Building point cloud gap filling viewpoint planning method, device, equipment and storage medium

By triangulating and mapping the point cloud of the building IFC model, calculating the gap score, and selecting high-value viewpoints, the problem of inaccurate point cloud defect localization in existing technologies is solved, thereby improving the quality and efficiency of building scanning.

CN123118955APending Publication Date: 2026-11-17INTELLIGENT RELOCATION SPACE (CHANGZHOU) TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611636187.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-22
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies cannot accurately locate defects in building point cloud data to the surface of specific components, resulting in a lack of targeting in the selection of supplementary scanning viewpoints, low efficiency, and easy omission of key areas.

Method used

By triangulating the building IFC model and establishing surface blocks, and combining the coordinate alignment and normal constraints of the 3D point cloud, the gap fraction is calculated, and high-value fill viewpoints are selected to cover the gap area.

Benefits of technology

It enables precise positioning and reasonable supplementary scanning of gaps in building point clouds, improving scanning quality and efficiency, and reducing the risk of repeated scanning and missing key areas.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123118955A_ABST
    Figure CN123118955A_ABST
Patent Text Reader

Abstract

This application discloses a method, apparatus, equipment, and storage medium for planning viewpoints for filling gaps in architectural point cloud scanning, relating to the field of 3D reconstruction technology. The method includes triangulation, point cloud mapping, gap targeting, and viewpoint selection. It utilizes the properties of normal dot product to measure the degree of offset between different triangular faces, and establishes first and second normal constraints accordingly to group triangular faces with smaller offsets into the same surface block. This constructs the component surface into several precise surface blocks, enabling the local quality gaps in the existing point cloud to be located to precise areas on the component surface. By combining the gap fraction, area, and corresponding observation quality of the surface blocks, the value of candidate viewpoints for filling gaps is calculated, allowing candidate viewpoints to preferentially cover areas with high gaps and good observation quality, thus more rationally selecting viewpoints and effectively improving the scanning quality of the building.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of 3D reconstruction technology, and in particular to a method, apparatus, equipment and storage medium for planning viewpoints for filling gaps in building point clouds. Background Technology

[0002] In 3D scanning projects for both interior and exterior buildings, due to limitations imposed by site conditions and equipment performance, single-station laser scanning or photogrammetry often cannot completely cover the entire surface of the target building. To obtain complete point cloud data, a multi-station, multi-view stitching acquisition strategy is typically required. However, even with careful path planning, the final point cloud model often suffers from defects such as missing data coverage, insufficient point cloud density, and a single observation direction on local building surfaces due to a combination of factors including mutual occlusion between components, excessive scanning incident angle, scanning distance exceeding the effective range, limited field of view, cumulative errors in multi-station point cloud registration, and physical accessibility limitations of the acquisition path.

[0003] Currently, the evaluation of scan quality mainly relies on global integrity indicators (such as overall point cloud coverage and average point spacing within the point cloud) or manual visual inspection. These methods only reflect the overall macroscopic completeness of the point cloud; they cannot precisely locate data defects to specific surface sections of particular IFC components (such as the lower flange of a steel beam or the inner corner of a wall), nor can they quantify the actual impact of each defective area on subsequent modeling and analysis tasks. This coarse-grained scan quality evaluation method directly leads to a lack of clear spatial targeting in subsequent supplementary scans. Engineers often have to rely on experience to perform random re-scans, which is not only inefficient but also prone to causing numerous duplicate scans or missing critical areas.

[0004] To address the aforementioned issues, Next Best View (NBV) planning theory has been introduced into the field of architectural scanning, aiming to calculate the next scan pose that maximizes information gain in order to fill data gaps. Some NBV methods primarily rely on geometric coverage, environmental unknowns, or other global benefit indicators. However, how to pinpoint the local quality gaps in existing point clouds to precise areas on the surface of IFC model components to more rationally select the supplementary scan viewpoint remains an unsolved problem in this field. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a method, apparatus, equipment and storage medium for planning the scanning viewpoint for filling gaps in building point clouds, so as to locate the local quality gaps in the existing point cloud to a precise area on the surface of the component to select the scanning viewpoint more reasonably and improve the scanning quality of the building.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] Firstly, this application provides a viewpoint planning method for filling gaps in building point clouds, including:

[0008] Obtain the IFC model of the building, which includes several components; divide the surface of each component into triangles to construct each component into several surface blocks composed of several triangular faces;

[0009] Obtain a 3D point cloud corresponding to the building, the 3D point cloud including a number of sampling points; align the 3D point cloud with the IFC model in coordinates, and match each sampling point with at most one surface block according to the minimum distance between each sampling point and each triangular face;

[0010] The geometric coverage and point cloud density of the three-dimensional point cloud on each of the surface blocks are calculated and weighted normalized to obtain the gap score of each surface block, and the surface block with the highest gap score is selected as the target surface block.

