Three-dimensional space-based interior and exterior wall recognition method, device, equipment and storage medium

Abstract

This application relates to a method, device, equipment and storage medium for identifying interior and exterior walls based on a three-dimensional space, and specifically relates to the field of computer-aided design. The method includes: constructing a three-dimensional space node tree according to the three-dimensional model of building components, where the three-dimensional space node tree is used to describe the three-dimensional space in the three-dimensional model; based on the three-dimensional space node tree, identifying the space types of different three-dimensional spaces in the three-dimensional model, and the space types include: component space, indoor space and outdoor space; based on the space types of the three-dimensional spaces contacted by the target component in the building components, identifying the wall types of the target component, and the wall types include: interior wall, exterior wall. In the above solution, the information of the three-dimensional space is introduced for interior and exterior wall identification, which is beneficial to improving the accuracy of identification.

Description

Technical Field

[0001] The present invention relates to the field of computer-aided design, and particularly to a method, device, equipment and storage medium for identifying interior and exterior walls based on a three-dimensional space. Background Art

[0002] In construction projects, in the past, during the construction drawing design stage and the engineering quantity calculation stage, the data was fragmented: after the construction drawing design was completed, it was necessary to import two-dimensional design drawings and create a three-dimensional model in the engineering quantity calculation software by modeling on the two-dimensional base map for engineering quantity calculation. In the engineering quantity calculation software, the interior and exterior walls were drawn using different function entrances, and the interior and exterior walls were naturally distinguished, and then the quantities of the interior and exterior walls were calculated separately to ensure the accuracy of the data in the engineering quantity calculation stage.

[0003] In order to connect the design and quantity calculation in an integrated manner, it is necessary to distinguish the interior and exterior walls of the building represented by the three-dimensional model in the construction drawing design stage, that is, to identify the interior and exterior walls, so that the three-dimensional model in the construction drawing design stage can be directly imported into the engineering quantity calculation software for engineering quantity calculation, thereby avoiding the previous data fragmentation situation.

[0004] In the related art, the three-dimensional model in the construction drawing design stage is projected to obtain a two-dimensional outer contour, and the exterior wall is found through the two-dimensional outer contour, so as to realize the identification of the interior and exterior walls. In some scenarios, the accuracy of the above-mentioned method for identifying the interior and exterior walls is not high enough. Summary of the Invention

[0005] The present application provides a method, device, equipment and storage medium for identifying interior and exterior walls based on a three-dimensional space, which improves the accuracy of identification. The technical solution is as follows.

[0006] On the one hand, a method for identifying interior and exterior walls based on a three-dimensional space is provided. The method includes:

[0007] Constructing a three-dimensional space node tree according to the three-dimensional model of the building component, where the three-dimensional space node tree is used to describe the three-dimensional space in the three-dimensional model;

[0008] Based on the three-dimensional space node tree, identifying the space types of different three-dimensional spaces in the three-dimensional model, where the space types include: component space, indoor space and outdoor space;

[0009] Based on the space type of the three-dimensional space contacted by the target component in the building component, identifying the wall type of the target component, where the wall types include: interior wall, exterior wall.

[0010] On the other hand, a device for identifying interior and exterior walls based on a three-dimensional space is provided. The device includes:

[0011] A node tree construction module for constructing a three-dimensional space node tree based on the three-dimensional model of building components, where the three-dimensional space node tree is used to describe the three-dimensional space in the three-dimensional model;

[0012] A space type recognition module for identifying the space types of different three-dimensional spaces in a three-dimensional model based on the three-dimensional space node tree, where the space types include: component space, indoor space, and outdoor space;

[0013] A wall type recognition module for identifying the wall type of a target component based on the space type of the three-dimensional space contacted by the target component in the building components, where the wall types include: interior wall and exterior wall.

[0014] In a possible implementation, the node tree construction module includes: a bounding box division sub-module and a node tree construction sub-module;

[0015] The bounding box division sub-module is used to divide bounding boxes for the components in the building components according to the three-dimensional model, and one component corresponds to at least one bounding box;

[0016] The node tree construction sub-module is used to construct the three-dimensional space node tree using the divided bounding boxes.

