms textual specifications into calculable structured data. Thus, the theoretical minimum setback distance between each building and its associated roads is found. The theoretical minimum setback distances corresponding to different types of roads and different building heights can be found in Table 1.

[0054] Table 1 Theoretical minimum setback distance between buildings and roads

[0055]

[0056] S5. For each building base outline and road red line, two point sets are generated using adaptive high-density interpolation. The two point sets are then used to accelerate nearest neighbor search through spatial indexing. The shortest Euclidean distance between all point pairs in the two point sets is calculated, and the shortest distance point pair is located.

[0057] Understandably, for each building, the building and each associated road are extracted, and the precise shortest distance between the building and each road is calculated using a high-density interpolation method.

[0058] To address the challenge of calculating the shortest distance between complex building contours (such as curved or recessed balconies) and irregular road boundaries (such as curved or discontinuous sections), this step employs an adaptive high-density interpolation algorithm. This method overcomes the limitation of traditional algorithms that only check vertex distances. By discretizing continuous line elements into a high-density point cloud and combining it with spatial indexing technology, it achieves millimeter-level measurement accuracy, effectively solving the shortest distance localization problem for complex contours such as recessed lobbies. The detailed implementation process is as follows:

[0059] (1) The implementation of the high-density interpolation algorithm includes: First, parameter configuration and line feature preprocessing are performed. Parameter initialization: Define the interpolation density parameter. (Default 0.5 meters, configurable by the system administrator). Obtain the building base outline. and road red line The geometric data is used to verify the topological correctness of the input line features and to fix issues such as self-intersection.

[0060] For the building base outline and road red line Line feature classification processing: The input line features are classified into straight line segments and curve segments. Straight line segments are line segments defined by two endpoints with curvature κ=0. Curve segments are line segments defined by parametric equations (such as circular arcs and Bézier curves), and their curvature needs to be calculated.

[0061] Special case handling: For zero-length line segments, skip them automatically; for discontinuous road boundary data, perform spatial connection processing first; for small holes (<0.1 meters) in building outlines, ignore them as noise.

[0062] Two point sets are generated separately for the building base outline and the associated road boundary line using an adaptive high-density interpolation method. The specific steps include:

[0063] Define the interpolation density parameter And to obtain the building base outline and associated road boundary lines;

[0064] Obtain the straight and curved segments in the building base outline and the road boundary line respectively;

[0065] For line segments, an adaptive high-density interpolation method is used to generate point sets, including:

[0066] Calculate the length L of the straight line segment and determine the number of insertion points. , To round down;

[0067] Based on the number of insertion points N, insertion points are placed at equal intervals on the line segment. arrive Calculate the interpolation point position ;

[0068] Generate a point set based on the two endpoints and the insertion point of the line segment;

[0069] For curve segments, an adaptive high-density interpolation method is used to generate point sets, including:

[0070] Calculate curve segment curvature , where t is a parameter variable;

[0071] Dynamically adjust the interpolation density and calculate the interpolation step size. , where α is the curvature influence coefficient;

[0072] Starting from t=0, insert points according to the dynamic step size until t≥1;

[0073] A point set is generated based on the two endpoints and the interpolation point of the curve segment.

[0074] An adaptive high-density interpolation algorithm is used to interpolate the building outline and the road boundary line respectively, generating a set of building outline points and a set of road boundary line points.

[0075] Next, check whether the generated point density meets the accuracy requirements and verify whether the generated points completely cover the original line features. Automatically increase the point density for key areas (such as concave structures) to ensure that the shortest distance points are not missed.

[0076] In some embodiments of the present invention, spatial indexing is used to accelerate nearest neighbor search for two point sets, the shortest Euclidean distance between all pairs of points in the two point sets is calculated, and the shortest distance point pairs are located, including:

[0077] Based on the building base outline point set and the road red line point set, an R-tree is constructed respectively. The index item of the R-tree includes the point coordinates (x, y), the ID of the original line segment where the point is located, and the parameter position s of the original line segment.

