A three-dimensional cadastral survey result checking method

Abstract

The application discloses a three-dimensional cadastral survey result checking method, and relates to the technical field of data processing. The method comprises the following steps: inputting a preset three-dimensional building model and a preset three-dimensional land parcel model into a preset three-dimensional GIS topological analysis engine to perform spatial relationship judgment, determining a land parcel falling detection result based on a spatial relationship judgment result and a right ownership detection result; determining an out-of-bound distance between a building and a land parcel and an overlapping volume between the building and a neighboring building based on a preset spatial conflict quantification model, and determining a building falling detection result according to the out-of-bound distance, the overlapping volume and a planning detection result; and obtaining a target checking result according to the land parcel falling detection result, the building falling detection result and a building number falling detection result, so that the accuracy of cadastral survey result detection can be improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method for checking the results of a three-dimensional cadastral survey. Background Technology

[0002] Current technologies primarily rely on 2D GIS for related inspections. During the three-stage inspection (land parcel, building, and land number), for spatial conflicts in the vertical direction and those spanning elevations, only planar information can be displayed graphically, with relevant information indicated through elevation markings and textual descriptions. However, this approach cannot visually represent the vertical ownership intersections when two land parcels do not overlap in plan but overlap vertically, especially in land parcel conflicts. Furthermore, in terms of building conflicts, it cannot intuitively display the spatial form of the vertical boundary-crossing portion of a building, making it easy to overlook boundary-crossing issues.

[0003] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main purpose of this application is to provide a method for inspecting three-dimensional cadastral survey results, aiming to solve the technical problem of how to improve the accuracy of cadastral survey result detection.

[0005] To address the aforementioned problems, this application provides a method for inspecting three-dimensional cadastral survey results, which includes:

[0006] Input the preset 3D building model and the preset 3D land parcel model into the preset 3D GIS topology analysis engine to determine the spatial relationship. Based on the spatial relationship determination results and the ownership detection results, determine the land parcel detection results.

[0007] The boundary distance between the building and the land parcel is determined based on the preset spatial conflict quantification model, as well as the overlapping volume between the building and the adjacent building. The building drop detection result is determined based on the boundary distance, the overlapping volume, and the planning detection result.

[0008] Based on the results of the land parcel detection, the building detection, and the number detection, the target inspection results are obtained.

[0009] In one embodiment, the step of inputting a preset 3D building model and a preset 3D land parcel model into a preset 3D GIS topology analysis engine for spatial relationship determination, and determining the land parcel detection result based on the spatial relationship determination result and the ownership detection result includes:

[0010] Based on the 3D GIS topology analysis engine, the spatial relationship is determined. If the spatial relationship is an inclusion relationship, the spatial location is determined to be compliant.

[0011] The building ownership information and land parcel ownership information are compared with the ownership registration information to determine the ownership test results;

[0012] If the spatial location is compliant and the ownership test results are consistent, then the land registration test result is determined to be qualified.

[0013] In one embodiment, the step of determining the spatial relationship based on the 3D GIS topology analysis engine, and determining the compliance of the spatial location if the spatial relationship is an inclusion relationship, includes:

[0014] Obtain the minimum bounding box of the building and the minimum bounding box of the land parcel, and divide the minimum bounding box of the building and the minimum bounding box of the land parcel into multiple subspaces based on the preset side length;

[0015] Obtain the building vertices in each of the subspaces. If the number of intersections between the ray corresponding to the positive Z-axis direction of the building vertex and the land boundary is odd, then the building vertex is determined to be inside the land, and an internal vertex list is generated.

[0016] If the number of intersection points is even, then the building vertex is determined to be outside the land parcel, and a list of superboundary vertices is generated;

[0017] If all the building vertices are internal vertices and there are no superboundary vertices, then the spatial relationship between the building and the land parcel is determined to be an inclusion relationship, and the spatial relationship between the building and the land parcel is compliant.

[0018] In one embodiment, the step of determining the transboundary distance between a building and a land parcel, and the overlapping volume between the building and adjacent buildings, based on a preset spatial conflict quantification model, and determining the building drop detection result according to the transboundary distance, the overlapping volume, and the planning detection result includes:

[0019] Traverse the building vertices and land parcel vertices, obtain and determine the hyperboundary distance based on the building Z-value range and the land parcel Z-value range;

[0020] Based on the building number of the building, obtain the adjacent buildings that intersect with the building, and obtain the overlapping volume of the building and the adjacent buildings;

[0021] If the superboundary distance is less than or equal to a preset superboundary distance threshold, and the overlapping volume is less than or equal to a preset overlapping volume threshold, then it is determined that there is no spatial conflict.

[0022] When the number of floors deviation is zero and the building area deviation rate is less than or equal to the preset building area deviation rate threshold, the planning indicators are deemed compliant.

[0023] If the planning indicators are confirmed to be compliant and there are no spatial conflicts, then the building inspection result is determined to be qualified.

[0024] In one embodiment, before the step of determining that the planning indicators are compliant when the floor number deviation is zero and the building area deviation rate is less than or equal to a preset building area deviation rate threshold, the three-dimensional cadastral survey result inspection method further includes:

[0025] Based on the building number, obtain the associated permit information, compare the approved number of floors in the permit information with the number of floors in the three-dimensional building model, and obtain the floor deviation value;

[0026] The approved building area in the license information is compared with the building area in the three-dimensional building model to obtain the building area deviation rate.

[0027] In one embodiment, before the step of inputting a preset 3D building model and a preset 3D land parcel model into a preset 3D GIS topology analysis engine to determine the spatial conflict check results between the land parcel and the building, the 3D cadastral survey result checking method further includes:

[0028] Initial point cloud data is acquired using a 3D laser device. The point cloud data is then denoised using a preset point cloud density clustering algorithm to remove vegetation points and noise points, thereby obtaining target point cloud data.

[0029] The initial image is acquired using a photographic device, and the initial image is stitched together according to a preset feature point matching algorithm to obtain the target image.

[0030] Based on the target point cloud data and the target image, the three-dimensional land parcel model and the three-dimensional building model are constructed.

[0031] In one embodiment, the step of constructing the 3D land parcel model and the 3D building model based on the target point cloud data and the target image includes:

[0032] Based on a preset boundary extraction algorithm, the set of land parcel boundary points is extracted from the target point cloud data, and the set of land parcel boundary points is matched and fused with the coordinates of the boundary points to generate the land parcel boundary.

[0033] Based on the vertical height of the boundary point along the Z-axis, the land parcel boundary is stretched along the Z-axis to generate the three-dimensional land parcel model;

[0034] The target point cloud data is sliced ​​and feature-recognized to obtain the building outline and the boundary of the partition wall;

[0035] The texture information of the target image is mapped onto the building outline to obtain the three-dimensional building model.

[0036] In one embodiment, the step of performing point cloud slicing and feature recognition on the target point cloud data to obtain the building outline and the boundary of the partition wall includes:

[0037] Along the Z-axis, the building point cloud is sliced ​​according to a preset thickness to obtain interval slices, and the median Z value of the interval slice is determined as the slice feature value.

[0038] According to the slice feature values ​​in ascending order, the slice feature value difference between adjacent interval slices is obtained. If the slice feature value difference is greater than the preset floor height threshold, the lower limit of the Z value interval of the previous interval slice is determined as the floor boundary line.

[0039] In one embodiment, the method for checking the results of a three-dimensional cadastral survey further includes:

[0040] Upon receiving real-time data, a data identifier is generated based on the parcel code and update timestamp of the real-time data.

[0041] Based on the data identifier, the corresponding hash value is determined, and the hash value is matched in the database to obtain the new data;

[0042] Based on the model difference algorithm and the newly added data, the three-dimensional building model and the three-dimensional land parcel model are updated.