[0011] The target surface is divided into blocks to generate several candidate supplementary viewpoints. The value of each candidate supplementary viewpoint is calculated based on the view frustum coverage of each candidate supplementary viewpoint, and the candidate supplementary viewpoint with the highest value is selected as the target supplementary viewpoint.

[0012] Optionally, the step of triangulating the surface of each of the components to construct each component into a plurality of surface blocks composed of a plurality of triangular faces includes:

[0013] The surface of each component is triangularly divided to construct a plurality of triangular faces on the surface of each component.

[0014] Based on the adjacency and normal relationships between the triangular faces corresponding to each component, the triangular faces corresponding to each component are summarized respectively, so that the surface of each component is constructed as a number of surface blocks summarized from the corresponding triangular faces.

[0015] Optionally, the step of summarizing the triangular faces corresponding to each component based on the adjacency and normal relationships between the triangular faces corresponding to each component, so as to construct the surface of each component into several surface blocks summarized from the corresponding triangular faces, includes:

[0016] In any of the aforementioned components, a triangular face that is not included in an existing surface block is selected as a seed triangular face, and the remaining triangular faces that are not included in an existing surface block and have a common edge with the seed triangular face are selected as candidate triangular faces.

[0017] Based on the normals of the seed triangle and the candidate triangle in the component, a first normal constraint and a second normal constraint are established. The seed triangle and the candidate triangles satisfying the first and second normal constraints are then summarized one by one to obtain corresponding surface blocks. The first normal constraint is that the dot product of the normals of the candidate triangles with a common edge is not less than a preset value. The second normal constraint is that the dot product of the normals of the seed triangle and the candidate triangle is not less than a preset value.

[0018] The IFC model is traversed, and the above steps are repeated for each component until there are no triangular faces in any component that have not been included in the existing surface blocks, so that the surface of each component is constructed as a number of surface blocks that are included by the corresponding number of triangular faces.

[0019] Optionally, matching each sampling point with at most one surface block based on the minimum distance between each sampling point and each of the triangular faces includes:

[0020] For any of the sampling points, calculate the minimum distance between it and each of the triangular faces, and select the surface block to which the triangular face corresponding to the minimum distance is classified as the nearest surface block of the sampling point. When the minimum distance corresponding to the sampling point does not exceed a preset value, match the sampling point with at most one of the nearest surface blocks corresponding to it.

[0021] The three-dimensional point cloud is traversed, and the above steps are performed on each of the sampling points to match each of the sampling points with at most one of the surface blocks.

[0022] Optionally, the step of calculating the geometric coverage and point cloud density of the 3D point cloud on each of the surface blocks and performing weighted normalization to obtain the gap fraction of each of the surface blocks includes:

[0023] In any of the surface segments, the geometric coverage of the 3D point cloud on the surface segment is calculated based on the proportion of the observed portion of the surface segment to the total surface segment.

[0024] The point cloud density of the three-dimensional point cloud on the surface block is calculated based on the ratio of the number of sampling points matched by the surface block to the area of ​​the surface block.

[0025] The geometric coverage and point cloud density of the three-dimensional point cloud on the surface block are assigned corresponding weights and then weighted and normalized to obtain the gap fraction of the surface block.

[0026] The above steps are performed on each of the surface blocks in the IFC model to obtain the notch fraction of each surface block.

[0027] Optionally, the weighted normalization formula for the gap score is:

[0028] ;

[0029] In the formula, In the IFC model, the first... The number of notches in each of the aforementioned surface segments. , These respectively represent the three-dimensional point cloud at the th... Geometric coverage and point cloud density on each of the aforementioned surface segments This represents the preset desired density of the three-dimensional point cloud for any of the surface segments. and All are weighted factors and .