[0078] Using the R-trees corresponding to the two point sets, search for the shortest distance from each building outline point to the road boundary line;

[0079] Based on the shortest distance from all building outline points to the road boundary line, determine the shortest distance from the building to the road boundary line, and record the nearest point on the building outline and the nearest point on the road boundary line corresponding to the shortest distance point, forming a shortest distance point pair.

[0080] Specifically, after generating the building outline point set and the road boundary line point set, R-Tree spatial indexes are constructed separately. First, the index structure is designed: R-Tree is used as the primary spatial index structure, suitable for two-dimensional spatial data. R-Trees are constructed for the building outline point cloud P_b and the road boundary line point cloud P_r respectively. The index entries include point coordinates (x, y), the original line segment ID, and the parameter position s.

[0081] An R*-Tree variant is used to optimize node overlap and area. The node capacity is set to M=100 (which can be adjusted according to memory). Point data is inserted in the order of the Hilbert curve to optimize spatial locality.

[0082] After constructing R-trees for the building outline point set and the road boundary point set, an efficient nearest neighbor search algorithm is used to search and determine the shortest distance between the building and each road.

[0083] The first step is to prioritize searching the area close to the road in the building outline point cloud. Using the convex hull property, the shortest distance between the building outline convex hull and the road boundary line is calculated first. For areas where the distance is greater than the theoretical minimum setback distance, the search is terminated in advance.

[0084] For the initially found shortest distance point pairs, a secondary verification is performed by expanding the neighborhood range to eliminate local extrema interference. During the search process, multithreading can be used to process multiple buildings simultaneously, leveraging parallel computing to accelerate distance calculation. Furthermore, to further accelerate distance calculation during the search process, distance lower bound pruning is implemented: if the minimum possible distance from the current point to the road point cloud is already greater than the current shortest distance, the search for that point is terminated early.

[0085] Precise location of the search distance includes global minimum location. Specifically, using the search algorithm described above, the shortest distance from each point in the building outline point cloud to the road point cloud is recorded. The shortest distances of all points are compared, and the global minimum is determined as the precise shortest distance from the building to the road, forming a pair of points with the shortest distance between the building and the road.

[0086] Then, record the coordinates of the building outline point corresponding to the shortest distance point, its corresponding line segment ID, and the parameter position t_b of the corresponding line segment (which can be understood as the number index of the corresponding line segment among all line segments), as well as the coordinates of the road red line point corresponding to the shortest distance point, its corresponding line segment ID, and the parameter position t_r of the corresponding line segment. Using the line segment ID and parameter position, the specific location of the shortest distance on the building outline and the road is accurately identified.

[0087] Finally, for the determined shortest distance point pair, neighborhood verification is performed (checking points within a range of ±d / 2) to confirm that the point pair indeed represents the global shortest distance, avoiding interference from local extrema. Traceable positioning information is generated to support subsequent 3D mapping and visualization.

[0088] By searching and identifying the shortest distance pair between each building and each associated road, the actual road width can be determined. Specifically, starting from the building center and pointing towards the nearest road point, an extended ray is emitted. This ray traverses the current road until it intersects the opposite road boundary line, thus defining the complete road extent. Connecting the boundary points on both sides of the same road establishes a complete road map. Based on this, the road's perpendicular bisector can be calculated, and the actual road width can be verified or calculated using map scale or attribute data.

[0089] S6: Determine if the shortest distance between each building and the road is within the warning distance range. If yes, end the process; otherwise, generate a review report and proceed to S7.

[0090] Understandably, after finding the shortest distance from the building to each road, it checks whether the shortest distance is within the warning distance range (theoretical minimum setback distance). If it is, no action is taken. If the shortest distance is within the warning distance range, the road width, minimum road setback distance, and road red line configuration are returned to the front end. The front end generates a visualization result, generates a compliance review report, and executes subsequent S7 steps.

[0091] S7. Based on the nearest point from each building to the road and the corresponding edge length, determine the linear range of the road red line along the road direction. Based on the linear range of the road red line, stretch it towards the building direction to generate a local three-dimensional road buffer model with the same height as the building and the same depth as the theoretical minimum setback distance.