[0043] In addition, to achieve the above objectives, this application also proposes a three-dimensional cadastral survey result inspection method and apparatus, the apparatus comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the three-dimensional cadastral survey result inspection method as described above.

[0044] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the three-dimensional cadastral survey result inspection method described above.

[0045] This application provides a method for checking three-dimensional cadastral survey results. It uses a three-dimensional GIS topology analysis engine to determine the spatial relationship between three-dimensional building models and three-dimensional land parcel models, and combines this with ownership detection to accurately determine the land parcel registration results, effectively ensuring the accuracy of matching building and land parcel ownership. By using a spatial conflict quantification model to calculate boundary distances and overlapping volumes, and integrating planning detection results to determine building registration results, it can accurately identify spatial conflicts between buildings and land parcels, and between adjacent buildings. Finally, by integrating the land parcel registration, building registration, and number registration results, the target inspection result is obtained, providing accurate and reliable evidence for fields such as architectural planning and ownership management. Attached Figure Description

[0046] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A schematic diagram of the first type of process for examining the three-dimensional cadastral survey results of this application;

[0049] Figure 2 A second flowchart illustrating the method for examining the three-dimensional cadastral survey results in this application;

[0050] Figure 3 A third flowchart illustrating the method for examining the three-dimensional cadastral survey results in this application;

[0051] Figure 4 This is a schematic diagram of the hardware operating environment involved in the three-dimensional cadastral survey results inspection method in this application embodiment.

[0052] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0053] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0054] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0055] To achieve the above objectives, this application proposes a method for inspecting three-dimensional cadastral survey results. The method includes: inputting a preset three-dimensional building model and a preset three-dimensional land parcel model into a preset three-dimensional GIS topology analysis engine for spatial relationship judgment; determining the land parcel detection result based on the spatial relationship judgment result and ownership detection result; determining the boundary distance between the building and the land parcel, and the overlapping volume between the building and adjacent buildings, based on a preset spatial conflict quantification model; determining the building detection result based on the boundary distance, the overlapping volume, and the planning detection result; and obtaining the target inspection result based on the land parcel detection result, the building detection result, and the land number detection result.

[0056] Current technologies primarily rely on 2D GIS for related inspections. During the three-stage inspection (land parcel, building, and land number), for spatial conflicts in the vertical direction and those spanning elevations, only planar information can be displayed graphically, with relevant information indicated through elevation markings and textual descriptions. However, this approach cannot visually represent the vertical ownership intersections when two land parcels do not overlap in plan but overlap vertically, especially in land parcel conflicts. Furthermore, in terms of building conflicts, it cannot intuitively display the spatial form of the vertical boundary-crossing portion of a building, making it easy to overlook boundary-crossing issues.

[0057] This application provides a method for checking three-dimensional cadastral survey results. It uses a three-dimensional GIS topology analysis engine to determine the spatial relationship between three-dimensional building models and three-dimensional land parcel models, and combines this with ownership detection to accurately determine the land parcel registration results, effectively ensuring the accuracy of matching building and land parcel ownership. By using a spatial conflict quantification model to calculate boundary distances and overlapping volumes, and integrating planning detection results to determine building registration results, it can accurately identify spatial conflicts between buildings and land parcels, and between adjacent buildings. Finally, by integrating the land parcel registration, building registration, and number registration results, the target inspection result is obtained, providing accurate and reliable evidence for fields such as architectural planning and ownership management.

[0058] It should be noted that the executing entity in this embodiment can be a computing service device with network communication and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or apparatus capable of performing the above functions. The following description uses a three-dimensional cadastral survey result inspection method and device as an example to illustrate this embodiment and the subsequent embodiments.

[0059] Based on this, the embodiments of this application provide a method for checking three-dimensional cadastral survey results, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the three-dimensional cadastral survey results inspection method of this application.

[0060] In this embodiment, the method for checking the three-dimensional cadastral survey results includes steps S10 to S30:

[0061] Step S10: Input the preset 3D building model and the preset 3D land parcel model into the preset 3D GIS topology analysis engine to determine the spatial relationship. Based on the spatial ownership relationship determination result and the ownership detection result, determine the land parcel detection result.

[0062] It should be noted that a land parcel is a plot of land enclosed by ownership boundaries and possessing independent use value. For a land parcel to be eligible for registration, the three-dimensional spatial extent of the real estate unit must be completely contained within the three-dimensional spatial extent of the corresponding land parcel, and the land ownership type must be consistent with the use of the real estate unit.

[0063] In this embodiment, the 3D model of the building includes all vertex coordinates and surface composition information, and the 3D model of the land parcel includes boundary vertex coordinates and elevation information. The ownership information of the building and the land parcel includes text information such as the owner's name and land use right certificate number.

[0064] Specifically, step S10 includes steps S11 to S13:

[0065] Step S11: Determine the spatial relationship based on the 3D GIS topology analysis engine. If the spatial relationship is an inclusion relationship, then determine that the spatial location is compliant.

[0066] In one feasible implementation, step S11 includes: obtaining the minimum bounding box of the building and the minimum bounding box of the land parcel; dividing the minimum bounding box of the building and the minimum bounding box of the land parcel into multiple subspaces based on a preset side length; obtaining the building vertices in each of the subspaces; if the number of intersections between the ray corresponding to the positive Z-axis direction of the building vertex and the boundary of the land parcel is odd, then the building vertex is determined to be inside the land parcel, and an internal vertex list is generated; if the number of intersections is even, then the building vertex is determined to be outside the land parcel, and a superboundary vertex list is generated; if all the building vertices are internal vertices and there are no superboundary vertices, then the spatial relationship between the building and the land parcel is determined to be an inclusion relationship, and the spatial relationship between the building and the land parcel is compliant.

[0067] In this implementation, all vertices of the building model are traversed, and the x-coordinate of each vertex is extracted. The minimum and maximum values ​​are found and used as the building's boundary along the x-axis. The same method is used to find the minimum and maximum values ​​along the y-axis and z-axis. These six values ​​(minimum x, maximum x, minimum y, maximum y, minimum z, maximum z) are combined to form the minimum cuboid parameters that can completely enclose the building, i.e., the building's minimum bounding box. The same operation is performed on the land parcel model to calculate the land parcel's minimum bounding box.

[0068] Based on the minimum bounding box of the land parcel, the 3D space is divided into several subspaces at equal intervals along the x, y, and z axes. The side length of each subspace is set according to the model's precision, such as 1 meter. For each surface of a building, its subspace is determined by checking which subspaces all the vertices of the surface fall into, and the surface is added to the index list of those subspaces. The same operation is performed on each surface of the land parcel, adding it to the corresponding subspace index list, and recording which building surfaces and land parcel surfaces are included in each subspace.

[0069] Establish a mapping table between subspaces and surfaces. For each subspace, query its adjacent subspaces, including those in the top, bottom, left, right, front, back, and diagonal directions, and record the identifiers of these associated subspaces.

[0070] Given any surface, find its subspace through the mapping table, then find all possible related subspaces through the association lookup table, and finally obtain all surfaces within these subspaces.

[0071] For each vertex of a building, sequentially check whether it is located inside the land parcel. Extract all boundary lines from the land parcel surface, construct a 3D polygon boundary, and emit a detection ray in the positive z-axis direction of the vertex. Count the number of intersections with the land parcel boundary. If the number of intersections is odd, the vertex is determined to be inside the land parcel; if it is even, it is outside. For external vertices, calculate the vertical distance to the nearest land parcel boundary, and record the coordinates and specific distance values ​​of all vertices whose distance exceeds the preset allowable out-of-bounds distance.

