3D Model Construction Method, System and Storage Medium for Architectural Design

Abstract

The present invention relates to the technical field of model construction, and particularly relates to a three-dimensional model construction method, system and storage medium for architectural design. The method includes: removing abnormal point cloud data of an abnormal wall part in a point cloud cluster according to a first direction vector and a first Euclidean distance between adjacent point cloud data in the point cloud cluster, so as to obtain a target point cloud cluster belonging to a connection part of a building component; determining a regular area of a non-wall connected to a wall; determining the identity of the structures of the point cloud data on both sides of the wall within the target point cloud cluster; in the case where the identity is greater than or equal to a first threshold, determining that the building components formed by the point cloud data on both sides of the wall are the same building component; connecting the point cloud data belonging to the same building component on both sides of the wall to obtain point cloud data of a complete building component; constructing a three-dimensional model of the building component according to the point cloud data of the complete building component. The present invention improves the construction efficiency and accuracy of the three-dimensional model of the building component.

Description

Technical Field

[0001] The present invention relates to the technical field of model construction, and particularly to a three-dimensional model construction method, system and storage medium for architectural design. Background Art

[0002] Building components refer to the basic units or parts that make up a building, which can be structural, functional or decorative. Each component has its specific function and needs to be precisely defined and manufactured during the architectural design and construction process. The three-dimensional models of building components can visually display design concepts, improve design accuracy and efficiency, assist in construction simulation, optimize structural design, and perform performance analysis, etc. The three-dimensional models of building components are mainly constructed by independent design or by obtaining three-dimensional data after scanning existing buildings.

[0003] In some scenarios, usually, the point cloud data of an existing building is obtained by scanning it, and then hierarchical division is performed according to the obtained point cloud data to obtain the building components of the existing building, and a three-dimensional model for architectural design is constructed based on the point cloud data of the building components. Since the connection between different building components is tight and the surface substances are easy to fall off, it is difficult to find the division boundaries when performing hierarchical division of building components through the point cloud data obtained by scanning the building, and there are redundant and irrelevant detection data points in the obtained point cloud data, thus increasing the division difficulty of building components and reducing the division accuracy, and further resulting in low construction efficiency and accuracy of the three-dimensional models of building components. Summary of the Invention

[0004] In order to solve the technical problems of high division difficulty and low division accuracy of building components, which in turn lead to low construction efficiency and accuracy of the three-dimensional models of building components, the purpose of the present invention is to provide a three-dimensional model construction method, system and storage medium for architectural design.

[0005] The specific technical solutions adopted to solve the above technical problems are as follows:

[0006] An embodiment of the present invention provides a three-dimensional model construction method for architectural design, including: clustering the point cloud data of a building to obtain a plurality of point cloud clusters, removing the abnormal point cloud data of the abnormal wall part in the point cloud clusters according to the first direction vector and the first Euclidean distance between adjacent point cloud data in the point cloud clusters, so as to obtain the target point cloud clusters belonging to the connection parts of building components; determining the regular areas of non-walls connected to the wall according to the second Euclidean distance and the second direction vector between adjacent point cloud data in the target point cloud clusters; determining the identity of the structures of the point cloud data on both sides of the wall in the target point cloud clusters according to the point cloud data on the inner and outer sides of the regular areas on the same vector on both sides of the wall and the regular coefficients of the target point cloud clusters where the point cloud data on the inner and outer sides are located; when the identity is greater than or equal to the first threshold, determining that the building components formed by the point cloud data on both sides of the wall are the same building component; connecting the point cloud data on both sides of the wall belonging to the same building component to obtain the point cloud data of the complete building component; constructing a three-dimensional model of the building component according to the point cloud data of the complete building component.

[0007] Optionally, removing the abnormal point cloud data of the abnormal wall part in the point cloud clusters according to the first direction vector and the first Euclidean distance between adjacent point cloud data in the point cloud clusters, so as to obtain the target point cloud clusters belonging to the connection parts of building components includes: determining the point cloud density parameter of the point cloud cluster according to the first quantity of the point cloud data in the point cloud cluster, the spatial volume of the point cloud data, and the average value of the Euclidean distances between the point cloud data in the point cloud cluster; when the point cloud density parameter is greater than or equal to the second threshold, determining that the point cloud cluster is the initial target point cloud cluster belonging to the connection part of the building component; obtaining the first direction vector and the first Euclidean distance between adjacent point cloud data at the same height in the initial target point cloud cluster; determining the first dispersion parameter of the adjacent point cloud data at the same height according to the first Euclidean distance between the adjacent point cloud data and the first included angle parameter between the adjacent first direction vectors; determining the regular coefficient of the initial target point cloud cluster according to the first dispersion parameter of the adjacent point cloud data at each height in the initial target point cloud cluster, the maximum height value in the initial target point cloud cluster, and the point cloud density parameter of the initial target point cloud cluster; when the regular coefficient of the initial target point cloud cluster is greater than or equal to the third threshold, determining that the initial target point cloud cluster is the target point cloud cluster belonging to the connection part of the building component, and when the regular coefficient of the initial target point cloud cluster is less than the third threshold, determining that the initial target point cloud cluster belongs to the abnormal part of the building wall.

[0008] Optionally, determining the point cloud density parameter of the point cloud cluster according to the first quantity of the point cloud data in the point cloud cluster, the spatial volume of the point cloud data, and the average value of the Euclidean distances between the point cloud data in the point cloud cluster includes: calculating a first ratio between the first quantity and the spatial volume, and the reciprocal of the average value of the Euclidean distances; calculating a first product between the first ratio and the reciprocal; performing a normalization process on the first product to obtain the point cloud density parameter of the point cloud cluster.

[0009] Optionally, determining the first dispersion parameter of adjacent point cloud data at the same height according to the first Euclidean distance between adjacent point cloud data and the first angle parameter between adjacent first direction vectors includes: calculating the absolute value of a first difference between the first Euclidean distances between adjacent point cloud data, and a second ratio between the first angle parameters between adjacent first direction vectors; calculating a second product between the absolute value of the first difference and the second ratio, and superimposing the second products to obtain a first superimposed value; calculating a second difference between the number of point cloud data in the initial target point cloud cluster and a preset value; determining a third ratio between the first superimposed value and the second difference as the first dispersion parameter of adjacent point cloud data at the same height.