[0092] In some embodiments of the present invention, step S7, using the closest point from each building to the road and its corresponding edge length as a reference, determines the linear range of the road red line along the road direction. Using the linear range of the road red line as a reference, it stretches towards the building direction to generate a local three-dimensional road buffer model with the same height as the building and the same depth as the theoretical minimum setback distance, including:

[0093] Find the closest point of the building outline in the shortest distance point pair and the building edge it is located on;

[0094] Calculate the length of the building's edge line, and extend half the length of the building's edge line to both sides from the nearest point of the building's outline to form the building's side influence range;

[0095] By using normal vector projection technology, the influence range of the building side is mapped to the road red line space to determine the linear range of the road red line;

[0096] Based on the defined linear range of the road red line, parametric stretching is performed in the direction of the building. The stretching depth is the theoretical minimum setback distance, and the stretching height is the building height in the building BIM model, generating a local three-dimensional buffer model of the road.

[0097] Specifically, this step, as a key technical link in realizing 2D and 3D collaborative analysis, innovatively solves the problem of serious waste of computing resources in traditional 3D review. By accurately identifying potential violation areas (i.e., buildings and roads whose shortest distance to the road is within the warning distance range), a highly customized 3D buffer model is generated only for this area, rather than modeling the entire road, which greatly improves the efficiency and accuracy of subsequent collision detection.

[0098] To determine the baseline range, a precise spatial relationship is first established based on the nearest point of the building outline and its corresponding building edge. Specifically, the actual length of the building edge containing the nearest point is calculated, and half the edge length (maximum 5 meters) is extended to both sides of the nearest point to form the building's side influence range. Subsequently, this range is precisely mapped onto the road red line space using normal vector projection technology to determine the corresponding linear range on the road red line. For curved roads, the mapping parameters are dynamically adjusted according to the road curvature to ensure accurate definition of the buffer range even in curved areas. At road intersections, based on the angle threshold filtering results of S2, it is intelligently determined whether the buffer range needs to be extended to adjacent roads to avoid the risk of missed detection due to the complexity of road topology.

[0099] During parametric modeling, the system employs a highly accurate geometric generation strategy. Based on the defined linear range of the road boundary, parametric stretching is performed towards the building direction. The stretching depth strictly matches the calculated theoretical minimum setback distance, ensuring the model depth is completely consistent with planning requirements. The stretching height precisely corresponds to the "building height" attribute value in the building BIM model, guaranteeing vertical detection accuracy. For road terrain variations, the bottom surface of the buffer model is dynamically tilted based on Digital Elevation Model (DEM) data, ensuring a perfect fit between the model and the actual ground shape. For high-rise buildings, wind load effects are also considered, with a tapered convergence design in the top 5% height area of ​​the model to more realistically reflect the spatial relationship between the building and the road. In terms of geometric representation, NURBS surfaces are used to accurately describe the buffer boundaries of road curve segments, ensuring model geometric accuracy. Simultaneously, intelligent mesh simplification is implemented in non-critical areas to reduce data volume.

[0100] During the model optimization and verification phase, topology checks are performed to ensure that the generated 3D buffer volume is a manifold geometry without self-intersections or holes. Coordinate system transformation is used to align the model to a local coordinate system consistent with the building BIM model. Bounding box technology is employed to pre-calculate the spatial extent of the model. The final generated local 3D road buffer model only includes the narrowest spaces most likely to violate regulations, with a data volume only 5-15% of the full road segment model, yet retaining the geometric details of key areas, laying the foundation for 3D collision detection.

[0101] The model is stretched according to the linear range of the road red line to generate a local three-dimensional road buffer model.

[0102] S8 uses a bounding box-based fast collision detection algorithm to detect whether building components intrude into the road buffer zone by combining the generated local road 3D buffer model with the building BIM model.

[0103] In some embodiments of the present invention, step S8 involves using a bounding box-based fast collision detection algorithm to detect whether building components intrude into the road buffer zone, combining the generated local road 3D buffer model with the building BIM model.

[0104] The local road 3D buffer model and the building BIM model are subjected to spatial intersection analysis using a collision detection algorithm, wherein the collision algorithm includes a coarse screening stage and a fine measurement stage.