[0072] Calculate the proportion of internal vertices to the total number of vertices in the building. If the proportion is 100% and there are no outbound vertices, it is considered a complete containment. If the proportion is 100% but there are outbound vertices, and the maximum outbound distance is less than or equal to the preset allowed outbound distance, it is also considered a complete containment, and the spatial relationship between the land parcel and the building is compliant. If there are surface pairs that intersect and the proportion of internal vertices is less than 100%, it is considered a partial intersection. If all surface pairs do not intersect and the minimum distance is less than or equal to a threshold, they are considered adjacent. If the minimum distance is greater than the adjacent threshold, they are considered to have no spatial relationship.

[0073] Step S12: Compare the building ownership information and land parcel ownership information with the ownership registration information to determine the ownership detection results.

[0074] Step S13: If the spatial location is compliant and the ownership detection results are consistent, then the land registration detection result is determined to be qualified.

[0075] In this embodiment, data is read from the building attribute database, including fields such as: building unique identifier (ID_B), owner's name (OWNER_B), land use right certificate number (CERT_ID_B), and registration date. Data is also read from the land parcel attribute database, including fields such as: land parcel unique identifier (ID_Z), owner's name (OWNER_Z), land use right certificate number (CERT_ID_Z), and registration date. Finally, data is read from the relational database, including fields such as: association ID, building ID, land parcel ID, association type, and effective date. Through the spatial association ID between the building and the land parcel, the ownership data of buildings and land parcels in the same spatial unit are bound together, resulting in standardized ownership information and temporary association IDs between the building and the land parcel, containing cleaned owner and certificate number ownership information.

[0076] The system performs a character-by-character comparison between the building owner and the land parcel owner, checking for homophone substitutions and relationships between abbreviations and full names, generating owner comparison results. For building certificate numbers and land parcel certificate numbers, a string equality comparison is performed to check for complete identicalness. The certificate number format is verified for legality, checking if it conforms to encoding rules (e.g., the first two digits are the province code, and the last six digits are the year). If multiple certificate numbers are associated, such as multiple certificate numbers for the same land parcel, the system checks if the building certificate number is in the list of associated land parcel certificate numbers, outputting the certificate number verification result.

[0077] Using the land parcel ID in the building-land parcel ID pair as the query key, retrieve all associated building IDs in the association table. If the number of returned building IDs is greater than 1, it is determined to be a case of multiple rights for one land parcel. Using the owner information corresponding to the building ID as the query key, retrieve all associated land parcel IDs. If the number of returned land parcel IDs is greater than 1, it is determined to be a case of multiple land parcels for one right. Obtain the association rule verification result and a list of specific IDs of the illegal association.

[0078] Integrate all verification results to generate a binary verification matrix. For example, [1,1,0] indicates that the ownership holder is consistent, the document number is consistent, and there is a violation of the association. If all verification items pass, the ownership association is compliant; if any one or more items fail, the ownership association is determined to be abnormal, and the abnormality type is marked. Generate an abnormality detail report, recording the mismatched fields and the illegal association relationships.

[0079] Step S20: Determine the boundary distance between the building and the land parcel, and the overlapping volume between the building and adjacent buildings, based on the preset spatial conflict quantification model. Determine the building drop detection result based on the boundary distance, the overlapping volume, and the planning detection result.

[0080] It should be noted that, according to the "Real Estate Measurement Standard" GB / T 17986-2000, the determination of a building requires verification that its three-dimensional volume is consistent with the planning permit and that the building does not exceed the three-dimensional boundary of the land parcel.

[0081] In this embodiment, the existence of a conflict is determined by calculating the Z-axis boundary difference between the building and the land parcel, as well as the three-dimensional overlap volume between the building and adjacent buildings.

[0082] Specifically, step S20 includes steps S21 to S26:

[0083] Step S21: Traverse the building vertices and land parcel vertices, obtain and determine the superboundary distance based on the building Z-value range and the land parcel Z-value range;

[0084] In this embodiment, all vertices of the building are traversed, the maximum Z value (Z_bmax at the top of the building) and the minimum Z value (Z_bmin at the bottom of the building) are extracted, and the top out-of-bounds value is calculated as follows: ΔZ_up = Z_bmax - Z_max. If the result is positive, it indicates that the building is out of bounds; if negative, it indicates that the building is not out of bounds. The bottom out-of-bounds value is calculated as follows: ΔZ_down = Z_min - Z_bmin. If the result is positive, it indicates that the building is out of bounds; if negative, it indicates that the building is not out of bounds. The top out-of-bounds distance and the bottom out-of-bounds distance are obtained, with positive and negative signs indicating the direction.

[0085] Step S22: Based on the building number of the building, obtain the adjacent buildings that intersect with the building, and obtain the overlapping volume of the building and the adjacent buildings;

[0086] Step S23: When the cross-boundary distance is less than or equal to a preset cross-boundary distance threshold, and the overlapping volume is less than or equal to a preset overlapping volume threshold, it is determined that there is no spatial conflict.

[0087] In this embodiment, adjacent buildings that intersect with the bounding box of the target building are filtered using a spatial index. A range query interface is called from the spatial index database, inputting the x, y, and z coordinate ranges of the building's bounding box. The R-tree or octree structure within the spatial index is traversed to find all buildings whose bounding boxes intersect with the target building's bounding box, i.e., satisfying the condition that the query building's x_min ≤ the target building's x_max, and the query building's x_max ≥ the target building's x_min. The buildings themselves are then filtered by their unique IDs to obtain a list of candidate adjacent buildings and their complete 3D model data.

[0088] For each candidate building, a fine-grained boundary box verification is first performed, calculating the intersection volume of the boundary boxes of two buildings. If the volume is 0, the building is directly excluded. For buildings that pass the verification, a 3D Boolean intersection operation is performed, converting the 3D models of the two buildings into closed polyhedral representations. The divide-and-conquer method is used to calculate the intersecting line segments on the surfaces of the two models sequentially, forming the boundary of the intersection region. A 3D mesh model of the intersection region is constructed based on the intersecting line segments. If the number of vertices in the mesh model of the intersection region is 0, it is determined that there is no spatial intersection. A list of adjacent building IDs with spatial intersection and the 3D mesh models of each intersection region are obtained.

[0089] Calculate the minimum bounding box of the intersection region to determine the spatial range of voxelization. Divide the space into a uniform array of cubic voxels at preset intervals, such as one dividing point every 0.1m in the x-direction. For each voxel, take the coordinates of its center point and check if the point is inside the 3D mesh of the intersection region. If the center point is inside, mark the voxel as a valid voxel and count the number of all valid voxels. Calculate the volume of a single voxel and use the product of the number of valid voxels and the volume of a single voxel as the total overlap volume. Sort adjacent buildings in descending order of overlap volume to obtain a list of adjacent building IDs and the 3D overlap volume between each adjacent building and the target building.

[0090] Establish a mapping relationship between conflict types and regulatory clauses, such as Z-axis boundary overflow corresponding to Article X regarding elevation deviation; extract the allowable error thresholds in the regulations to obtain the corresponding regulatory clauses and applicable compliance thresholds for the conflicts. For Z-axis boundary conflicts, compare the overflow distance with a preset overflow distance threshold. If the overflow distance is less than or equal to the preset overflow distance threshold, it is determined to be a compliant conflict; otherwise, it is a non-compliant conflict. For spatial overlap conflicts, compare the approved building area with the building area information in the 3D building model. If the overflow distance is less than or equal to a preset building area deviation rate, it is determined to be a compliant conflict; otherwise, it is a non-compliant conflict.

[0091] Step S24: Obtain the associated permit information based on the building number, compare the approved number of floors in the permit information with the number of floors in the three-dimensional building model to obtain the floor deviation value, and compare the approved building area in the permit information with the building area in the three-dimensional building model to obtain the building area deviation rate.