[0010] Optionally, determining the rule coefficient of the initial target point cloud cluster according to the first dispersion parameter of adjacent point cloud data at each height in the initial target point cloud cluster, the maximum height value in the initial target point cloud cluster, and the point cloud density parameter of the initial target point cloud cluster includes: superimposing the first dispersion parameters of adjacent point cloud data at each height in the initial target point cloud cluster to obtain a second superimposed value; calculating a fourth ratio between the second superimposed value and the maximum height value, and a third product between the fourth ratio and the point cloud density parameter of the initial target point cloud cluster; performing a normalization process on the third product using the derivative function of the sigmoid function to obtain the rule coefficient of the initial target point cloud cluster.

[0011] Optionally, determining a regular area of non-wall components connected to a wall according to the second Euclidean distance and the second direction vector between adjacent point cloud data in a target point cloud cluster includes: obtaining the second Euclidean distance and the second direction vector between adjacent point cloud data in the target point cloud cluster; determining whether the plane where the adjacent point cloud data is located is a wall plane by using the second Euclidean distance and the second direction vector; in the case of a wall plane, determining a third direction vector in the wall plane according to the adjacent point cloud data of the wall plane; determining that the point cloud data corresponding to the second direction vector is a non-wall component connected to the wall according to the second direction vector between adjacent point cloud data in the target point cloud cluster and the fourth direction vector of adjacent point cloud data of a target building component; determining a second dispersion parameter of adjacent point cloud data at the same height according to a second included angle parameter between a third Euclidean distance and a fifth direction vector between adjacent point cloud data of the non-wall component; in the case where the second dispersion parameter is a predetermined value, determining that the shape formed by the adjacent point cloud data is a regular shape, and the area corresponding to the regular shape is a regular area; extending from the left side of the point cloud data at the contact part between the regular shape and the wall along the normal vector of the third direction vector to determine whether there is non-wall point cloud data outside the wall; in the case where there is non-wall point cloud data, determining a third dispersion parameter of the non-wall point cloud data according to a fourth Euclidean distance and a sixth direction vector between adjacent point cloud data of the non-wall point cloud data; in the case where the third dispersion parameter is equal to the predetermined value, determining that the shape formed by the adjacent point cloud data of the non-wall point cloud data is a regular shape, and the area corresponding to the regular shape is a regular area.

[0012] Optionally, determining the identity of the point cloud data structures on both sides of a wall within a target point cloud cluster according to the point cloud data on the inner and outer sides of a regular area on the same vector on both sides of the wall and the regular coefficient of the target point cloud cluster where the point cloud data on the inner and outer sides is located includes: from the regular areas on the same vector on both sides of the wall, respectively taking the point cloud data on the inner and outer sides of the wall as a first starting point and a second starting point, obtaining a fifth Euclidean distance and a seventh direction vector between the point cloud data adjacent to the first starting point, and obtaining a sixth Euclidean distance and an eighth direction vector between the point cloud data adjacent to the second starting point; determining the identity of the point cloud structures on both sides of the wall within the target point cloud cluster where the first starting point and the second starting point are located according to the fifth Euclidean distance, the seventh direction vector, the sixth Euclidean distance, the eighth direction vector, and the regular coefficient of the target point cloud cluster where the first starting point and the second starting point are located.

[0013] Optionally, determining the identity of the point cloud structures on both sides of the wall within the target point cloud cluster where the first starting point and the second starting point are located, based on the fifth Euclidean distance, the seventh direction vector, the sixth Euclidean distance, the eighth direction vector, and the rule coefficient of the target point cloud cluster where the first starting point and the second starting point are located, includes: determining the distance difference parameter between the first starting point and the second starting point according to the fifth Euclidean distance and the sixth Euclidean distance; determining the direction difference parameter between the first starting point and the second starting point according to the seventh direction vector and the eighth direction vector; determining the identity of the point cloud structures on both sides of the wall within the target point cloud cluster where the first starting point and the second starting point are located according to the distance difference parameter, the direction difference parameter, and the rule coefficient of the target point cloud cluster where the first starting point and the second starting point are located.

[0014] In a second aspect, an embodiment of the present invention provides a three-dimensional model construction system for architectural design, including: a processor and a memory; wherein, the memory is used to store a computer program that can run on the processor; the processor is used to execute the program stored on the memory to implement the steps of the three-dimensional model construction method for architectural design as mentioned in the first aspect.

[0015] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, where the computer-readable medium stores one or more programs, and when the one or more programs are executed by an electronic device including a plurality of application programs, the electronic device is caused to execute the steps of the three-dimensional model construction method for architectural design as mentioned in the first aspect.

[0016] The present invention has the following beneficial effects: First, cluster the point cloud data of the building to obtain a plurality of point cloud clusters, remove the abnormal point cloud data of the abnormal wall part in the point cloud cluster according to the first direction vector and the first Euclidean distance between adjacent point cloud data in the point cloud cluster, and obtain the target point cloud cluster belonging to the connection part of the building components; and determine the regular area of the non-wall connected to the wall according to the second Euclidean distance and the second direction vector between adjacent point cloud data in the target point cloud cluster; then determine the identity of the structure of the point cloud data on both sides of the wall within the target point cloud cluster according to the point cloud data on the inner and outer sides of the regular area on the same vector on both sides of the wall and the rule coefficient of the target point cloud cluster where the point cloud data on the inner and outer sides are located; in the case where the identity is greater than or equal to the first threshold, determine that the building components formed by the point cloud data on both sides of the wall are the same building components; secondly, connect the point cloud data on both sides of the wall belonging to the same building component to obtain the point cloud data of the complete building component; finally, construct a three-dimensional model of the building component according to the point cloud data of the complete building component.