[0017] In a possible implementation, the node tree construction sub-module is further used to:

[0018] Obtain the bounding box reference values of each divided bounding box in three dimensions to form three first coordinate sequences, where the first coordinate sequences correspond to the dimensions one by one, and the sequence values in each first coordinate sequence include: the bounding box reference values of all the bounding boxes in one dimension, and the bounding box reference value includes: the maximum dimension value and the minimum dimension value of the bounding box;

[0019] Based on the sequence reference values in each first coordinate sequence, obtain a first merged bounding box, and use the first merged bounding box as the root node of the three-dimensional space node tree, where the sequence reference value includes: the maximum sequence value and the minimum sequence value in the first coordinate sequence;

[0020] Use the three first coordinate sequences, and perform recursive division with the root node as the initial node until a first recursive termination condition is met to obtain the three-dimensional space node tree.

[0021] In a possible implementation, the node tree construction sub-module is further used to:

[0022] Determine a target node and the target bounding box corresponding to the target node, where the target node is the node to be recursively divided currently;

[0023] From the three first coordinate sequences, respectively extract the sequence values included in the target bounding box to form three sub-coordinate sequences;

[0024] Take the coordinate axis of the target sub-coordinate sequence as the division direction, take the median of the sequence values in the target sub-coordinate sequence, divide the target bounding box to form two child nodes of the target node, and the target sub-coordinate sequence is the one with the largest number of sequence values among the three sub-coordinate sequences;

[0025] Repeat the above steps until all nodes meet the first recursive termination condition to obtain the three-dimensional space node tree;

[0026] Among them, the initial node of the recursive division is the root node.

[0027] In a possible implementation, the first recursive termination condition includes:

[0028] The number of sequence values included in the target bounding box in each of the first coordinate sequences is less than or equal to 2.

[0029] In a possible implementation, the leaf nodes in the three-dimensional space node tree include: leaf nodes of the component space type and leaf nodes of the empty type. The leaf nodes of the component space type correspond to the component space, and the leaf nodes of the empty type correspond to the indoor space or the outdoor space;

[0030] The space type recognition module is further configured to:

[0031] Extract the first merged bounding box corresponding to the root node in the three-dimensional space node tree;

[0032] Perform an outward thickening process on the outer surface of the first merged bounding box to obtain six outer bounding boxes, and use the six outer bounding boxes as the initial traversal queue;

[0033] Use the traversal queue to perform a traversal process on the three-dimensional space node tree to identify the leaf nodes of the outdoor space type among the leaf nodes of the empty type.

[0034] In a possible implementation, the space type recognition module is further configured to:

[0035] Extract a bounding box from the traversal queue and determine the bounding boxes adjacent to the extracted bounding box in the three-dimensional model;

[0036] Update the leaf nodes of the empty type corresponding to the bounding boxes adjacent to the extracted bounding box in the three-dimensional space node tree to leaf nodes of the outdoor space type;

[0037] Add the bounding box corresponding to the leaf node whose type is updated in the three-dimensional space node tree to the traversal queue;

[0038] Repeat the above steps until the traversal queue is empty.

[0039] In a possible implementation, the bounding box division sub-module is further configured to:

[0040] For the target component in the building component, extract the feature points of the target component according to the three-dimensional model;

[0041] Obtain the coordinate values of the feature points in three dimensions as sequence values, and form three second coordinate sequences, which correspond to the dimensions one by one;

[0042] Based on the sequence reference values in each of the second coordinate sequences, obtain a second merged bounding box, where the sequence reference values include: the maximum sequence value and the minimum sequence value in the second coordinate sequence;

[0043] Use the three second coordinate sequences to recursively divide the second merged bounding box until the second recursive termination condition is met, to obtain all the bounding boxes corresponding to the target component.