[0105] The coarse screening stage includes:

[0106] The space is divided using a dynamic octree. Axial bounding boxes are calculated for each component in the building BIM model and the 3D buffer model of local roads. Through bounding box overlap testing, object pairs that do not overlap in space are excluded. The remaining components in the building BIM model are then subjected to fine-tuning as potential collision objects.

[0107] The precision measurement stage includes:

[0108] Geometric intersection is determined for potential collision objects. The depth and volume of each building component intruding into the road are calculated. The collision component ID, position coordinates and intrusion depth are recorded. The three-dimensional intrusion depth is converted into a two-dimensional plane setback distance insufficient value through a geometric projection algorithm.

[0109] Understandably, this step involves performing a spatial intersection analysis between the local road 3D buffer model generated by S7 and the building BIM model. A two-stage collision detection algorithm is employed, detecting only potentially illegal areas to improve computational efficiency.

[0110] In the initial screening stage, the system uses a dynamic octree to partition the space and calculates axial bounding boxes (AABBs) for each component in the building BIM model and the road buffer model. By performing bounding box overlap tests, clearly non-overlapping object pairs are quickly eliminated. The remaining components are then re-detected as potential collision objects, and possible collision areas are predicted using the results of the previous detection. Parallel computing technology improves processing efficiency.

[0111] In the fine measurement phase, the system performs precise geometric intersection judgment on potential collision object pairs that pass the coarse screening. After quality inspection of the component triangular mesh, a precise intersection test is performed to calculate the intrusion depth and volume. A multi-resolution strategy is adopted to first determine the approximate collision location and then perform fine calculations to ensure measurement accuracy.

[0112] Key information such as the collision component ID, location coordinates, and intrusion depth is recorded. The 3D intrusion depth is then converted into a 2D planar insufficient clearance distance value using a geometric projection algorithm, and cross-validated with the 2D calculation results from step S6. The final collision results are used to identify illegal intrusions by cantilevered components, canopies, and other three-dimensional elements, providing a basis for subsequent review reports.

[0113] S9 calculates and outputs the collision detection results, identifies the component ID, location coordinates, and intrusion depth value that intrude into the road buffer zone in the building BIM model, and outputs a 3D schematic diagram of the collision area.

[0114] Understandably, S8 generates collision detection results after performing spatial collision detection on the local road 3D buffer model and the building BIM model. This step analyzes the collision detection results to accurately identify and quantify the collision situations. It identifies the specific components (such as cantilevered balconies, canopies, etc.) that intrude into the road buffer zone in the BIM model and their corresponding triangular facet sets, establishing a precise mapping relationship between the components and the intruded area.

[0115] The automatically generated structured conflict result data includes: the unique identifier of the colliding component (corresponding to the IFC GUID in the BIM model), the three-dimensional spatial coordinate range of the collision area (represented by the minimum bounding box), the maximum depth of the intruding component into the buffer zone, and key parameters such as the intrusion volume.

[0116] An automatic perspective optimization algorithm is used to calculate the optimal viewing angle based on the spatial characteristics of the collision area, and generate a 3D visualization image of the collision area to ensure that key violation features are clearly visible.

[0117] All results data are organized in accordance with planning review specifications, providing accurate basis for the generation of subsequent rectification suggestions, and supporting cross-validation with two-dimensional review results to ensure consistency between two-dimensional and three-dimensional analysis results.

[0118] The review results (including road width, minimum setback distance, theoretical distance, and collision information) are transmitted to a front-end web visualization engine (such as Cesium or Three.js). The front-end interface enables 2D and 3D linkage: illegal buildings and related roads are highlighted on the 2D map; in the 3D scene, the building BIM model is rendered, and the local road buffer model and intruding building components that collide are highlighted with semi-transparent red, providing an intuitive and immersive review experience, and supporting clicking to query detailed information.

[0119] This includes S10 after S10, which generates a comprehensive report based on the collision detection results.