[0030] Optionally, the step of generating a plurality of candidate supplementary viewpoints by dividing the target surface into blocks, and calculating the value of each candidate supplementary viewpoint based on the view frustum coverage of each candidate supplementary viewpoint, includes:

[0031] For any candidate supplementary viewpoint, the value of the candidate supplementary viewpoint is calculated based on the sum of the products of the notch fraction, area, and corresponding observation quality of each surface patch within the cone coverage area of ​​the candidate supplementary viewpoint; the observation quality corresponding to the surface patch is obtained based on the orthographic, distance quality, and resolution quality of the surface patch from the candidate supplementary viewpoint.

[0032] The above steps are performed on each of the candidate supplementary viewpoints to calculate the value of each candidate supplementary viewpoint.

[0033] Secondly, this application provides a viewpoint planning device for filling gaps in building point clouds, comprising:

[0034] A triangular construction module is used to obtain the IFC model of a building, which includes several components; the surface of each component is triangularly divided to construct each component into several surface blocks composed of several triangular faces.

[0035] The point cloud mapping module is used to acquire the three-dimensional point cloud corresponding to the building, the three-dimensional point cloud including a number of sampling points; the three-dimensional point cloud is aligned with the IFC model in coordinates, and each sampling point is matched with at most one surface block according to the minimum distance between each sampling point and each triangular face;

[0036] The gap targeting module is used to calculate the geometric coverage and point cloud density of the three-dimensional point cloud on each of the surface blocks and perform weighted normalization to obtain the gap score of each surface block, and select the surface block with the highest gap score as the target surface block.

[0037] The viewpoint selection module is used to generate several candidate supplementary viewpoints by dividing the target surface into blocks, calculate the value of each candidate supplementary viewpoint based on the view frustum coverage of each candidate supplementary viewpoint, and select the candidate supplementary viewpoint with the highest value as the target supplementary viewpoint.

[0038] Thirdly, this application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the aforementioned building point cloud gap filling viewpoint planning method.

[0039] Fourthly, this application provides a computer-readable storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the aforementioned building point cloud gap filling viewpoint planning method.

[0040] The present application adopts the above technical solution, and its beneficial effects compared with the prior art are as follows:

[0041] This application utilizes the properties of normal dot product to measure the degree of offset between different triangular faces, and establishes a first normal constraint and a second normal constraint accordingly. Triangular faces with smaller offsets are grouped into the same surface block, thereby constructing the component surface into several precise surface blocks. This allows for the local quality gaps of existing point clouds to be located to precise areas on the component surface. By combining the gap fraction, area, and corresponding observation quality of the surface blocks, the value of candidate supplementary scanning viewpoints is calculated, enabling candidate supplementary scanning viewpoints to preferentially cover areas with high gaps and good observation quality. This allows for more rational selection of supplementary scanning viewpoints and effectively improves the scanning quality of the building. Attached Figure Description

[0042] 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, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0043] Figure 1 A flowchart of a viewpoint planning method for filling gaps in building point clouds provided in this application embodiment.

[0044] Figure 2 This is a schematic diagram of a component surface provided in an embodiment of this application.

[0045] Figure 3 This is a schematic diagram of a surface segmentation provided in an embodiment of this application.

[0046] Figure 4 This is a schematic diagram of a candidate supplementary scan viewpoint provided in an embodiment of this application.

[0047] Figure 5 This is a schematic diagram of a candidate supplementary scan viewpoint provided in an embodiment of this application.

[0048] Figure 6 A structural diagram of a viewpoint planning device for filling gaps in building point clouds provided in this application embodiment.

[0049] Figure 7 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0050] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0051] See Figure 1 This application discloses a viewpoint planning method for filling gaps in building point clouds, comprising the following steps:

[0052] S1. Obtain the IFC model of the building, which includes several components; triangulate the surface of each component to construct each component into several surface blocks composed of several triangular faces.

[0053] Figure 2 This is a schematic diagram of a component surface provided in an embodiment of this application. See also... Figure 2 After triangulating the surface of the component, six triangular faces numbered A1-A6 are obtained.

[0054] Figure 3 This is a schematic diagram of surface segmentation provided in an embodiment of this application. See also... Figure 3 ,right Figure 2 After summarizing the six triangular faces shown, we get surface blocks (shaded areas) formed by summarizing the three triangular faces numbered A1-A3.