[0017] Thus, in the embodiments of the present invention, by clustering the point cloud data and removing the abnormal point cloud data according to the direction vectors and Euclidean distances of the point cloud data in the clustering clusters, the redundant and irrelevant detection data points in the point cloud data are eliminated, thereby reducing the difficulty of dividing the building components and improving the division accuracy. Further, based on the Euclidean distances and direction vectors between the normal point cloud data after removing the abnormal point cloud data, the connection parts of the building components are further distinguished to determine the regular areas of non-walls connected to the walls, so as to accurately find the division boundaries and accurately divide the building components. Finally, based on the complete point cloud data of the divided building components, modeling is performed, which improves the construction efficiency and accuracy of the three-dimensional model of the building components. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 Flowchart of a three-dimensional model construction method for building design provided by an embodiment of the present invention;

[0020] Figure 2 Schematic diagram of the distribution of detection stations outside a building provided by an embodiment of the present invention;

[0021] Figure 3 Schematic diagram of the scanning route of the point cloud data inside a building provided by an embodiment of the present invention;

[0022] Figure 4 Schematic diagram of the point cloud data of the normal area and abnormal area of a wall provided by an embodiment of the present invention;

[0023] Figure 5 Schematic diagram of the point cloud data of the complete building structure of a building component provided by an embodiment of the present invention;

[0024] Figure 6 Schematic diagram of the structure of a three-dimensional model construction system for building design provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features, and effects of a three-dimensional model construction method, system, and storage medium for architectural design proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0027] The following specifically describes the specific solution of a three-dimensional model construction method for architectural design provided by the present invention in conjunction with the accompanying drawings.

[0028] Embodiment 1:

[0029] Please refer to Figure 1 , which shows a flowchart of a three-dimensional model construction method for architectural design provided by an embodiment of the present invention, including:

[0030] Step S101: Cluster the point cloud data of the building to obtain multiple point cloud clusters, and remove the abnormal point cloud data of the abnormal wall part in the point cloud clusters according to the first direction vector and the first Euclidean distance between adjacent point cloud data in the point cloud clusters, so as to obtain the target point cloud clusters belonging to the joints of building components.

[0031] Specifically, for a building, it includes the interior and the exterior. Since the structures of the interior and exterior of the building are different, different methods are adopted in the embodiments of the present invention to obtain the point cloud data of the interior and exterior of the building. Among them, a handheld three-dimensional laser scanner is used to scan the interior of the building to obtain the three-dimensional data information of the interior of the building. A stationary laser scanner is used to scan the exterior of the building to obtain the three-dimensional data information of the exterior of the building. For the exterior, a stationary laser scanner is used, and the corner inflection points of the exterior of the building are used as the basic detection stations. At the same time, a rectangular array is formed as much as possible between the basic detection stations, and the remaining stations are ordinary detection stations. The ordinary detection stations are evenly distributed between adjacent basic detection stations, with an interval distance of 15 m and a maximum interval distance not exceeding 18 m. For the interior of the building, a handheld scanner is used, and the movement route inside the building needs to be planned in advance, and the detection route is determined in advance to avoid repeated routes as much as possible. At the same time, the scanner is 0.8 m away from the wall. Exemplarily, as shown in Figure 2 and Figure 3 shown, Figure 2 is a schematic diagram of the distribution of detection stations for the exterior of a building provided by an embodiment of the present invention, Figure 3A schematic diagram of the scanning route of the point cloud data inside a building provided by an embodiment of the present invention.

[0032] Figure 2 Among them, the corner inflection points outside the building are used as the basic detection stations. At the same time, a rectangular array is formed as much as possible between the basic detection stations, and the remaining stations are ordinary detection stations, and the ordinary detection stations are evenly distributed between adjacent basic detection stations.

[0033] Furthermore, for the three-dimensional data information obtained by scanning the inside and outside of the building, the TRW software is used to preprocess the three-dimensional data information, including operations such as point cloud data extraction, coordinate system conversion, and point cloud denoising. Among them, for the coordinate system conversion, the straight line where the adjacent basic detection stations of the external station type are located can be used as the coordinate axis side. Taking the height h of the building as the height coordinate value of the data points in the point cloud data, and then obtaining the point cloud data of the building. During the three-dimensional scanning of the building, there are differences in the point cloud data scanned in different parts. For example, the point cloud data of the wall is less and the distribution between the data points is relatively uniform, while the point cloud data obtained at the artistic part or the connection of different components is more and the distribution between the data points is relatively complex, and the part with more in the building is the building wall. Therefore, in order to better achieve the hierarchical division of building components, it is necessary to remove the wall areas with low-density point clouds.

[0034] Furthermore, the embodiment of the present invention first uses the K-Means Clustering Algorithm (K-means) to perform clustering operations on the point cloud data of the building, uses the elbow method to obtain the number of cluster centers required for the point cloud data of the building, and then starts to traverse the point cloud data until the iteration stops, forming the final point cloud cluster set. Subsequently, the spatial volume of each point cloud cluster is obtained through the circumscribed rectangle, and the first quantity of the point cloud data in each point cloud cluster is counted. Furthermore, the point cloud density parameters of each point cloud cluster are obtained.

[0035] Further, as an optional embodiment of the present invention, removing the abnormal point cloud data of the abnormal wall part in the point cloud cluster according to the first direction vector and the first Euclidean distance between adjacent point cloud data in the point cloud cluster, and obtaining the target point cloud cluster belonging to the connection part of the building component includes: determining the point cloud density parameter of the point cloud cluster according to the first quantity of the point cloud data in the point cloud cluster, the spatial volume of the point cloud data, and the average value of the Euclidean distances between the point cloud data in the point cloud cluster; when the point cloud density parameter is greater than or equal to the second threshold, determining that the point cloud cluster is the initial target point cloud cluster belonging to the connection part of the building component; obtaining the first direction vector and the first Euclidean distance between adjacent point cloud data at the same height in the initial target point cloud cluster; determining the first dispersion parameter of the adjacent point cloud data at the same height according to the first Euclidean distance between the adjacent point cloud data and the first included angle parameter between the adjacent first direction vectors; determining the rule coefficient of the initial target point cloud cluster according to the first dispersion parameter of the adjacent point cloud data at each height in the initial target point cloud cluster, the maximum height value in the initial target point cloud cluster, and the point cloud density parameter of the initial target point cloud cluster; when the rule coefficient of the initial target point cloud cluster is greater than or equal to the third threshold, determining that the initial target point cloud cluster is the target point cloud cluster belonging to the connection part of the building component, and when the rule coefficient of the initial target point cloud cluster is less than the third threshold, determining that the initial target point cloud cluster belongs to the abnormal part of the building wall.