[0092] Step S25: When the deviation value of the number of floors is zero and the deviation rate of the building area is less than or equal to the preset building area deviation rate threshold, the planning indicators are determined to be compliant.

[0093] Step S26: When it is confirmed that the planning indicators are compliant and there is no spatial conflict, the building inspection result is determined to be qualified.

[0094] In this embodiment, non-numeric characters in the number are removed using regular expressions to unify the encoding format.

[0095] Perform Chinese word segmentation on the project name, such as converting "XX Community Building 3" to ["XX Community", "Building 3"], and extract the core identifier words. Construct a hash index table using the cleaned building number and license number as keys. Directly compare the building number and license number using the hash index; if the strings are completely identical, they are considered a match.

[0096] The approved number of floors and approved building area are parsed from the matched permit records. The number of floors in the 3D building model is obtained by counting the layers along the Z-axis of the building model. The area of ​​each floor's horizontal projection is calculated using polygons, and then summed, minus wall thickness, to obtain the building area in the 3D building model. The floor number deviation and area deviation rate are calculated. When the floor number deviation = 0 and the relative deviation rate ≤ a preset deviation rate threshold, the planning indicators are marked as compliant; otherwise, they are marked as non-compliant, and the excess items are recorded.

[0097] Step S30: Based on the results of the land parcel detection, the building parcel detection, and the number parcel detection, the target inspection result is obtained.

[0098] It should be noted that the numbering check refers to the requirement that, in accordance with the "Rules for Setting and Compiling Codes for Real Estate Units" GB / T37346-2019, the real estate unit number adopts a 19-digit coding structure of "land parcel code + building number + household number". The numbering must ensure that the number is unique and corresponds one-to-one with the three-dimensional spatial unit.

[0099] In this implementation, the detection status of each detection result is extracted. The target result is judged as qualified only when all three detection results (landing certificate, building, and number) are compliant; if any one detection result is abnormal, the target result is judged as unqualified, and all abnormal information is associated. A structured target inspection report is generated, which includes whether the target inspection result is qualified or unqualified, as well as a summary of abnormality types and quantitative indicators for all detection steps.

[0100] In this implementation, a 3D GIS topology analysis engine is used to determine the spatial relationship between the 3D building model and the 3D land parcel model. Combined with ownership detection, the land parcel detection results are accurately determined, effectively ensuring the accuracy of matching building and land parcel ownership. By using a spatial conflict quantification model to calculate the boundary distance and overlapping volume, and by comprehensively planning detection results, the building parcel detection results are determined, which can accurately identify spatial conflicts between buildings and land parcels, and between adjacent buildings. Finally, the target inspection results are obtained by combining the land parcel, building, and number detection results, providing accurate and reliable basis for fields such as architectural planning and ownership management.

[0101] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 Before step S10, steps S01 to S03 may also be included:

[0102] Step S01: Acquire initial point cloud data based on a 3D laser device, and perform point cloud denoising on the point cloud data according to a preset point cloud density clustering algorithm to remove vegetation points and noise points, thereby obtaining target point cloud data.

[0103] In this embodiment, a laser scanner emits a laser beam to acquire high-point cloud data of buildings and land parcel boundaries. The initial point cloud data is read, and its core information is extracted: three-dimensional coordinates (X / Y / Z), reflection intensity values, and scanning timestamps. Points with Z values ​​exceeding the reasonable elevation range of the survey area and points with empty coordinates are identified as outliers and deleted, retaining only valid points with complete coordinates and intensity values. According to preset coordinate transformation parameters, the initial point cloud's device local coordinate system is converted to the target coordinate system corresponding to the project, such as converting the initial point cloud's device local coordinate system to the 2000 national geodetic coordinate system based on a transformation matrix of GNSS (Global Navigation Satellite System) control points.

[0104] The DBSCAN algorithm is used as the preset clustering algorithm to classify point cloud types based on density connectivity. A preset neighborhood radius and minimum number of points are set according to the average density of the point cloud in the survey area; optionally, the neighborhood radius is set to 1.5-2 times the average spacing of the point cloud. A point is considered a core point when the number of points in its neighborhood is greater than or equal to the minimum number of points. The preprocessed point cloud is traversed, and the number of points in the neighborhood of each point is calculated, marking all core points. Points with fewer points in their neighborhood than the preset minimum number of points, falling within the neighborhood of a core point, are marked as boundary points. Points that are neither core points nor boundary points and have insufficient points in their neighborhood are marked as noise points. Starting from the core points, interconnected core points (i.e., core points indirectly connected through their neighborhoods) are grouped into the same cluster, while boundary points are grouped into the cluster of their respective core points, forming multiple independent point cloud clusters. Based on the characteristics of clusters (density, morphology, reflection intensity), preset classification rules are established: noise points are isolated points not belonging to any cluster; vegetation points are points in small clusters with far fewer points than the target point cluster, and their reflection intensity values ​​are below a preset threshold; target points are points belonging to large clusters, where the number of points within a cluster is large, the density is uniform, the reflection intensity value is above the preset threshold, and the Z-direction coordinate is relatively stable. For initially identified vegetation point clusters, the normal vector direction of the points within the cluster is calculated. If the normal vector direction is chaotic, it is confirmed as a vegetation point; if the normal vector direction is relatively uniform, further screening is performed based on the reflection intensity. Points with low intensity are classified as vegetation, and those with high intensity are reclassified as target points. According to the clustering results, noise points and vegetation point clusters are filtered and deleted, while target point clusters are retained. The processed target point cloud is saved in a preset format.

[0105] Step S02: Acquire an initial image based on the photographic equipment, and stitch the initial image together according to a preset feature point matching algorithm to obtain the target image.

[0106] In this embodiment, a drone equipped with a multi-lens camera acquires oblique images of the real estate. The drone is equipped with an RTK module, such as the DJI M300 RTK. The multi-lens camera includes forward-looking, backward-looking, and front-looking lenses. GNSS control points are evenly distributed in the survey area, and the coordinates (X / Y / Z) of the control points in the 2000 National Geodetic Coordinate System are collected using a GNSS receiver as the reference for subsequent data correction. The survey area is imported, and the flight altitude, overlap rate, and lens angle are set. For example, the lens angle is set so that the front-looking lens is vertically downward, and the forward / backward-looking lenses are at 45° to the vertical direction, generating an automatic flight path. The drone flies along the flight path, and the multiple lenses simultaneously capture images. After each flight (e.g., 15 minutes of flight), the number of images is checked to ensure they match the plan, and the clarity is free of blurring and obstruction. During the flight, the drone's RTK module records the position (X / Y / Z) and attitude (roll angle, pitch angle, yaw angle) of each image capture in real time and stores it as POS data (position and attitude data). Remove blurry, overexposed, or occluded invalid images, retaining only clear and overlapping images that meet the requirements. Obtain camera intrinsic parameters (focal length, principal point coordinates, distortion coefficients) from the camera calibration report to eliminate the impact of lens distortion on image geometric accuracy. Identify GNSS control points in the original images and read their pixel coordinates. Combine the image pixel coordinates, the geodetic coordinates (X / Y / Z) of the control points, and the UAV POS data to correlate the image coordinates with the geodetic coordinates.

[0107] The common features between oblique image data are identified by a preset feature point matching algorithm, the spatial correspondence between different images is determined and stitched together to obtain the target image; optionally, the feature extraction algorithm can be SIFT, SURF, or ORB.