[0055] In some implementations of this embodiment, the surfaces of each component are triangularly divided to construct several triangular faces. Then, based on the adjacency and normal relationships between the triangular faces of each component, the triangular faces of each component are summarized to construct several surface blocks formed by summarizing the corresponding triangular faces.

[0056] Specifically, in any component, a triangular face that is not included in the existing surface blocks is selected as a seed triangular face, and the remaining triangular faces that are not included in the existing surface blocks and have a common edge with the seed triangular face are selected as candidate triangular faces.

[0057] by Figure 2 Taking the component surface shown as an example, the undirected edge V1 between vertex E1 and vertex E2 is the common edge of triangle A1 and triangle A4, the undirected edge V2 between vertex E1 and vertex E3 is the common edge of triangle A4 and triangle A5, and the undirected edge V3 between vertex E2 and vertex E3 is the common edge of triangle A4 and triangle A6.

[0058] And with Figure 3 Taking the surface segmentation shown as an example, under the condition that triangle A1 is selected as the seed triangle, since triangles A2-A4 have a common edge with triangle A1, and triangles A5 and A6 do not have a common edge with triangle A1, triangles A2-A4 are selected as candidate triangles.

[0059] Subsequently, based on the normals of the seed triangle and the candidate triangle in the component, a first normal constraint and a second normal constraint are established. The seed triangle and the candidate triangles that satisfy the first normal constraint and the second normal constraint are summarized one by one to obtain the corresponding surface blocks. The above steps are repeated for each component until there are no triangles in any component that have not been summarized into the existing surface blocks, so that the surface of each component is constructed as several surface blocks summarized from the corresponding triangles.

[0060] The first normal constraint is that the dot product of the normals of candidate triangles with a common edge is not less than a preset value, and the second normal constraint is that the dot product of the normals of the seed triangle and the candidate triangle is not less than a preset value.

[0061] The normal vector of a triangular face is a unit vector perpendicular to the plane containing the triangle. The dot product of two normal vectors yields a numerical value. When two normal vectors are orthogonal, their dot product is 0. When two normal vectors coincide, their dot product is maximized if their directions are the same, and minimized if their directions are opposite. Ignoring the case where the angle between the two normal vectors is obtuse, a smaller dot product indicates a larger angle between them, and thus a greater offset between them.

[0062] by Figure 3 Taking the surface segmentation shown as an example, the normal of the seed triangle A1 is... The normal of candidate triangle A2 is The normal of candidate triangle A3 is The normal of candidate triangle A4 is Without considering that the included angle between the two normals is obtuse, the inductibility of candidate triangles A2-A4 can be judged one by one according to their numbering order. For candidate triangle A2, since this candidate triangle only shares an edge with the seed triangle A1 and satisfies:

[0063] ;

[0064] in, Indicates the preset dot product threshold; This indicates that the dot product of the normal vectors of candidate triangle A2 and seed triangle A1 is not less than a preset value, therefore candidate triangle A2 satisfies the second normal constraint;

[0065] After summarizing the candidate triangle A2 and the seed triangle A1, for the candidate triangle A3 that shares a common edge with candidate triangle A2, the following condition is met:

[0066] ;

[0067] in, This indicates that the dot product of the normal vectors of candidate triangle A2 and candidate triangle A3 is not less than a preset value. This indicates that the dot product of the normal vectors of candidate triangle A3 and seed triangle A1 is not less than a preset value. Therefore, candidate triangle A3 satisfies both the first normal constraint and the second normal constraint.

[0068] After summarizing the candidate triangles A2, A3, and seed triangle A1, for candidate triangle A4 that shares a common edge with candidate triangle A2, the following condition is met:

[0069] , and / or ;

[0070] in, This indicates that the dot product of the normal vectors of candidate triangle A2 and candidate triangle A4 is less than a preset value. This indicates that the dot product of the normal vectors of candidate triangle A4 and seed triangle A1 is less than a preset value. Candidate triangle A4 does not satisfy either the first normal constraint or the second normal constraint, and therefore cannot be grouped into the same surface block as candidate triangle A2, candidate triangle A3 and seed triangle A1.