[0036] Specifically, when determining the point cloud density parameter of the point cloud cluster, first calculate the first ratio between the first quantity and the spatial volume, and the reciprocal of the average value of the Euclidean distances; then calculate the first product between the first ratio and the reciprocal; finally, perform a normalization process on the first product to obtain the point cloud density parameter of the point cloud cluster.

[0037] Among them, the embodiment of the present invention specifically uses the following formula to calculate the point cloud density parameter of the point cloud cluster:

[0038]

[0039] In the above formula, P is the point cloud density parameter of the point cloud cluster. M is the first quantity of the data points in the point cloud cluster. V is the spatial volume of the point cloud cluster. It is the ratio of the number of data points in the point cloud cluster to the occupied space, representing the density of the data points. The larger this value is, the greater the density of the data points and the higher the complexity of the building, and vice versa. is the average value of the Euclidean distances between the point cloud data in the point cloud cluster. norm represents the normalization function, which is used to perform a normalization process on .

[0040] Further, in the embodiment of the present invention, all point cloud clusters are processed by the above method to obtain the point cloud density parameter of each point cloud cluster. By setting a second threshold, when the point cloud density parameter of the point cloud cluster is greater than or equal to the second threshold, the point cloud cluster may be the connection part of the building component, and the point cloud cluster is marked; otherwise, it is not marked. Among them, the second threshold can be set according to the actual situation, and in the embodiment of the present invention, the value is 0.5.

[0041] Further, over time, a paint layer will peel off or flake on the surface layer of the building wall, which will record the three-dimensional information data of the abnormal wall area formed by the peeling or flaking during three-dimensional scanning, resulting in a large number of point cloud data and high density on the wall surface, so that the initial point cloud cluster marked contains abnormal wall parts that are not the connection parts of the building components. Exemplarily, as Figure 4 shown, Figure 4 is a schematic diagram of the point cloud data of the normal area and the abnormal area of the wall provided by an embodiment of the present invention. Figure 4 In it, for the high-density point cloud cluster formed by the abnormal area of the wall, the similarity between adjacent point cloud data is low, the cracks formed are irregular, and the degree of dispersion is large. For the high-density point cloud cluster designed or at the building connection, such as when there are doors and windows in the wall, the data point line formed between the wall and the doors and windows is smooth. Therefore, by analyzing the degree of regularity between high-density point cloud data points, if there is a regular change, it is the connection part of the building; otherwise, it is an abnormal phenomenon generated by objects such as walls.

[0042] Further, the wall stands vertically on the ground, and the abnormal area of the wall takes the wall as the abnormal foundation. Therefore, in the embodiment of the present invention, the change relationship between the point cloud data at the same height h is used to analyze the rule coefficient of the point cloud cluster. First, for the point cloud data at the same height, the Euclidean distance and the direction vector between adjacent point cloud data are obtained. Then, according to the direction vector, the included angle parameter between adjacent direction vectors is obtained.

[0043] Specifically, in the embodiment of the present invention, the following formula is specifically used to calculate the included angle parameter between adjacent direction vectors:

[0044]

[0045] In the above formula, θ is the first included angle parameter between adjacent direction vectors. i is the abscissa number of the point cloud data. represents the direction vector between the point cloud data i and its adjacent point cloud data. represents the direction vector between the adjacent point cloud data i + 1 of the point cloud data i and its adjacent point cloud data. The arccos() function is used to obtain the included angle degree between adjacent direction vectors.

[0046] Further, if the Euclidean distances between adjacent point cloud data are close and the direction vectors are similar, then this area may be a building connection. Conversely, if the Euclidean distances between adjacent point cloud data vary greatly and the direction vectors are significantly different, then this area is a wall anomaly area.

[0047] Further, as an optional embodiment of the present invention, determining the first dispersion parameter of adjacent point cloud data at the same height according to the first Euclidean distance between adjacent point cloud data and the first included angle parameter between adjacent first direction vectors includes: calculating the absolute value of the first difference between the first Euclidean distances between adjacent point cloud data, and the second ratio between the first included angle parameters between adjacent first direction vectors; calculating the second product of the absolute value of the first difference and the second ratio, and superimposing the second products to obtain a first superimposed value; calculating the second difference between the number of point cloud data in the initial target point cloud cluster and a preset value; determining the third ratio between the first superimposed value and the second difference as the first dispersion parameter of adjacent point cloud data at the same height.

[0048] Specifically, the embodiment of the present invention specifically uses the following formula to calculate the first dispersion parameter of adjacent point cloud data at the same height:

[0049]

[0050] In the above formula, F h is the first dispersion parameter of adjacent point cloud data at height h. h is the height coordinate of the point cloud cluster. n is the number of point cloud data in the point cloud cluster, and its value is a natural number greater than or equal to 2. i is the horizontal coordinate number of the point cloud data, Δd i is the first difference between the first Euclidean distances between the point cloud data adjacent to the i-th point cloud data. Δd i+1 is the first difference between the first Euclidean distances between the point cloud data adjacent to the (i + 1)-th point cloud data adjacent to the i-th point cloud data.

[0051] |Δd i -Δd i+1 | is the distance difference coefficient between adjacent point cloud data. θ i is the first included angle parameter between adjacent first direction vectors between the point cloud data adjacent to the i-th point cloud data. θ i+1 is the first included angle parameter between adjacent first direction vectors between the point cloud data adjacent to the (i + 1)-th point cloud data adjacent to the i-th point cloud data, and is replaced by 0.1 when the value is 0 to avoid a denominator of 0. is the included angle difference coefficient between adjacent point cloud data. It is the product of the distance difference coefficient and the angle difference coefficient between adjacent point cloud data, representing the relationship parameter of the point cloud data at the same height. It is the mean value of the relationship parameters of the point cloud data at the same height, representing the dispersion parameter. The larger the value, the greater the degree of dispersion, and vice versa.

[0052] Furthermore, use the above embodiments of the present invention to process the point cloud data at the remaining heights within the point cloud cluster, obtain the dispersion parameters between the point cloud data, and then obtain the regular coefficient of the entire cluster according to the dispersion parameters at each height. Among them, as an optional embodiment of the present invention, determining the regular coefficient of the initial target point cloud cluster according to the first dispersion parameter of the adjacent point cloud data at each height in the initial target point cloud cluster, the maximum height value in the initial target point cloud cluster, and the point cloud density parameter of the initial target point cloud cluster includes: superimposing the first dispersion parameters of the adjacent point cloud data at each height in the initial target point cloud cluster to obtain a second superimposed value; calculating the fourth ratio between the second superimposed value and the maximum height value, and the third product between the fourth ratio and the point cloud density parameter of the initial target point cloud cluster; using the derivative function of the sigmoid function to normalize the third product to obtain the regular coefficient of the initial target point cloud cluster.