[0120] Understandably, this step, as the technical endpoint and value output link of the entire 2D / 3D collaborative review process, innovatively constructs a semantic conversion model from geometric data to standardized expressions, achieving seamless integration between review results and planning management requirements. Through multi-dimensional data fusion and intelligent mapping algorithms, complex 3D geometric collision results are transformed into a structured and quantitative review report that meets the requirements of the "Guidelines for Review of Construction Engineering Design Schemes," completely solving the problem of the disconnect between review results and standard clauses in existing technologies.

[0121] During the data integration phase, the consistency verification of the two-dimensional and three-dimensional results is first performed: the two-dimensional shortest distance calculated by S5 and the three-dimensional intrusion depth determined by S9 are cross-validated using a geometric projection algorithm. The calculation formula is: two-dimensional planar distance = three-dimensional intrusion depth × cosθ (where θ is the angle between the component's normal and the vertical direction). This distance is the insufficient setback distance value. Based on the insufficient setback distance value, the building continues to be set back by this distance value in the direction away from the road. Subsequently, a complete data association chain is constructed, accurately mapping geometric data such as building component ID, spatial location, and intrusion depth with management data such as building attributes, road type, and regulatory requirements, forming a complete review evidence chain.

[0122] In the semantic transformation stage, a rule-based reasoning engine is used to convert geometric data into normative expressions. This engine comprises three key components: first, a setback distance converter, which accurately maps the three-dimensional intrusion depth to insufficient planar setback distances; second, a normative clause matcher, which automatically matches corresponding legal provisions from a built-in normative knowledge base based on multi-dimensional parameters such as road type, building height, and regional characteristics; and finally, a rectification instruction generator, which calculates the precise rectification distance based on the intrusion depth and generates a quantitative expression that meets the requirements of administrative documents. For example, the system not only identifies "the F5 cantilever slab at the southeast corner intrudes into the road red line buffer zone," but also accurately calculates the rectification distance of "needing to be moved 0.52 meters inward." This value is derived from the difference between the theoretical minimum setback distance and the actual setback distance after geometric correction, ensuring the scientific validity and operability of the rectification recommendations.

[0123] In terms of report structure design, the system adopts a modular report generation framework, outputting standardized documents that meet the requirements of planning administration. The report consists of five core parts:

[0124] (1) The review summary section provides an overview of the overall compliance conclusions and key indicators;

[0125] (2) The detailed data section lists the actual measured setback distance, theoretical required distance and difference for each review point in tabular form, and indicates the data source (two-dimensional calculation or three-dimensional verification).

[0126] (3) The violation list section describes each violation in detail, including the component IFC GUID, spatial coordinates, intrusion depth, and violation type (such as three-dimensional intrusion or planar insufficiency).

[0127] (4) The section on violations not only lists the relevant legal provisions number, but also provides the original text of the provisions and specific application instructions;

[0128] (5) The attached diagram automatically generates a two-dimensional planar schematic diagram, a three-dimensional collision view and a detailed data table of the violation location, supporting multi-angle viewing and interactive query;

[0129] (6) It realizes the two-way association between the report content and the BIM model. When the reviewer clicks on any violation item in the report, the corresponding area will be highlighted in the three-dimensional model, which greatly improves the review efficiency.

[0130] This invention provides a CIM-based method for reviewing building setback distances, which has the following advantages:

[0131] (1) By decomposing the boundaries of the plots and finding the unit normal vector and the high-density interpolation method, the problem that the traditional two-dimensional method cannot accurately locate the shortest distance point of the midpoint of the edge line is solved, which meets the accuracy requirements of the "Guidelines for Review of Construction Engineering Design Schemes".

[0132] (2) By using the two-dimensional and three-dimensional collaborative mechanism, the waste of computing resources caused by full-model three-dimensional detection is avoided, and the review efficiency is improved.