[0108] In one feasible implementation, feature point matching is performed based on the ORB algorithm. Each pixel in the image is traversed, and a neighborhood of a preset size (e.g., 3×3) is selected centered on that pixel. If the grayscale difference between the center pixel and a preset number of pixels within the neighborhood exceeds a preset threshold, that pixel is marked as a candidate corner point. Non-maximum suppression is applied to the candidate corner points, and the response values ​​of the candidate corner points are compared with those of surrounding pixels. The corner point with the largest response value is retained, resulting in a stable FAST corner point set, which serves as the initial feature point set. A neighborhood of a preset size is selected centered on each feature point in the initial feature point set. For each preset number of random point pairs, the grayscale values ​​of each pair are compared to generate a binary descriptor of a preset number of bits. The descriptor undergoes a preset normalization process to enhance its robustness to illumination changes and rotation, resulting in a feature point-to-descriptor mapping table.

[0109] A reference image is selected from the image set. This reference image can be the center of the selected survey area or the first image. The remaining images are used as images to be matched, and each image is matched against the reference image. For each feature descriptor of the reference image, a preset distance is calculated between it and all descriptors of the images to be matched. The descriptor with the smallest distance is found and recorded as the initial matching pair. For each initial matching pair, the ratio of the nearest neighbor distance to the second nearest neighbor distance is calculated. If the ratio is lower than a preset threshold, the matching pair is retained; otherwise, it is considered an incorrect match and deleted. A preset number of optimized matching pairs are randomly selected, and the homography matrix between the two images is solved. The feature point coordinates of the reference image are substituted into the matrix, and their projected coordinates in the image to be matched are calculated. If the reprojection error between the projected coordinates and the actual feature point coordinates is lower than a preset pixel value, it is classified as an inlier and confirmed as a correct match; otherwise, it is classified as an outlier and confirmed as an incorrect match. The sampling is repeated a preset number of times, retaining the homography matrix and inlier set with the largest number of inliers to obtain a high-precision correct matching pair. Based on the homography matrix, a preset type of spatial transformation is performed on the images to be matched, aligning their coordinate systems to the reference image coordinate system to ensure complete overlap of matching feature points between the two images. The extreme coordinate values ​​of the reference image and all registered images to be matched are taken, and the overall range of the registered image is calculated. A blank target image canvas is created, and the reference image is first drawn onto the corresponding position on the canvas. Through coordinate range intersection calculation, areas with forward and lateral overlap rates exceeding preset proportions are identified as overlapping areas of adjacent images. A weighted average fusion algorithm is used to calculate the distance from each pixel within the overlapping area to the edges of the two images, and the final grayscale value is calculated according to a preset formula. Color equalization is performed on the fused overall image to avoid color block issues caused by differences in shooting lighting. The fused image is saved in a preset format to generate a complete target image.

[0110] Step S03: Based on the target point cloud data and the target image, construct the three-dimensional land parcel model and the three-dimensional building model.

[0111] In this embodiment, based on point cloud data and the target image corresponding to the target image, the boundary information of the land parcel is extracted to determine the extent of the land parcel. Using algorithms such as triangulation and Poisson reconstruction, the discrete point cloud data is transformed into a continuous surface model, constructing a 3D land parcel model. The generated 3D land parcel model is then optimized through smoothing, detail enhancement, and other operations to improve the model's accuracy and completeness.

[0112] Point cloud data of buildings is extracted from point cloud data, and preliminary filtering is performed using methods such as height and vegetation filters. Feature information of the buildings, such as roofs, walls, doors, and windows, is extracted to provide a foundation for subsequent modeling. Based on the extracted building feature information, 3D building models are constructed using realistic 3D modeling software such as 3ds Max, Revit, and Geomagic.

[0113] Unique IDs are generated according to hierarchical coding rules, such as land parcel ID: ZD-001, building ID: BJ-ZD-001-01, and unit ID: FY-BJ-ZD-001-01-101. These IDs are embedded into the corresponding model's attribute fields in the modeling software. Using R-tree indexing technology, a spatial index is created for the minimum boundary rectangles of the land parcel and building models. The model's MBR coordinates are stored in the spatial index table, enabling fast model retrieval by spatial range. A mapping relationship is established between the model ID and the database table primary key, such as land parcel model ID = land parcel attribute table land parcel ID, and unit model ID = unit unit attribute table unit ID. The building model containing hierarchical, unit, and component attributes is exported to IFC format, preserving component hierarchy and attributes. The IFC model is imported into GIS software, and spatial attributes from the GIS, such as the land parcel number and coordinate range of the building, are written into the IFC model's attribute fields, completing the integration of BIM and GIS.

[0114] Specifically, step S03 may include steps S031 to S034:

[0115] Step S031: Extract the parcel boundary point set from the target point cloud data based on the preset boundary extraction algorithm, and match and fuse the parcel boundary point set with the boundary point coordinates to generate the parcel boundary.

[0116] In this embodiment, the parcel boundary points are extracted based on the α-shape algorithm. First, ground points are filtered from the target point cloud by classification code or Z-value range (e.g., ground point classification code is 2; points of buildings and vegetation are excluded by Z-value range), resulting in a subset of the ground point cloud containing only the area where the parcel is located. Second, Delaunay triangulation is performed on the ground point cloud subset to generate a non-overlapping triangular mesh that maximizes the minimum angle, ensuring that the mesh completely covers the ground area of ​​the parcel. The circumcircle radius r of each triangle is calculated, corresponding to an α value of 1 / r; a preset α threshold is set, and the edges of triangles with α values ​​≥ the preset α threshold are retained. These edges are connected to form the initial set of parcel boundary points.

[0117] Read the known list of boundary point coordinates. For each point in the initial boundary point set, calculate its Euclidean distance to all other boundary points. If the distance between a boundary point and another boundary point is less than a preset threshold, replace that boundary point with the first boundary point to ensure the boundary passes through all boundary points. Rearrange the matched boundary points in clockwise or counterclockwise order. Connect the first and last points to generate a closed parcel boundary polygon with no self-intersections and at least 3 vertices.

[0118] Step S032: Based on the vertical height of the boundary point along the Z-axis, the land parcel boundary is stretched along the Z-axis to generate the three-dimensional land parcel model.

[0119] In this embodiment, the Z values ​​of all boundary points are read, and the minimum value Zmin and the maximum value Zmax are taken. If Zmax - Zmin is less than a preset difference threshold, the terrain of the parcel is considered to be gentle, and the stretching height is set from Zmin to Zmin + 0.1m. If the difference is greater than or equal to the difference threshold, the stretching height is directly set as Zmin - Zmax. Taking the closed parcel boundary as the bottom surface, it is stretched along the positive Z-axis to the Z = Zmax plane to generate the top surface. Connecting the corresponding vertices of the bottom surface and the top surface forms the side surfaces. Each side surface is a rectangle, and the number of side surfaces is the same as the number of sides of the bottom surface, thus forming a closed three-dimensional parcel model consisting of the bottom surface, top surface, and side surfaces.

[0120] Step S033: Perform point cloud slicing and feature recognition on the target point cloud data to obtain the building outline.

[0121] In this embodiment, building points are filtered from the target point cloud by classification code or Z-value range. For example, the classification code of a building point is 6; or the Z-value range is higher than Zmax+0.5m of the ground point, thus obtaining a subset of the building point cloud. The building point cloud is sliced ​​along the Z-axis according to a preset thickness (e.g., 0.5m, adapted to the floor height), resulting in multiple horizontal slice layers. Each slice layer corresponds to a horizontal section of a certain height of the building.

[0122] For each slice layer, the RANSAC algorithm is used to fit the vertical plane and extract the boundary lines of the plane. The boundary lines of all slice layers are sorted according to the Z-axis height. The complete boundary of the bottom slice is taken as the bottom outline of the building, the boundary of the top slice is taken as the top outline of the building, and the boundary of the middle slice is used for auxiliary correction, finally forming a closed three-dimensional outline of the building.