[0053] Specifically, the embodiments of the present invention specifically use the following formula to calculate the regular coefficient of the initial target point cloud cluster:

[0054]

[0055] In the above formula, R k is the regular coefficient of the initial target point cloud cluster k. k is the number of the point cloud cluster, and h is the height coordinate of the point cloud cluster. is the minimum height value of the initial target point cloud cluster k. is the maximum height value of the initial target point cloud cluster k. F h is the first dispersion parameter of the adjacent point cloud data at height h. is the mean value of the first dispersion parameters of the whole initial target point cloud cluster k. The larger this value, the more irregular the cluster, and vice versa. P k is the point cloud density parameter of the initial target point cloud cluster k. Use the derivative function of the sigmoid function to perform normalization processing, so that the larger the mean value of the first dispersion parameter, the larger the regular coefficient R k value, the more regular the cluster, and vice versa.

[0056] Further, traverse all the point cloud clusters using the above embodiments of the present invention to obtain the rule coefficients of each point cloud cluster. The embodiments of the present invention set a third threshold. When the rule coefficient is greater than or equal to the third threshold, the initial target point cloud cluster is the connection of the building; otherwise, it is the abnormal wall area. Among them, the third threshold in the embodiments of the present invention can be valued according to the actual situation, and the value in the embodiments of the present invention is 0.4.

[0057] Step S102: Determine the regular area of the non-wall connected to the wall according to the second Euclidean distance and the second direction vector between adjacent point cloud data in the target point cloud cluster.

[0058] Specifically, for the target point cloud cluster at the connection of the selected building, the structure formed between the point cloud data within the target point cloud cluster is sticky. For example, there are obvious angular areas between the point cloud data formed at the connection of the wall and the window, and the building surface is smooth within the adjacent angular areas, or the surface is smooth once and the slope of the curve change is similar the other time within the angular area, that is, the building at the connection of the building has a regular shape. Therefore, to better achieve the division of the building at the connection of the building, it is necessary to analyze according to the curve curvature parameters formed by the points within the target point cloud cluster, and divide the curves with similar curvatures into the same building component.

[0059] Further, as an alternative embodiment of the present invention, determining a regular area of a non-wall connected to a wall according to the second Euclidean distance and the second direction vector between adjacent point cloud data in a target point cloud cluster includes: obtaining the second Euclidean distance and the second direction vector between adjacent point cloud data in the target point cloud cluster; using the second Euclidean distance and the second direction vector to determine whether the plane where the adjacent point cloud data is located is a wall plane; in the case of being a wall plane, determining a third direction vector in the wall plane according to the adjacent point cloud data of the wall plane; determining that the point cloud data corresponding to the second direction vector is a non-wall component connected to the wall according to the second direction vector between the adjacent point cloud data in the target point cloud cluster and the fourth direction vector of the adjacent point cloud data of the target building component; determining a second dispersion parameter of the adjacent point cloud data at the same height according to the second included angle parameter between the third Euclidean distance and the fifth direction vector between the adjacent point cloud data of the non-wall component; in the case where the second dispersion parameter is a predetermined value, determining that the shape formed by the adjacent point cloud data is a regular shape, and the area corresponding to the regular shape is a regular area; extending from the left side of the point cloud data at the contact part between the regular shape and the wall along the normal vector of the third direction vector to determine whether there is non-wall point cloud data outside the wall; in the case where there is non-wall point cloud data, determining a third dispersion parameter of the non-wall point cloud data according to the fourth Euclidean distance and the sixth direction vector between the adjacent point cloud data of the non-wall point cloud data; in the case where the third dispersion parameter is equal to the predetermined value, determining that the shape formed by the adjacent point cloud data of the non-wall point cloud data is a regular shape, and the area corresponding to the regular shape is a regular area.

[0060] Specifically, the connection between all building components is based on the support of the wall. Therefore, by obtaining the second Euclidean distance D and the second direction vector A between adjacent point cloud data in the target point cloud cluster, the variance S of the Euclidean distance between adjacent point cloud data in the target point cloud cluster is further determined. D and the variance S of the direction vector A . If the variance S of the Euclidean distance between adjacent point cloud data D and the variance S of the direction vector A are close to 0, it indicates that the plane where the adjacent point cloud data is located is a wall plane, otherwise it is not. Further, for the plane where the point cloud data of the wall is located, according to the height change between the height values of the point cloud data, the third direction vector A' of the wall is obtained. Specifically, after obtaining the point cloud data of the wall, then for the position relationship of the adjacent point cloud data on the h-axis, from low to high is the third direction vector of the wall.

[0061] Further, due to the angular relationship between the wall and the connected target building component, such as the angle between the wall and the window frame. Therefore, obtain the direction vector of the adjacent point cloud data within the target point cloud cluster that forms an angle 180>θ′>0 with the window frame Then the structure where the adjacent point cloud data is located may be a building component connected to the wall, that is, a non-wall component connected to the wall.

[0062] Further, extend the adjacent point cloud data of the non-wall component connected to the wall uniformly along the horizontal axis i plane to obtain the difference value Δd of the third Euclidean distance between the adjacent point cloud data of the non-wall component j , and the second included angle parameter θ between the fifth direction vectors j , and substitute them into the calculation formula of the first dispersion parameter to obtain the second dispersion parameter F of the adjacent point cloud data at the same height of the non-wall component i . Among them, the predetermined value can be taken according to the actual situation. In the embodiment of the present invention, the value is 0. If F i =0, then the curve formed between the adjacent point cloud data is smooth, and it is a regular shape. Otherwise, it is irregular. Subsequently, mark the regular shape.

[0063] Further, due to the obstruction of the wall, it will cause a disconnection phenomenon when performing 3D scanning on the inside and outside of the building that belongs to the same building. Therefore, when constructing a 3D model of the building, it is necessary to combine the identity between the inside and outside of the building on both sides of the wall to achieve building integration and improve the integrity of the building.