[0133] (3) Through the semantic transformation model, the review results and planning management requirements are directly connected, avoiding manual secondary parsing. It should be noted that in the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0134] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0135] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0136] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0137] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0138] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0139] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for reviewing building setback distances based on CIM, characterized in that, include: S1, acquire building foundation data and road boundary data; S2, decompose the boundary of each plot to which each building belongs, obtain multiple line segments of the plot boundary, and calculate the unit direction vector and normal vector of each line segment; S3, taking each building as the center, emits query rays in the direction of the normal vector of each line segment of the corresponding sub-plot boundary, and establishes the relationship between buildings and roads through spatial intersection analysis; S4. Obtain the height of each building. Based on the height of each building and the associated road boundary attribute information, obtain the theoretical minimum setback distance of each building relative to the road by querying. S5. For each building base outline and road red line, two point sets are generated using adaptive high-density interpolation. The two point sets are then used to accelerate the nearest neighbor search through spatial indexing. The shortest Euclidean distance between all point pairs in the two point sets is calculated, and the shortest distance point pairs are located. S6: Determine if the shortest distance between each building and the road is within the warning distance range. If yes, end the process; otherwise, generate a review report and proceed to S9. S7. Based on the nearest point from each building to the road and the corresponding edge length, determine the linear range of the road red line along the road direction. Based on the linear range of the road red line, stretch it towards the building direction to generate a local three-dimensional road buffer model with the same height as the building and the same depth as the theoretical minimum setback distance. S8. The generated local road 3D buffer model and building BIM model are compared using a bounding box-based fast collision detection algorithm to detect whether building components intrude into the road buffer zone. S9 calculates and outputs the collision detection results, identifies the component ID, location coordinates, and intrusion depth value that intrude into the road buffer zone in the building BIM model, and outputs a 3D schematic diagram of the collision area.

2. The method for reviewing building setback distances according to claim 1, characterized in that, S1, acquiring building foundation data and road boundary data, includes: The projected plane outline of each building foundation and all road boundary line data are obtained from the CIM platform. The projected plane outline is a polygon and the road boundary line data is linear feature data.

3. The method for reviewing building setback distances according to claim 1, characterized in that, S2 involves decomposing the boundary of each building's sub-plot to obtain multiple line segments of the sub-plot boundary, calculating the unit direction vector and normal vector of each line segment, and prior to this, establishing the relationship between each building's foundation and the sub-plot boundary to verify the geometric validity of each building's foundation and the sub-plot boundary. Traverse all building base polygons and verify whether each building base polygon is contained within the boundary of its sub-plot by judging spatial relationships. If so, execute S3; otherwise, rectify buildings that exceed the sub-plot boundary.

4. The method for reviewing building setback distances according to claim 1, characterized in that, S2 involves decomposing the boundary of each building's sub-plot to obtain multiple line segments of the sub-plot boundary, and calculating the unit direction vector and normal vector of each line segment, including: Obtain polygon data of the boundaries of the land parcels; Extracting vertex coordinate sequences from polygon data of plot boundaries ,in ; Construct a set of line segment units in vertex order: , where each line segment From the starting point and the end point definition, From 0 to ,when At 1 o'clock, the destination is ; Assign a unique identifier SID = [plot ID]_[segment number] to each line segment; For line segments Its starting coordinates are The endpoint coordinates are Calculate line segments Direction vector Calculate line segments length ; Calculate line segments Unit normal vector: ; Verify line segment Is the unit normal vector correct? From line segment midpoint The direction of the verification normal vector is taken as the distance. checkpoint Is it located outside the building's base polygon? If so, line segment The unit normal vector is correct; if not, the line segment... The unit normal vector is incorrect; the line segment should be taken. The unit normal vector is .

5. The method for reviewing building setback distances according to claim 1, characterized in that, The preceding S3 includes: Identify the road segments at the corners of each sub-plot boundary, calculate the angle between the direction vectors of the two road segments at the corner, and if the angle between the direction vectors is within the angle threshold range, the two road segments belong to the normal road bends, filter out the two road segments, and do not participate in S3. S3, taking each building as the center, emits query rays in the direction of the normal vector of each line segment of the corresponding sub-plot boundary, and establishes the relationship between buildings and roads through spatial intersection analysis, including: Centered on each building, a query ray is emitted in the direction of the normal vector of each line segment of the corresponding sub-plot boundary. Using the ray intersection function of the spatial database, the nearest intersection point between each query ray and the road red line is calculated. Based on the intersection point, the spatial relationship between each building and the road red line is established, and the setback direction and target road of each building are recorded.