[0071] Based on the above technical solution, this embodiment uses the property of normal dot product to measure the degree of offset between different triangular faces, and establishes a first normal constraint and a second normal constraint to group the triangular faces with smaller offsets into the same surface block, thereby constructing the component surface into several precise surface blocks so as to locate the local mass gaps of the existing dot cloud to the precise area of ​​the component surface.

[0072] S2. Obtain the 3D point cloud corresponding to the building. The 3D point cloud includes several sampling points. Align the 3D point cloud with the IFC model in coordinates. Based on the minimum distance between each sampling point and each triangular face, match each sampling point with at most one surface block.

[0073] Specifically, for any sampling point, calculate the minimum distance between it and each triangular face, and select the surface block to which the triangular face corresponding to the minimum distance is classified as the nearest surface block of the sampling point. When the minimum distance corresponding to the sampling point does not exceed a preset value, match the sampling point with its at most one nearest surface block. Traverse the three-dimensional point cloud and perform the above steps for each sampling point to match each sampling point with at most one surface block.

[0074] Understandably, when the distance between a sampling point and more than one triangular facet is the minimum distance, if these triangular faces are grouped into the same nearest surface block, the sampling point is matched to that nearest surface block; if these triangular faces are grouped into different nearest surface blocks, the sampling point is matched to the nearest surface block with more triangular faces corresponding to the aforementioned minimum distance; if these triangular faces are grouped into different nearest surface blocks and each nearest surface block has the same number of triangular faces corresponding to the aforementioned minimum distance, then any nearest surface block is selected for matching, thereby establishing a reliable mapping relationship between the 3D point cloud and the IFC model.

[0075] S3. Calculate the geometric coverage and point cloud density of the 3D point cloud on each surface block and perform weighted normalization to obtain the gap score of each surface block, and select the surface block with the highest gap score as the target surface block.

[0076] Specifically, in any surface block, the geometric coverage of the 3D point cloud on that surface block is calculated based on the proportion of the observed portion of that surface block. For example, evaluation points are arranged on the surface block at preset intervals, and the calculation is based on the proportion of evaluation points actually observed by existing scanning stations. The point cloud density of the 3D point cloud on that surface block is calculated based on the ratio of the number of sampling points matched to the area of ​​that surface block. Thus, corresponding weights are set for the geometric coverage and point cloud density of the 3D point cloud on that surface block, and weighted normalization is performed to obtain the gap score of that surface block. The above steps are performed on each surface block through the IFC model to obtain the gap score of each surface block.

[0077] The weighted normalization formula for the gap score is:

[0078] ;

[0079] In the formula, In the IFC model, the first The number of gaps in each surface segment. , These represent the three-dimensional point cloud at the th... Geometric coverage and point cloud density on each surface patch This represents the preset desired density of a 3D point cloud for any surface segment. and All are weighted factors and .

[0080] Geometric coverage measures whether a building surface is adequately covered, while point cloud density measures whether the sampling points in the covered area of ​​the building surface are sufficiently dense. To give geometric coverage a relatively higher priority in defect assessment, we can set... To give sampling density a relatively higher priority in defect evaluation, one can set... In practical applications, In this way, for surface patches with obvious unscanned areas, their gap scores will be preferentially increased; for surface patches that have achieved some coverage but have insufficient point cloud density, the insufficient density will still contribute to the gap score. Preferably, , .

[0081] For the formula above Item, it is understandable, if If the point cloud density reaches the desired density, then the value of this term is 0. At this time, the gap fraction of the surface block is determined only by the geometric coverage of the 3D point cloud on the surface block.

[0082] S4. Divide the target surface into blocks to generate several candidate supplementary viewpoints. Calculate the value of each candidate supplementary viewpoint based on its view frustum coverage. Select the candidate supplementary viewpoint with the highest value as the target supplementary viewpoint.

[0083] Figure 4 This is a schematic diagram of a candidate supplementary scan viewpoint provided for an embodiment of this application. See also... Figure 4 Using the target surface block as the reference center for candidate viewpoint generation, a reference axis that is not parallel to the normal of the target surface block is first selected. When the angle between the target surface normal and the vertical direction is greater than a preset value, the vertical direction is used as the reference axis; otherwise, the global X-axis or Y-axis is selected as the reference axis.