[0064] Further, in the embodiment of the present invention, according to the left side of the point cloud data of the contact part between the known regular shape and the wall, extend along the normal vector N of the third direction vector A′ of the wall to determine whether there are non-wall data points outside the wall. If so, proceed to the next step. If not, discard. For the building component with non-wall data points outside the wall, based on the fourth Euclidean distance and the sixth direction vector between the adjacent point cloud data, obtain the third dispersion parameter F i ′ of the non-wall. If F i ′=0, then the curve formed between the adjacent point cloud data is smooth, and it is a regular shape. Otherwise, it is irregular.

[0065] Step S103, determine the identity of the structure of the point cloud data on both sides of the wall within the target point cloud cluster according to the point cloud data on the inner and outer sides of the regular area on the same vector on both sides of the wall and the regular coefficient of the target point cloud cluster where the point cloud data on the inner and outer sides is located.

[0066] Specifically, due to the obstruction of the wall, there will be a disconnection phenomenon when performing 3D scanning on the interior and exterior of the same building. Therefore, when constructing a 3D model of a building, it is necessary to combine the identity between the two sides of the wall to achieve building integration and improve the integrity of the building.

[0067] Further, as an optional embodiment of the present invention, determining the identity of the point cloud data structures on both sides of the wall within the target point cloud cluster according to the point cloud data on the inner and outer sides of the regular area on the same vector on both sides of the wall and the regular coefficients of the target point cloud clusters where the point cloud data on the inner and outer sides are located includes: from the regular area on the same vector on both sides of the wall, taking the point cloud data on the inner and outer sides of the wall as the first starting point and the second starting point respectively, obtaining the fifth Euclidean distance and the seventh direction vector between the point cloud data adjacent to the first starting point, and obtaining the sixth Euclidean distance and the eighth direction vector between the point cloud data adjacent to the second starting point; determining the identity of the point cloud structures on both sides of the wall within the target point cloud cluster where the first starting point and the second starting point are located according to the fifth Euclidean distance, the seventh direction vector, the sixth Euclidean distance, the eighth direction vector, and the regular coefficients of the target point cloud clusters where the first starting point and the second starting point are located.

[0068] Specifically, for the regular area on the same vector on both sides of the wall, taking the point cloud data m and m' on the inner and outer sides of the wall as the starting point O respectively, obtaining the fifth Euclidean distance d between the starting point m and the point cloud data adjacent to it O,m and the seventh direction vector the sixth Euclidean distance d between the starting point m' and the point cloud data adjacent to it O,m ' and the eighth direction vector

[0069] Further, as an optional embodiment of the present invention, determining the identity of the point cloud structures on both sides of the wall within the target point cloud cluster where the first starting point and the second starting point are located according to the fifth Euclidean distance, the seventh direction vector, the sixth Euclidean distance, the eighth direction vector, and the regular coefficients of the target point cloud clusters where the first starting point and the second starting point are located includes: determining the distance difference parameter between the first starting point and the second starting point according to the fifth Euclidean distance and the sixth Euclidean distance; determining the direction difference parameter between the first starting point and the second starting point according to the seventh direction vector and the eighth direction vector; determining the identity of the point cloud structures on both sides of the wall within the target point cloud cluster where the first starting point and the second starting point are located according to the distance difference parameter, the direction difference parameter, and the regular coefficients of the target point cloud clusters where the first starting point and the second starting point are located.

[0070] Specifically, the embodiments of the present invention obtain the distance difference parameter between the first starting point and the second starting point through the following formula:

[0071]

[0072] In the above formula, δ d represents the distance difference parameter between the first starting point and the second starting point. The larger the value of δ d , the greater the difference. Conversely, the smaller the value. d O,m represents the fifth Euclidean distance between the starting point m and its adjacent point cloud data. d O,m ′ represents the sixth Euclidean distance between the starting point m′ and its adjacent point cloud data.

[0073] Similarly, in the embodiment of the present invention, the direction difference parameter between the first starting point and the second starting point is determined by the following formula:

[0074]

[0075] In the above formula, represents the direction difference parameter between the first starting point and the second starting point. represents the seventh direction vector between the starting point m and its adjacent point cloud data. represents the eighth direction vector between the starting point m′ and its adjacent point cloud data.

[0076] Further, by comparing the distance difference parameter and the direction difference parameter between the point cloud data on both sides of the wall, if the difference values of the distance difference parameter and the direction difference parameter are closer, the identity is higher. Conversely, the identity is lower.

[0077] Further, in the embodiment of the present invention, the following formula is specifically used to calculate the identity of the point cloud structures on both sides of the wall within the point cloud cluster where the first starting point and the second starting point are located:

[0078]

[0079] In the above formula, G k is the identity of the point cloud structures on both sides of the wall within the target point cloud cluster k. δ d represents the distance difference parameter between the first starting point and the second starting point on both sides of the wall. is the direction difference parameter between the first starting point and the second starting point on both sides of the wall. is its difference value. The larger this value is, the lower the identity. Conversely, the higher the identity. And when this value reaches 0, it is replaced with 0.1 to avoid the denominator being 0; R k is the rule coefficient of the target point cloud cluster k, which is used as the cluster screening coefficient here. The tanh function is used to perform normalization processing. The tanh function is the hyperbolic tangent function.

[0080] Step S104, when the identity is greater than or equal to the first threshold, determine that the building components formed by the point cloud data on both sides of the wall are the same building component.

[0081] Specifically, in the embodiment of the present invention, the first threshold can be set according to the actual situation, and the value is 0.8 in the embodiment of the present invention. When G k ≥0.8, the data points on both sides of the wall form the same building component, otherwise they are not.

[0082] Step S105, connect the point cloud data belonging to the same building component on both sides of the wall to obtain the point cloud data of the complete building component, and construct a three-dimensional model of the building component according to the point cloud data of the complete building component.

[0083] Specifically, for the point cloud data belonging to the same building component after division, connect the point cloud data belonging to the same building component on both sides of the wall to obtain the point cloud data of the complete building structure of the building component. Exemplarily, as Figure 5 shown, Figure 5 is a schematic diagram of the point cloud data of the complete building structure of a building component provided by an embodiment of the present invention. Figure 5 In this figure, the point cloud data of the indoor part and the outdoor part on both sides of the wall are connected through the connecting part to obtain the point cloud data of the complete building component. Subsequently, the model construction system constructs a three-dimensional model in the model construction system according to the coordinate information of the point cloud data.