[0123] For each slice layer, the point cloud density variation is analyzed. The point cloud density in the partition wall area is higher than the thickness of the interior space wall, resulting in point cloud accumulation. The location of the partition wall is located by using a density threshold. A vertical plane is fitted to the point cloud in the partition wall area, and the plane boundary line is extracted as the partition wall boundary. The partition wall boundaries of each slice layer are connected along the Z-axis to form a complete three-dimensional boundary of the partition wall to separate different partition units.

[0124] Step S034: Map the texture information of the target image onto the building outline to obtain the three-dimensional building model.

[0125] In this embodiment, the corresponding texture areas are cropped from the target image according to the building outline. The wall texture is cropped from the side-view image, the roof texture is cropped from the top-view target image, and the detailed textures of windows, balconies, etc., are cropped from the high-resolution target image. The 3D outline of the building is triangulated, dividing the outline surface into a fine triangular mesh. A correspondence is established between the vertices of the triangles and the texture pixels of the target image. For example, the vertices of the triangular mesh of the front wall of the building are bound one by one to the cropped texture pixels of the front wall to ensure that the texture and mesh size are consistent. The matched texture is then applied to the 3D outline surface of the building, and the texture edges are smoothed to eliminate seams, generating a 3D building model with realistic textures.

[0126] Optionally, step S034 may further include: slicing the building point cloud along the Z-axis according to a preset thickness to obtain interval slices, and confirming the median Z-value of the interval slice as the slice feature value; obtaining the slice feature value difference between adjacent interval slices in ascending order of the slice feature values; if the slice feature value difference is greater than a preset floor height threshold, then determining the lower limit of the Z-value interval of the previous interval slice as the floor boundary line;

[0127] In this embodiment, adjacent floors are separated by floor slabs, forming a structure of wall-floor-wall of the next floor. Reflected in point cloud data, this manifests as a sudden jump in the Z-value from a continuous distribution range of the wall to the distribution range of the wall of the next floor, with the jump magnitude matching the normal floor height. The three-dimensional data is converted into a two-dimensional height distribution along the Z-axis by slicing the point cloud data. Then, statistical analysis is used to locate the Z-value abrupt change point that matches the floor height range; this point is the floor boundary line.

[0128] Along the Z-axis, the point cloud of the main building is horizontally sliced ​​with a fixed thickness, forming a series of Z-value interval slices parallel to the ground. The fixed thickness can be 0.1m-0.5m, and must be less than the floor slab thickness. Each slice corresponds to a fixed Z-value range and includes the point cloud of all main buildings within that range. For each slice, the median or mean Z-value of the point cloud within the slice is identified as the slice feature Z-value. Calculate the difference in Z-values ​​between adjacent valid slices in ascending order of feature Z-values. The formula is: Difference = Z-value of the previous slice - Z-value of the current slice. If the difference is less than a preset threshold, it means the two slices correspond to the same building component, such as different height segments of the same wall, with continuous Z-values ​​without abrupt changes. If the difference is greater than or equal to the preset threshold, which corresponds to the minimum floor height of a typical residential / office building, it means the two slices correspond to different building components. The current slice is the top of the wall on the previous floor, and the previous slice is the bottom of the wall on the next floor. The Z-value shows an abrupt change consistent with the floor height pattern. Record the lower limit of the Z-value interval of the previous slice as a candidate point for abrupt change. The corresponding lower limit of the Z-value interval of the previous slice is the floor boundary line, which corresponds to the top height of the floor slab on the previous floor and also the ground height of the next floor. For example, if the current slice Z = 4.8m-4.9m, the previous slice Z = 7.6m-7.7m, the difference is 2.8m, and the abrupt change point is Z = 7.6m.

[0129] Optionally, for a mutation point, it is necessary to verify the point cloud density change of the slices above and below it. The slice below the mutation point should be a high-density point cloud slice, corresponding to the top of the wall of the previous floor. The slice above the candidate point should also be a high-density point cloud slice, corresponding to the bottom of the wall of the next floor. The point cloud density of the low-density / empty slices between the two should be extremely low, which is consistent with the structural characteristics of thin floor slabs and few point clouds. This verification can eliminate false mutations caused by non-building structures, such as local Z-value jumps caused by window openings.

[0130] Optionally, step S034 may further include: segmenting the three-dimensional building model based on the floor boundary line to obtain floor sub-models; identifying the partition walls through vertical plane clustering, and dividing the floor sub-model into multiple partition units based on the boundary coordinates of the partition walls to obtain a partition model;

[0131] In this embodiment, the overall building model is horizontally cut along each floor boundary line, dividing the model into multiple independent floor sub-models. Each sub-model contains only building components within the corresponding Z-value range; for example, a floor sub-model contains one wall and one floor slab, resulting in a set of floor sub-models with floor numbers. The floor number and unit number are used as association keys to input ownership information into the attribute database of the corresponding unit.

[0132] The RANSAC algorithm was used to perform planar fitting on the floor point cloud. A planar fitting error threshold was set, and all vertical planes were extracted and identified as wall planes. The number of point clouds, plane equation (Ax+By+Cz+D=0), and boundary range of each plane were recorded. Apartment partition walls were filtered based on thickness and function, while non-partition walls within the floor were excluded. The thickness of the plane was calculated by taking the distance between adjacent points on both sides. If the thickness was within the preset partition wall thickness range, it was marked as a candidate partition wall. If the point clouds on both sides of the wall belonged to two independent spaces and the wall extended to the floor boundary, it was confirmed as a partition wall; if the wall only separated areas within the same room, it was identified as a non-partition wall and excluded.

[0133] Optionally, a region growing algorithm is used to divide independent spaces based on point cloud density and spatial connectivity. An unlabeled point is randomly selected from the point cloud as a seed point, and all adjacent points within its preset radius are searched. If the Z-axis difference between an adjacent point and the seed point is less than or equal to a preset difference, and the point cloud density is greater than or equal to a preset point cloud density threshold (e.g., 5 points / m²), ensuring it is an effective space rather than an empty area, then the adjacent points are grouped into the same region. The seed point selection and adjacent point merging process is repeated until all points are labeled, ultimately generating multiple non-overlapping spatially connected regions. Each region corresponds to a potential independent space, and N spatially connected regions within a single floor are output. Each region is accompanied by a unique ID, region boundary coordinates, and the set of point clouds it contains.

[0134] For each candidate partition wall, its planar equation Ax+By+Cz+D=0 is extracted. Extending 0.1m to each side along the wall's plane normal vector, this extends to cover the wall and adjacent spaces on both sides, generating a buffer zone for the wall. All connected spatial regions are traversed, and it is determined whether points within these regions intersect with the wall's buffer zone. If the point cloud of a region intersects with the buffer zone, then that region is considered an adjacent spatial region of the wall. The number of adjacent spatial regions for each candidate wall is counted. If the number is 2, meaning there is one different connected spatial region on each side of the wall, then the wall is considered to separate two independent spaces and is marked as a partition wall to be verified. If the number is 1, meaning both sides of the wall belong to the same connected spatial region, then the wall only separates areas within the same room and is marked as a non-partition wall and excluded.

[0135] For each partition wall to be verified, its 2D projected boundary on the floor plan is extracted. The two endpoints of the projected boundary are extracted, and the direction vector of the line connecting the endpoints is used to determine the wall's extension direction. Extension segments are generated along this direction. It is then determined whether the extension segments intersect the floor boundary. If both endpoints intersect the floor boundary, the wall is considered to extend to the floor boundary and is ultimately marked as a partition wall. If only one endpoint intersects the floor boundary, or neither intersects, the wall is considered not to extend to the floor boundary, is marked as a non-partition wall, and is excluded. The final official partition wall simultaneously satisfies both conditions: separating two independent spaces and extending to the floor boundary. It also includes the IDs of the connected regions of the two spaces and the complete boundary coordinates.