[0084] In this local coordinate system, the normal of the target surface block is used as the initial candidate direction. Multiple candidate directions are obtained by applying different azimuth and elevation angle offsets. Then, the position of the candidate supplementary scan viewpoint is determined along each candidate direction according to one or more preset candidate distances. The orientation of each candidate supplementary scan viewpoint is pointed towards the center of the target surface block.

[0085] Specifically, for any candidate supplementary viewpoint, the value of the candidate supplementary viewpoint is calculated by summing the products of the notch fraction, area, and corresponding observation quality of each surface patch within the cone coverage area of ​​the candidate supplementary viewpoint; the observation quality corresponding to the surface patch is obtained based on the orthographic, distance quality, and resolution quality of the surface patch from the candidate supplementary viewpoint; the above steps are performed on each candidate supplementary viewpoint to calculate the value of each candidate supplementary viewpoint.

[0086] In this embodiment, the orthographic projection is calculated based on the normal of the surface patch and the unit direction vector pointing from the center of the surface patch to the candidate supplementary viewpoint. The more consistent the directions of the two, the closer the candidate supplementary viewpoint is to viewing the surface patch orthographically, and the higher the orthographic projection. For any surface patch that can be observed by the candidate supplementary viewpoint, the orthographic projection is first determined based on the degree of consistency between the normal of the surface patch and the observation direction pointing from the surface patch to the candidate supplementary viewpoint.

[0087] Distance quality is determined based on the degree of deviation between the distance between the candidate supplementary scan viewpoint and the surface patch relative to the preferred working distance of the scanning equipment. Distance quality is highest near the preferred working distance; it gradually decreases as the actual observation distance deviates from the preferred working distance.

[0088] Resolution quality is determined by the observation distance. As the distance between the candidate supplementary scan viewpoint and the surface patch increases, the effective spatial sampling capability of the scanning device for a unit surface decreases, thus resulting in a corresponding reduction in resolution quality.

[0089] Multiplying the aforementioned frontal view, distance quality, and resolution quality yields the observation quality of the candidate rescan viewpoint for the corresponding surface patch. Therefore, a poor performance in any of these factors will correspondingly reduce the overall observation quality of the candidate rescan viewpoint for that surface patch.

[0090] Figure 5 This is a schematic diagram of a candidate supplementary viewpoint provided in an embodiment of this application. See also: Figure 5 The rectangles marked B1-B4 are all surface blocks, and the shaded area is the cone coverage of the candidate supplementary viewpoint. Among them, the surface blocks that meet the conditions of positivity, field of view, and range are included in the value calculation of the candidate supplementary viewpoint.

[0091] The field of view is determined by the field of view parameters of the scanning equipment used. For a candidate supplementary scan viewpoint, the angle between the candidate supplementary scan viewpoint and the direction from the candidate supplementary scan viewpoint to the corresponding surface segment is determined; when the angle is within the field of view of the scanning equipment, the surface segment is considered to satisfy the field of view condition.

[0092] The range condition is determined based on the effective ranging range of the scanning equipment. The spatial distance between the candidate supplementary scan viewpoint and the corresponding surface block is calculated. When this distance is within the effective range of the scanning equipment, the range condition is considered to be satisfied.

[0093] Based on the above technical solution, this embodiment calculates the value of candidate supplementary scanning viewpoints by combining the notch fraction, area and corresponding observation quality of surface blocks, so that candidate supplementary scanning viewpoints can preferentially cover areas with high notches and good observation quality, thereby selecting supplementary scanning viewpoints more rationally and effectively improving the scanning quality of buildings.

[0094] See Figure 6 This application also discloses a viewpoint planning device for filling gaps in building point clouds, comprising:

[0095] The triangular construction module 601 is used to obtain the IFC model of the building. The IFC model includes several components. The surface of each component is triangularly divided to construct each component into several surface blocks composed of several triangular faces.

[0096] The point cloud mapping module 602 is used to acquire the three-dimensional point cloud corresponding to the building. The three-dimensional point cloud includes several sampling points. The three-dimensional point cloud is aligned with the IFC model in coordinates. Based on the minimum distance between each sampling point and each triangular face, each sampling point is matched with at most one surface block.

[0097] The notch targeting module 603 is used to calculate the geometric coverage and point cloud density of the 3D point cloud on each surface block and perform weighted normalization to obtain the notch score of each surface block, and select the surface block with the highest notch score as the target surface block.