[0044] In a possible implementation, the bounding box division sub-module is further configured to:

[0045] Determine the target bounding box to be recursively divided currently;

[0046] Respectively extract the sequence values included in the target bounding box from the three second coordinate sequences to form three sub-coordinate sequences;

[0047] Take the coordinate axis of the target sub-coordinate sequence as the division direction, and take the median of the sequence values in the target sub-coordinate sequence to divide the target bounding box to obtain two divided bounding boxes, where the target sub-coordinate sequence is the one with the largest number of sequence values among the three sub-coordinate sequences;

[0048] Repeat the above steps until all the bounding boxes meet the second recursive termination condition, to obtain all the bounding boxes corresponding to the target component.

[0049] In a possible implementation, the second recursive termination condition includes:

[0050] The target bounding box to be recursively divided currently is completely inside the target component;

[0051] Or,

[0052] The target bounding box to be recursively partitioned currently is completely outside the target component;

[0053] Or,

[0054] The target bounding box to be recursively partitioned currently intersects the target component partially, and the size of the target bounding box is less than a preset precision value.

[0055] In a possible implementation manner, the wall type recognition module is further configured to:

[0056] When there is a bounding box in the target component that is in contact with the outdoor space, recognize the wall type of the target component as the exterior wall;

[0057] When there is no bounding box in the target component that is in contact with the outdoor space, recognize the wall type of the target component as the interior wall.

[0058] In another aspect, a computer device is provided. The computer device includes a processor and a memory. At least one instruction, at least one program, a code set or an instruction set is stored in the memory. The at least one instruction, at least one program, the code set or the instruction set is loaded and executed by the processor to implement the above-mentioned method for identifying interior and exterior walls based on a three-dimensional space.

[0059] In yet another aspect, a computer-readable storage medium is provided. At least one instruction is stored in the storage medium. The at least one instruction is loaded and executed by the processor to implement the above-mentioned method for identifying interior and exterior walls based on a three-dimensional space.

[0060] In still another aspect, a computer program product or a computer program is provided. The computer program product or the computer program includes computer instructions. The computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the above-mentioned method for identifying interior and exterior walls based on a three-dimensional space.

[0061] The technical solution provided by this application may include the following beneficial effects:

[0062] Construct a three-dimensional space node tree according to the three-dimensional model of the building component. The three-dimensional space node tree can represent the three-dimensional space in the three-dimensional model, and then identify the space type of the three-dimensional space. According to the contact relationship between the component and the space type, identify the wall type of the component. Compared with the method of identifying interior and exterior walls based on the two-dimensional outer contour of the three-dimensional model, the information of the three-dimensional space is introduced for identifying interior and exterior walls, which is beneficial to improving the accuracy of identification. Description of the Drawings

[0063] To more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0064] Figure 1 It is a schematic diagram of a building including a patio shown according to an exemplary embodiment.

[0065] Figure 2 It is a schematic diagram of a building with an open balcony containing a low wall shown according to an exemplary embodiment.

[0066] Figure 3 It is a schematic diagram of a building including a wall hole (or slab hole) provided according to an exemplary embodiment.

[0067] Figure 4 It is a flowchart of a method for identifying interior and exterior walls based on a three-dimensional space shown according to an exemplary embodiment.

[0068] Figure 5 It is a schematic diagram of the recognition effect of space types shown according to an exemplary embodiment.

[0069] Figure 6 It is a flowchart of a method for identifying interior and exterior walls based on a three-dimensional space shown according to an exemplary embodiment.

[0070] Figure 7 It is a schematic diagram of constructing a three-dimensional space node tree with the properties of a balanced binary tree shown according to an exemplary embodiment.

[0071] Figure 8 It is a schematic diagram of identifying wall types shown according to an exemplary embodiment.

[0072] Figure 9 It is a flowchart of a method for constructing a three-dimensional space node tree shown according to an exemplary embodiment.

[0073] Figure 10 It is a flowchart of a method for identifying space types shown according to an exemplary embodiment.

[0074] Figure 11 It is a flowchart of a method for dividing the bounding box of components shown according to an exemplary embodiment.