[0136] For the confirmed sub-unit walls, their projected boundaries on the floor plan are extracted. The point cloud of the sub-unit walls is projected onto this plane to obtain a two-dimensional point set. The α-shape algorithm is used to extract the closed boundaries of the point set, resulting in the two-dimensional boundary polygons of the sub-unit walls. All sub-unit wall boundaries are superimposed with the building's exterior wall boundaries to form multiple closed two-dimensional polygon regions, each region being a sub-unit. Temporary unit numbers are assigned to the sub-units in a left-to-right, east-to-west order. Simultaneously, the coordinates of the boundary vertices of each sub-unit are recorded to ensure no overlap or omission.

[0137] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description and will not be repeated hereafter. Based on this, before step S30, steps A10 to A30 may be included:

[0138] Step A10: Traverse the real estate unit numbers, determine whether the real estate unit numbers are unique, and obtain a list of unique unit numbers.

[0139] Step A20: Obtain the ownership information associated with the unique unit number in the unique unit number list, and compare the ownership information with the ownership information in the real estate registration database.

[0140] Step A30: If the comparison results are consistent, then the serial number detection is deemed qualified.

[0141] In this embodiment, in accordance with the "Rules for Setting and Compiling Codes for Real Estate Units" (GB / T 37346-2019), the 19-digit code is split into 6 levels: administrative division (6 digits), cadastral district (3 digits), cadastral sub-district (3 digits), parcel (5 digits), fixtures (2 digits), and individual households (4 digits).

[0142] Enter the information for each level in sequence to generate the initial real estate unit number.

[0143] The length of the unit number is counted character by character. If it is not the specified number of digits, it is marked as an error due to incorrect code length; the actual length is multiplied by the number of digits. The code is split into 6 segments according to the coding rules. The administrative division segment must be numeric and must exist in the national administrative division code database. If it contains letters or is not in the database, the administrative division segment is marked as invalid. The cadastral area (digits 7-9) and cadastral sub-area (digits 10-12) are both numeric, ranging from 001 to 999. If they contain letters or exceed the range, the cadastral area / sub-area code is marked as invalid. The parcel number (digits 13-17) has the following characteristics: the first two digits are the parcel feature code (e.g., GB represents a state-owned construction land use right parcel), and the last three digits are the parcel sequence number. If the feature code is not a specified letter or the sequence number exceeds the range, the parcel number is marked as invalid. The fixture code (digit 18) is a specified letter (e.g., F represents a house). If it is any other letter or number, the fixture code is marked as invalid. The unit number (19th digit) is a number. If it exceeds the preset range, the unit number code is marked as invalid. The system queries the total number of unit models for the corresponding building. If the unit number is greater than the total number of models, it marks the unit number as mismatched with the actual number of units. This process yields a list of unit numbers that passed the rule validation, a list of unit numbers that failed the validation, and detailed error reasons.

[0144] For each unit number in the list of unit numbers that pass the rule validation, execute the SQL query: SELECT unit_code FROM unit_codes WHERE unit_code = 'Unit Number to be Verified'. If the query result set is empty, the unit number is determined to be unique and added to the list of unique unit numbers.

[0145] For each unit number in the unique unit number list, execute the SQL query: `SELECT owner_name,area, cert_no FROM ownership WHERE unit_code = 'Unit Number to be Verified'`. If the query result is empty, mark it as having no ownership association and add it to the number association anomaly list. If the result set has exactly one record, check field consistency: the unit number must completely match the `unit_code` in the ownership table. Information such as the owner, area, and identification number must be complete and conform to the format requirements (e.g., area must be positive, and identification number must conform to the encoding rules). If all conditions are met, the association is considered consistent, and it is added to the number association pass list. If the result set has multiple records, mark one unit as having multiple ownership, add it to the number association anomaly list, and record all associated ownership information.

[0146] Based on the first embodiment of this application, in the fourth embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3The above-mentioned method for checking the results of three-dimensional cadastral surveys may also include steps B10 to B30:

[0147] Step B10: When real-time data is received, a data identifier is generated based on the land parcel code and update timestamp of the real-time data;

[0148] In this embodiment, the real-time data received via the RESTful API includes land parcel codes, update timestamps, 3D coordinates, and attribute information. Required fields, such as the land parcel code and update timestamp, are checked for existence. The coordinate format is verified, ensuring x, y, and z are floating-point numbers and that the range conforms to the regional coordinate system. Invalid data is stored in an exception queue, triggering an alarm to notify maintenance personnel. The land parcel code and update timestamp are extracted, concatenated into a composite string, and a hash value is calculated using the SHA-256 algorithm to generate a unique data identifier.

[0149] Step B20: Determine the corresponding hash value based on the data identifier, match the hash value in the database, and obtain the new data;

[0150] In this embodiment, the historical data identifier index table is read from the database, and an index query is performed on the hash value of each real-time data entry: `SELECT data_id FROM identifier_index WHERE hash = 'current hash value'`. If the query result is empty, it is marked as new data, and the complete data content, including 3D coordinates and attributes, is recorded. If the query result is not empty, it is determined to be duplicate data, and the processing is skipped. This yields a list of new data containing 3D model data and attribute information.

[0151] Step B30: Update the three-dimensional building model and the three-dimensional land parcel model based on the model difference algorithm and the newly added data.

[0152] In this embodiment, for a newly added 3D model, the vertex set V_new (containing x, y, z coordinates) and the face set F_new (each face consists of 3 vertex indices) are extracted. The model's topological relationships (such as the inclusion relationship between land parcels and buildings, and the connection relationship between adjacent buildings) are analyzed. The minimum bounding box of the new model is calculated, and intersection detection is performed with the bounding boxes in the existing model library to locate the region to which the new model belongs in 3D space. The vertex set V_new of the new model is added to the global vertex library, and the face set F_new is added to the global face library, associating the new vertex IDs. The bounding box and vertex information of the new model are inserted into the R-tree index to ensure that subsequent queries can identify it.

[0153] Send the vertex and face data of the newly added model to the visualization engine, perform highlight rendering on the newly added region, and update the model attribute labels. Create a new version number for the updated model library, record the update time and the amount of new data (number of vertices, number of faces) to the version log, and obtain a real-time updated visualization interface.

[0154] This application provides a three-dimensional cadastral survey result inspection method and apparatus, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the three-dimensional cadastral survey result inspection method in the above embodiment 1.

[0155] The following is for reference. Figure 4 The diagram illustrates a structural schematic of an apparatus suitable for implementing the three-dimensional cadastral survey result inspection method in the embodiments of this application. The apparatus for the three-dimensional cadastral survey result inspection method in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, personal digital assistants (PDAs), tablet computers (PADs), and vehicle terminals (e.g., vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 4 The method and equipment for checking the three-dimensional cadastral survey results shown are merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0156] like Figure 4As shown, the three-dimensional cadastral survey result inspection method device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that performs various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The random access memory 1004 also stores various programs and data required for the operation of the three-dimensional cadastral survey result inspection method device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the three-dimensional cadastral survey result inspection method equipment to exchange data wirelessly or via wired communication with other devices. Although the figure shows a three-dimensional cadastral survey result inspection method equipment with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0157] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0158] The three-dimensional cadastral survey result inspection method and equipment provided in this application, employing the three-dimensional cadastral survey result inspection method in the above embodiments, can solve the technical problem of how to improve the accuracy of cadastral survey result detection. Compared with the prior art, the beneficial effects of the three-dimensional cadastral survey result inspection method and equipment provided in this application are the same as those of the three-dimensional cadastral survey result inspection method provided in the above embodiments, and other technical features in this three-dimensional cadastral survey result inspection method and equipment are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0159] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0160] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0161] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the three-dimensional cadastral survey result inspection method in the above embodiments.