[0098] The viewpoint selection module 604 is used to generate several candidate supplementary viewpoints by dividing the target surface into blocks, calculate the value of each candidate supplementary viewpoint based on the view cone coverage of each candidate supplementary viewpoint, and select the candidate supplementary viewpoint with the highest value as the target supplementary viewpoint.

[0099] It should be noted that the apparatus disclosed in the above embodiments is only illustrated by the division of the above functional modules when performing the building point cloud gap filling viewpoint planning method. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus disclosed in the above embodiments and the building point cloud gap filling viewpoint planning method embodiments belong to the same concept, and the implementation process is detailed in the method embodiments, which will not be repeated here.

[0100] See Figure 7 This application also discloses an electronic device, including: at least one processor 71, at least one memory 72, a power supply 73, a communication interface 74, an input / output interface 75, and a communication bus 76. The memory 72 stores a computer program, and when the processor 71 executes the computer program, it implements the relevant steps in the building point cloud gap filling viewpoint planning method disclosed in any of the foregoing embodiments. Furthermore, the electronic device 70 in this embodiment can specifically be an electronic computer.

[0101] In this embodiment, the power supply 73 is used to provide operating voltage for the various hardware devices on the electronic device 70; the communication interface 74 can create a data transmission channel between the electronic device 70 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 75 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0102] In addition, the memory 72, as a carrier for resource storage, can be a read-only memory, random access memory, disk, or optical disk, etc. The resources stored thereon can include an operating system 721, computer programs 722, etc., and the storage method can be temporary storage or permanent storage.

[0103] The operating system 721 is used to manage and control the various hardware devices on the electronic device 70 and the computer program 722, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the building point cloud gap filling viewpoint planning method executed by the electronic device 70 as disclosed in any of the foregoing embodiments, the computer program 722 may further include a computer program capable of performing other specific tasks.

[0104] Furthermore, this application also discloses a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the aforementioned building point cloud gap filling viewpoint planning method. Specific steps of this method can be found in the corresponding content of the foregoing embodiments, and will not be repeated here.

[0105] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The apparatuses disclosed in the embodiments are described simply because they correspond to the methods disclosed in the embodiments; relevant parts can be found in the method section.

[0106] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0107] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0108] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0109] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for planning viewpoints to fill gaps in building point clouds, characterized in that, include: Obtain the IFC model of the building, which includes several components; divide the surface of each component into triangles to construct each component into several surface blocks composed of several triangular faces; Obtain a 3D point cloud corresponding to the building, the 3D point cloud including a number of sampling points; align the 3D point cloud with the IFC model in coordinates, and match each sampling point with at most one surface block according to the minimum distance between each sampling point and each triangular face; The geometric coverage and point cloud density of the three-dimensional point cloud on each of the surface blocks are calculated and weighted normalized to obtain the gap score of each surface block, and the surface block with the highest gap score is selected as the target surface block. The target surface is divided into blocks to generate several candidate supplementary viewpoints. The value of each candidate supplementary viewpoint is calculated based on the view frustum coverage of each candidate supplementary viewpoint, and the candidate supplementary viewpoint with the highest value is selected as the target supplementary viewpoint.

2. The building point cloud gap filling viewpoint planning method according to claim 1, characterized in that, The step of triangulating the surface of each component to construct each component into several surface blocks composed of several triangular faces includes: The surface of each component is triangularly divided to construct a plurality of triangular faces on the surface of each component. Based on the adjacency and normal relationships between the triangular faces corresponding to each component, the triangular faces corresponding to each component are summarized respectively, so that the surface of each component is constructed as a number of surface blocks summarized from the corresponding triangular faces.

3. The building point cloud gap filling viewpoint planning method according to claim 2, characterized in that, The step of summarizing the triangular faces corresponding to each component based on their adjacency and normal relationships, and constructing the surface of each component into several surface blocks summarized from the corresponding triangular faces, includes: In any of the aforementioned components, a triangular face that is not included in an existing surface block is selected as a seed triangular face, and the remaining triangular faces that are not included in an existing surface block and have a common edge with the seed triangular face are selected as candidate triangular faces. Based on the normals of the seed triangle and the candidate triangle in the component, a first normal constraint and a second normal constraint are established. The seed triangle and the candidate triangles satisfying the first and second normal constraints are then summarized one by one to obtain corresponding surface blocks. The first normal constraint is that the dot product of the normals of the candidate triangles with a common edge is not less than a preset value. The second normal constraint is that the dot product of the normals of the seed triangle and the candidate triangle is not less than a preset value. The IFC model is traversed, and the above steps are repeated for each component until there are no triangular faces in any component that have not been included in the existing surface blocks, so that the surface of each component is constructed as a number of surface blocks that are included by the corresponding number of triangular faces.