[0084] It should be noted that for the abnormal part of the wall, through the vector relationship of the adjacent point cloud data of the wall, the point cloud data is arranged along the vector direction according to the distance between the point cloud data, and at the same time, the point cloud data that is not the wall is removed to obtain the complete wall data points.

[0085] Furthermore, during the data storage process, the point cloud data points of the building components that are in the same vector and have the same distance between the point cloud data are compressed by run-length encoding, and the remaining point cloud data is stored according to the original data. The three-dimensional model construction system constructs a three-dimensional solid model of the building component in the three-dimensional space according to the left side of the point cloud data of each building after layering, which is convenient for the system to use during building design.

[0086] In the embodiment of the present invention, by clustering the point cloud data and removing the abnormal point cloud data according to the direction vector and Euclidean distance of the point cloud data in the clustering cluster, the redundant and irrelevant detection data points in the point cloud data are removed, thereby reducing the difficulty of dividing the building components and improving the division accuracy. Further, based on the Euclidean distance and direction vector between the normal point cloud data after removing the abnormal point cloud data, the connection part of the building components is further distinguished, and the regular area of the non-wall connected to the wall is determined, so as to accurately find the division boundary and accurately divide the building components. Finally, based on the complete point cloud data of the divided building components, modeling is carried out, which improves the construction efficiency and accuracy of the three-dimensional model of the building components.

[0087] Embodiment 2:

[0088] Corresponding to the three-dimensional model construction method for building design provided in the above embodiment, based on the same technical concept, the embodiment of the present invention also provides a three-dimensional model construction system for building design. The three-dimensional model construction system for building design is used to execute the above three-dimensional model construction method for building design. Figure 6 FIG. is a schematic structural diagram of a three-dimensional model construction system for building design provided in another embodiment of the present invention, as Figure 6 shown. The three-dimensional model construction system for building design may vary greatly due to configuration or performance, and may include one or more processors 301 and a memory 302. The memory 302 is used to store computer programs that can run on the processor 301. The processor 301 is used to execute the programs stored in the memory 302 to implement each step in the above Figure 1 method embodiment. Among them, the memory 302 can be short-term storage or persistent storage. The application programs stored in the memory 302 may include one or more modules (not shown in the figure), and each module may include a series of computer-executable instructions for the three-dimensional model construction system for building design.

[0089] Furthermore, the processor 301 can be set to communicate with the memory 302 and execute a series of computer-executable instructions in the memory 302 on the three-dimensional model construction system for building design. The three-dimensional model construction system for building design may further include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input / output interfaces 305, and one or more keyboards 306.

[0090] Specifically, in this embodiment, the three-dimensional model construction system for architectural design includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the bus. The memory is used to store computer programs. The processor is used to execute the programs stored on the memory to implement the above Figure 1 steps in the method embodiments above, and has the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present invention will not be described in detail here.

[0091] It should be noted that the three-dimensional model construction system for architectural design provided by the embodiments of the present invention and the three-dimensional model construction method for architectural design provided by the embodiments of the present invention are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned three-dimensional model construction method for architectural design, and has the same or similar beneficial effects. The repeated parts will not be described in detail.

[0092] It should be noted that the above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0094] The embodiments of the present invention also propose a computer-readable storage medium. The computer-readable medium stores one or more programs. When the one or more programs are executed by an electronic device including multiple application programs, the electronic device is caused to execute Figure 1 the methods disclosed in the illustrated embodiments and implement the functions and beneficial effects of the various methods in the foregoing method embodiments, which will not be described in detail here.

[0095] Among them, the computer-readable storage medium includes a read-only memory (ROM for short), a random access memory (RAM for short), a magnetic disk, or an optical disc, etc.

Claims

1. A three-dimensional model construction method for architectural design, characterized in that: The three-dimensional model construction method for architectural design comprises: Clustering the point cloud data of the building to obtain a plurality of point cloud clusters, removing abnormal point cloud data of abnormal wall parts in the point cloud clusters according to first direction vectors and first Euclidean distances between adjacent point cloud data in the point cloud clusters, and obtaining target point cloud clusters belonging to the connection of building components; Determine a regular non-wall area connected to the wall according to a second Euclidean distance and a second direction vector between adjacent point cloud data in the target point cloud cluster; Determine the structural identity of the point cloud data on both sides of the wall in the target point cloud cluster according to the point cloud data on the inner and outer sides of the regular area on the same vector on both sides of the wall and the regularity coefficient of the target point cloud cluster where the point cloud data on the inner and outer sides are located; When the identity is greater than or equal to a first threshold, determining that the building components formed by the point cloud data on both sides of the wall are the same building component; Connecting the point cloud data belonging to the same building component on both sides of the wall to obtain the point cloud data of the complete building component; constructing a three-dimensional model of the building component according to the point cloud data of the complete building component; The determining of a regular non-wall area connected to a wall according to the second Euclidean distance between adjacent point cloud data in the target point cloud cluster and the second direction vector comprises: Acquire a second Euclidean distance and a second direction vector between adjacent point cloud data in the target point cloud cluster; Determining whether the plane where the adjacent point cloud data are located is a wall plane by using the second Euclidean distance and the second direction vector; In the case of the wall plane, determining a third direction vector in the wall plane according to adjacent point cloud data of the wall plane; Determine, according to a second direction vector between adjacent point cloud data in the target point cloud cluster and a fourth direction vector of adjacent point cloud data of the target building component, that the point cloud data corresponding to the second direction vector is a non-wall component connected to the wall; Determine a second dispersion parameter of adjacent point cloud data at the same height according to a third Euclidean distance between adjacent point cloud data of the non-wall component and a second angle parameter between the fifth direction vectors; When the second dispersion parameter is a predetermined value, determining that a shape formed by adjacent point cloud data is a regular shape, and an area corresponding to the regular shape is a regular area; Extending from the left side of the point cloud data of the portion of the regular shape in contact with the wall along the normal vector of the third direction vector to determine whether there is point cloud data of a non-wall portion outside the wall; In the case where the point cloud data of the non-wall exists, determining a third dispersion parameter of the non-wall according to a fourth Euclidean distance and a sixth direction vector between adjacent point cloud data of the point cloud data of the non-wall; When the third dispersion parameter is equal to the predetermined value, it is determined that the shape formed by adjacent point cloud data of the non-wall point cloud data is a regular shape, and the area corresponding to the regular shape is a regular area.