[0162] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, radio frequency (RF), etc., or any suitable combination thereof.

[0163] The aforementioned computer-readable storage medium may be included in the three-dimensional cadastral survey result inspection method device; or it may exist independently and not assembled into the three-dimensional cadastral survey result inspection method device. The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by the three-dimensional cadastral survey result inspection method device, the three-dimensional cadastral survey result inspection method device: inputs a preset three-dimensional building model and a preset three-dimensional parcel model into a preset three-dimensional GIS topology analysis engine to perform spatial relationship judgment; determines the parcel location detection result based on the spatial relationship judgment result and the ownership detection result; determines the boundary distance between the building and the parcel, and the overlapping volume between the building and adjacent buildings, based on a preset spatial conflict quantification model; determines the building location detection result based on the boundary distance, the overlapping volume, and the planning detection result; and obtains the target inspection result based on the parcel location detection result, the building location detection result, and the numbering detection result.

[0164] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the client computer, partially on the client computer, as a standalone software package, partially on the client computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the client computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0165] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0166] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0167] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described three-dimensional cadastral survey result inspection method, thereby solving the technical problem of how to improve the accuracy of cadastral survey result detection. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the three-dimensional cadastral survey result inspection method provided in the above embodiments, and will not be repeated here.

[0168] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for inspecting three-dimensional cadastral survey results, characterized in that, The methods for checking the results of the three-dimensional cadastral survey include: The preset 3D building model and preset 3D land parcel model are input into a preset 3D GIS topology analysis engine for spatial relationship judgment. Based on the spatial relationship judgment result and the ownership detection result, the land parcel detection result is determined, including: determining the spatial relationship based on the 3D GIS topology analysis engine; if the spatial relationship is an inclusion relationship, the spatial relationship is determined to be compliant; comparing the building ownership information and land parcel ownership information with the ownership registration information to determine the ownership detection result; if the spatial relationship is compliant and the ownership detection result is consistent, the land parcel detection result is determined to be qualified. The boundary distance between the building and the land parcel is determined based on the preset spatial conflict quantification model, as well as the overlapping volume between the building and the adjacent building. The building drop detection result is determined based on the boundary distance, the overlapping volume, and the planning detection result. Based on the results of the land parcel detection, the building detection, and the number detection, the target inspection results are obtained; The step of determining the spatial relationship based on the 3D GIS topology analysis engine, and determining the compliance of the spatial relationship if the spatial relationship is an inclusion relationship, includes: Obtain the minimum bounding box of the building and the minimum bounding box of the land parcel. Divide the minimum bounding box of the building and the minimum bounding box of the land parcel into multiple subspaces based on a preset side length. Obtain the building vertices in each of the subspaces. If the number of intersections between the ray corresponding to the positive Z-axis of the building vertex and the boundary of the land parcel is odd, then the building vertex is determined to be inside the land parcel, and an internal vertex list is generated. If the number of intersections is even, then the building vertex is determined to be outside the land parcel, and a superboundary vertex list is generated. If all the building vertices are internal vertices and there are no superboundary vertices, then the spatial relationship between the building and the land parcel is determined to be an inclusion relationship, and the spatial relationship between the building and the land parcel is compliant.

2. The method for checking three-dimensional cadastral survey results as described in claim 1, characterized in that, The steps for determining the boundary distance between a building and a land parcel, and the overlapping volume between the building and adjacent buildings, based on a preset spatial conflict quantification model, and determining the building drop detection result according to the boundary distance, the overlapping volume, and the planning detection result, include: Traverse the building vertices and land parcel vertices, obtain and determine the hyperboundary distance based on the building Z-value range and the land parcel Z-value range; Based on the building number of the building, obtain the adjacent buildings that intersect with the building, and obtain the overlapping volume of the building and the adjacent buildings; If the superboundary distance is less than or equal to a preset superboundary distance threshold, and the overlapping volume is less than or equal to a preset overlapping volume threshold, then it is determined that there is no spatial conflict. When the number of floors deviation is zero and the building area deviation rate is less than or equal to the preset building area deviation rate threshold, the planning indicators are deemed compliant. If the planning indicators are confirmed to be compliant and there are no spatial conflicts, then the building inspection result is determined to be qualified.

3. The method for checking three-dimensional cadastral survey results as described in claim 2, characterized in that, Before the step of determining that the planning indicators are compliant when the floor number deviation is zero and the building area deviation rate is less than or equal to a preset building area deviation rate threshold, the three-dimensional cadastral survey result inspection method further includes: Based on the building number, obtain the associated permit information, compare the approved number of floors in the permit information with the number of floors in the three-dimensional building model, and obtain the floor deviation value; The approved building area in the license information is compared with the building area in the three-dimensional building model to obtain the building area deviation rate.

4. The method for checking three-dimensional cadastral survey results as described in claim 1, characterized in that, Before the step of inputting the preset 3D building model and the preset 3D land parcel model into the preset 3D GIS topology analysis engine to determine the spatial conflict check results between the land parcel and the building, the 3D cadastral survey result checking method further includes: Initial point cloud data is acquired using a 3D laser device. The point cloud data is then denoised using a preset point cloud density clustering algorithm to remove vegetation points and noise points, thereby obtaining target point cloud data. The initial image is acquired using a photographic device, and the initial image is stitched together according to a preset feature point matching algorithm to obtain the target image. Based on the target point cloud data and the target image, the three-dimensional land parcel model and the three-dimensional building model are constructed.

5. The method for checking three-dimensional cadastral survey results as described in claim 4, characterized in that, The steps of constructing the 3D land parcel model and the 3D building model based on the target point cloud data and the target image include: Based on a preset boundary extraction algorithm, the set of land parcel boundary points is extracted from the target point cloud data, and the set of land parcel boundary points is matched and fused with the coordinates of the boundary points to generate the land parcel boundary. Based on the vertical height of the boundary point along the Z-axis, the land parcel boundary is stretched along the Z-axis to generate the three-dimensional land parcel model; The target point cloud data is sliced ​​and feature-recognized to obtain the building outline and the boundary of the partition wall; The texture information of the target image is mapped onto the building outline to obtain the three-dimensional building model.

6. The method for checking three-dimensional cadastral survey results as described in claim 5, characterized in that, The step of performing point cloud slicing and feature recognition on the target point cloud data to obtain the building outline and the boundary of the partition wall includes: Along the Z-axis, the building point cloud is sliced ​​according to a preset thickness to obtain interval slices, and the median Z value of the interval slice is determined as the slice feature value. According to the slice feature values ​​in ascending order, the slice feature value difference between adjacent interval slices is obtained. If the slice feature value difference is greater than the preset floor height threshold, the lower limit of the Z value interval of the previous interval slice is determined as the floor boundary line.

7. The method for checking three-dimensional cadastral survey results as described in claim 6, characterized in that, The method for checking the results of the three-dimensional cadastral survey also includes: The three-dimensional building model is divided based on the floor boundary lines to obtain floor sub-models; The floor sub-model is divided into multiple sub-units based on the boundary coordinates of the sub-unit walls by identifying the sub-unit walls through vertical plane clustering, thus obtaining the sub-unit model.

8. The method for checking three-dimensional cadastral survey results as described in claim 1, characterized in that, The method for checking the results of the three-dimensional cadastral survey also includes: Upon receiving real-time data, a data identifier is generated based on the parcel code and update timestamp of the real-time data. Based on the data identifier, the corresponding hash value is determined, and the hash value is matched in the database to obtain the new data; Based on the model difference algorithm and the newly added data, the three-dimensional building model and the three-dimensional land parcel model are updated.

Images