6. The method for reviewing building setback distances according to claim 1, characterized in that, In step S5, for each building base outline and road boundary line, two point sets are generated using an adaptive high-density interpolation method, including: Define the interpolation density parameter And to obtain the building base outline and associated road boundary lines; Obtain the straight and curved segments in the building base outline and the road boundary line respectively; For line segments, an adaptive high-density interpolation method is used to generate point sets, including: Calculate the length L of the straight line segment and determine the number of insertion points. , To round down; Based on the number of insertion points N, insertion points are placed at equal intervals on the line segment. arrive Calculate the interpolation point position ; Generate a point set based on the two endpoints and the insertion point of the line segment; For curve segments, an adaptive high-density interpolation method is used to generate point sets, including: Calculate curve segment curvature , where t is a parameter variable; Dynamically adjust the interpolation density and calculate the interpolation step size. , where α is the curvature influence coefficient; Starting from t=0, insert points according to the dynamic step size until t≥1; A point set is generated based on the two endpoints and the interpolation point of the curve segment.

7. The method for reviewing building setback distances according to claim 1, characterized in that, In step S5, spatial indexing is used to accelerate nearest neighbor search for two point sets, the shortest Euclidean distance between all point pairs in the two point sets is calculated, and the shortest distance point pairs are located, including: Based on the building base outline point set and the road red line point set, an R-tree is constructed respectively. The index item of the R-tree includes the point coordinates (x, y), the ID of the original line segment where the point is located, and the parameter position s of the original line segment. Using the R-trees corresponding to the two point sets, search for the shortest distance from each building outline point to the road boundary line; Based on the shortest distance from all building outline points to the road boundary line, determine the shortest distance from the building to the road boundary line, and record the nearest point on the building outline and the nearest point on the road boundary line corresponding to the shortest distance, forming a shortest distance point pair.

8. The method for reviewing building setback distances according to claim 1, characterized in that, Before S6, the following is also included: Based on the shortest distance point pair, starting from the building center, extend the ray towards the nearest point on the road red line in the shortest distance point pair, cross the corresponding road, and form two intersections with the road. The distance between the two intersections is the actual width of the road.

9. The method for reviewing building setback distances according to claim 1, characterized in that, S7, using the closest point from each building to the road and its corresponding edge length as a reference, determines the linear range of the road red line along the road direction. Based on this linear range, it stretches the model towards the building direction to generate a local three-dimensional road buffer model with the same height as the building and the same depth as the theoretical minimum setback distance, including: Find the closest point of the building outline in the shortest distance point pair and the building edge it is located on; Calculate the length of the building's edge line, and extend half the length of the building's edge line to both sides from the nearest point of the building's outline to form the building's side influence range; By using normal vector projection technology, the influence range of the building side is mapped to the road red line space to determine the linear range of the road red line; Based on the defined linear range of the road red line, parametric stretching is performed in the direction of the building. The stretching depth is the theoretical minimum setback distance, and the stretching height is the building height in the building BIM model, generating a local three-dimensional buffer model of the road.

10. The method for reviewing building setback distances according to claim 1, characterized in that, In step S8, the generated local road 3D buffer model and building BIM model are compared using a bounding box-based fast collision detection algorithm to detect whether building components intrude into the road buffer zone, including: The local road 3D buffer model and the building BIM model are subjected to spatial intersection analysis using a collision detection algorithm, wherein the collision algorithm includes a coarse screening stage and a fine measurement stage. The coarse screening stage includes: The space is divided using a dynamic octree. Axial bounding boxes are calculated for each component in the building BIM model and the local road 3D buffer model. Through bounding box overlap test, object pairs that do not overlap in space are excluded. The remaining components in the building BIM model are then measured as potential collision objects. The precision measurement stage includes: Geometric intersection is determined for potential collision objects. The depth and volume of each building component intruding into the road are calculated. The collision component ID, position coordinates and intrusion depth are recorded. The three-dimensional intrusion depth is converted into a two-dimensional plane setback distance insufficient value through a geometric projection algorithm.

Images