4. The building point cloud gap filling viewpoint planning method according to claim 1, characterized in that, The step of matching each sampling point with at most one surface block based on the minimum distance between each sampling point and each triangular face includes: For any of the sampling points, calculate the minimum distance between it and each of the triangular faces, and select the surface block to which the triangular face corresponding to the minimum distance is classified as the nearest surface block of the sampling point. When the minimum distance corresponding to the sampling point does not exceed a preset value, match the sampling point with at most one of the nearest surface blocks corresponding to it. The three-dimensional point cloud is traversed, and the above steps are performed on each of the sampling points to match each of the sampling points with at most one of the surface blocks.

5. The building point cloud gap filling viewpoint planning method according to claim 1, characterized in that, The step of calculating the geometric coverage and point cloud density of the 3D point cloud on each of the surface blocks and performing weighted normalization to obtain the gap fraction of each surface block includes: In any of the surface segments, the geometric coverage of the 3D point cloud on the surface segment is calculated based on the proportion of the observed portion of the surface segment to the total surface segment. The point cloud density of the three-dimensional point cloud on the surface block is calculated based on the ratio of the number of sampling points matched by the surface block to the area of ​​the surface block. The geometric coverage and point cloud density of the three-dimensional point cloud on the surface block are assigned corresponding weights and then weighted and normalized to obtain the gap fraction of the surface block. The above steps are performed on each of the surface blocks in the IFC model to obtain the notch fraction of each surface block.

6. The building point cloud gap filling viewpoint planning method according to claim 5, characterized in that, The weighted normalization formula for the gap fraction is: ; In the formula, In the IFC model, the first... The number of notches in each of the aforementioned surface segments. , These respectively represent the three-dimensional point cloud at the th... Geometric coverage and point cloud density on each of the aforementioned surface segments This represents the preset desired density of the three-dimensional point cloud for any of the surface segments. and All are weighted factors and .

7. The building point cloud gap filling viewpoint planning method according to claim 1, characterized in that, The step of dividing the target surface into blocks to generate several candidate supplementary viewpoints, and calculating the value of each candidate supplementary viewpoint based on its view frustum coverage, includes: For any candidate supplementary viewpoint, the value of the candidate supplementary viewpoint is calculated based on the sum of the products of the notch fraction, area, and corresponding observation quality of each surface patch within the cone coverage area of ​​the candidate supplementary viewpoint; the observation quality corresponding to the surface patch is obtained based on the orthographic, distance quality, and resolution quality of the surface patch from the candidate supplementary viewpoint. The above steps are performed on each of the candidate supplementary viewpoints to calculate the value of each candidate supplementary viewpoint.

8. A viewpoint planning device for filling gaps in building point clouds, characterized in that, include: A triangular construction module is used to obtain the IFC model of a building, which includes several components; the surface of each component is triangularly divided to construct each component into several surface blocks composed of several triangular faces. The point cloud mapping module is used to acquire the three-dimensional point cloud corresponding to the building, the three-dimensional point cloud including a number of sampling points; the three-dimensional point cloud is aligned with the IFC model in coordinates, and each sampling point is matched with at most one surface block according to the minimum distance between each sampling point and each triangular face; The gap targeting module is used to calculate the geometric coverage and point cloud density of the three-dimensional point cloud on each of the surface blocks and perform weighted normalization to obtain the gap score of each surface block, and select the surface block with the highest gap score as the target surface block. The viewpoint selection module is used to generate several candidate supplementary viewpoints by dividing the target surface into blocks, calculate the value of each candidate supplementary viewpoint based on the view frustum coverage of each candidate supplementary viewpoint, and select the candidate supplementary viewpoint with the highest value as the target supplementary viewpoint.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the building point cloud gap filling viewpoint planning method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the building point cloud gap filling viewpoint planning method as described in any one of claims 1-7.