2. The three-dimensional model construction method for architectural design according to claim 1, characterized in that: The step of removing abnormal point cloud data of abnormal wall parts in the point cloud cluster according to the first direction vector and the first Euclidean distance between adjacent point cloud data in the point cloud cluster to obtain a target point cloud cluster belonging to the connection of the building component comprises: Determine a point cloud density parameter of the point cloud cluster according to a first number of point cloud data in the point cloud cluster, a spatial volume of the point cloud data, and an average value of Euclidean distances between point cloud data in the point cloud cluster; In a case where the point cloud density parameter is greater than or equal to a second threshold, determining the point cloud cluster as an initial target point cloud cluster belonging to a connection of a building component; Acquire a first direction vector and a first Euclidean distance between adjacent point cloud data at the same height in the initial target point cloud cluster; Determine a first dispersion parameter of adjacent point cloud data at the same height according to a first Euclidean distance between adjacent point cloud data and a first angle parameter between adjacent first direction vectors; Determine a regularity coefficient of the initial target point cloud cluster according to a first dispersion parameter of adjacent point cloud data at each height in the initial target point cloud cluster, a maximum height value in the initial target point cloud cluster, and a point cloud density parameter of the initial target point cloud cluster; When the regularity coefficient of the initial target point cloud cluster is greater than or equal to a third threshold, the initial target point cloud cluster is determined to be a target point cloud cluster belonging to the connection of building components; when the regularity coefficient of the initial target point cloud cluster is less than the third threshold, the initial target point cloud cluster is determined to belong to an abnormal part of the wall of the building.

3. The three-dimensional model construction method for architectural design according to claim 2, characterized in that: Determining the point cloud density parameter of the point cloud cluster according to the first number of point cloud data in the point cloud cluster, the spatial volume of the point cloud data, and the average value of the Euclidean distance between the point cloud data in the point cloud cluster includes: calculating a first ratio between the first quantity and the spatial volume, and a reciprocal of an average value of the Euclidean distances; calculating a first product between the first ratio and the reciprocal; The first product is normalized to obtain a point cloud density parameter of the point cloud cluster.

4. The three-dimensional model construction method for architectural design according to claim 2, characterized in that: The determining of the first dispersion parameter of the adjacent point cloud data at the same height according to the first Euclidean distance between the adjacent point cloud data and the first angle parameter between the adjacent first direction vectors comprises: Calculating the absolute value of the first difference between the first Euclidean distances between adjacent point cloud data and the second ratio between the first angle parameters between adjacent first direction vectors; Calculating a second product of the absolute value of the first difference and the second ratio, and superimposing each of the second products to obtain a first superimposed value; Calculating a second difference between the amount of point cloud data in the initial target point cloud cluster and a preset value; A third ratio between the first superposition value and the second difference value is determined as a first dispersion parameter of adjacent point cloud data at the same height.

5. The three-dimensional model construction method for architectural design according to claim 2, characterized in that: The determining of the regularity coefficient of the initial target point cloud cluster according to the first dispersion parameter of the adjacent point cloud data at each height in the initial target point cloud cluster, the maximum height value in the initial target point cloud cluster and the point cloud density parameter of the initial target point cloud cluster comprises: Superimposing first dispersion parameters of adjacent point cloud data at each height in the initial target point cloud cluster to obtain a second superposition value; Calculating a fourth ratio between the second superposition value and the maximum height value, and a third product between the fourth ratio and a point cloud density parameter of the initial target point cloud cluster; The third product is normalized using a derivative function of a sigmoid function to obtain a regularity coefficient of the initial target point cloud clustering.

6. The three-dimensional model construction method for architectural design according to claim 1, characterized in that: The step of determining the structural identity of the point cloud data on both sides of the wall in the target point cloud cluster according to the point cloud data on both sides of the regular area on the same vector on both sides of the wall and the regularity coefficient of the target point cloud cluster where the point cloud data on both sides of the wall are located comprises: From the regular areas on the same vector on both sides of the wall, taking the point cloud data on the inner and outer sides of the wall as the first starting point and the second starting point respectively, obtaining the fifth Euclidean distance and the seventh direction vector between the point cloud data adjacent to the first starting point, and obtaining the sixth Euclidean distance and the eighth direction vector between the point cloud data adjacent to the second starting point; According to the fifth Euclidean distance, the seventh direction vector, the sixth Euclidean distance, the eighth direction vector, and the regularity coefficient of the target point cloud cluster where the first starting point and the second starting point are located, the identity of the point cloud structure on both sides of the wall in the target point cloud cluster where the first starting point and the second starting point are located is determined.

7. The three-dimensional model construction method for architectural design according to claim 6, characterized in that: The determining, based on the fifth Euclidean distance, the seventh direction vector, the sixth Euclidean distance, the eighth direction vector, and the regularity coefficient of the target point cloud cluster where the first starting point and the second starting point are located, the identity of the point cloud structure on both sides of the wall in the target point cloud cluster where the first starting point and the second starting point are located comprises: Determine a distance difference parameter between the first starting point and the second starting point according to the fifth Euclidean distance and the sixth Euclidean distance; Determine a direction difference parameter between the first starting point and the second starting point according to the seventh direction vector and the eighth direction vector; The identity of the point cloud structures on both sides of the wall in the target point cloud cluster where the first starting point and the second starting point are located is determined according to the distance difference parameter, the direction difference parameter, and the regularity coefficient of the target point cloud cluster where the first starting point and the second starting point are located.

8. A three-dimensional model construction system for architectural design, characterized in that: include: A processor and a memory; wherein the memory is used to store a computer program that can be run on the processor; The processor is used to execute the program stored in the memory to implement the steps of the three-dimensional model construction method for architectural design as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, which, when executed by an electronic device including multiple application programs, enable the electronic device to perform the steps of the three-dimensional model construction method for architectural design as described in any one of claims 1-7.

Images