[0075] Figure 12 It is a schematic diagram of dividing a bounding box for components shown according to an exemplary embodiment.

[0076] Figure 13 It is a structural block diagram of an interior and exterior wall recognition device based on a three-dimensional space shown according to an exemplary embodiment.

[0077] Figure 14 It is a schematic diagram of a computer device provided according to an exemplary embodiment of the present application. Specific embodiments

[0078] The technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0079] It should be understood that the "indication" mentioned in the embodiments of the present application can be a direct indication, an indirect indication, or a representation of an associated relationship. For example, A indicates B, which can mean that A directly indicates B. For example, B can be obtained through A; it can also mean that A indirectly indicates B. For example, A indicates C, and B can be obtained through C; it can also mean that there is an associated relationship between A and B.

[0080] In the description of the embodiments of the present application, the term "corresponding" can represent a direct or indirect corresponding relationship between two parties, can also represent an associated relationship between the two parties, or can be a relationship such as indication and being indicated, configuration and being configured, etc.

[0081] In the embodiments of the present application, "predefined" can be implemented by pre-saving corresponding codes, tables, or other means that can be used to indicate relevant information in a device (for example, including a terminal device and a network device). The present application does not limit its specific implementation manner.

[0082] Before describing each embodiment shown in the present application, the concepts involved in the present application will be introduced first.

[0083] Exterior wall: A wall with at least one side in direct contact with outdoor air.

[0084] Interior wall: A wall that serves as a space partition indoors and is not in direct contact with outdoor air.

[0085] Among them, direct contact means that there is no other wall or door and window separating the wall from outdoor air.

[0086] Distinguishing the interior and exterior walls of a building mainly serves the engineering quantity calculation stage and is used to distinguish the list quotas of interior and exterior walls.

[0087] In the previous construction drawing design stage and engineering quantity calculation stage, the data was fragmented: after the construction drawing design was completed, it was necessary to import two-dimensional design drawings and create a three-dimensional model in the engineering quantity calculation software by modeling on the two-dimensional base drawing for engineering quantity calculation. In the engineering quantity calculation software, different function entrances were used to draw the inner and outer walls, and the inner and outer walls were naturally distinguished.

[0088] In order to connect the design and quantity calculation into one, and enable the three-dimensional model in the construction drawing design stage to be directly imported into the engineering quantity calculation software for engineering quantity calculation, it is necessary to efficiently identify the inner and outer walls.

[0089] Among the existing software products on the market, there is a function to distinguish the inner and outer walls with one click. This software product identifies based on the two-dimensional outer contour after projection: project the three-dimensional model to obtain the two-dimensional outer contour, and find the outer wall through the two-dimensional outer contour.

[0090] In the above identification method, three-dimensional information is lacking, and some scenarios cannot be correctly identified.

[0091] Exemplarily, for example Figure 1 For the building containing a patio shown, it is difficult to identify the outer wall based on the two-dimensional outer contour. Exemplarily, for example Figure 2 For the building with an open balcony containing a low wall shown, it is difficult to identify the outer wall based on the two-dimensional outer contour. Exemplarily, for example Figure 3 For the building containing a wall hole (or slab hole) shown, it is difficult to identify the outer wall based on the two-dimensional outer contour.

[0092] In response to the above problems, the present application provides a method for identifying inner and outer walls based on three-dimensional space, using space to replace the expression of air. The spatial types of space include: component space, indoor space, and outdoor space. The computer device automatically identifies the inner and outer walls according to the contact relationship between the target component and the spaces of different spatial types. Compared with the method of identifying inner and outer walls based on the two-dimensional outer contour of the three-dimensional model, the information of the three-dimensional space is introduced for inner and outer wall identification, which is beneficial to improving the accuracy of identification.

[0093] Next, the technical solutions provided by the present application will be described in conjunction with the following several embodiments.

[0094] Figure 4 is a flowchart of a method for identifying inner and outer walls based on three-dimensional space shown according to an exemplary embodiment. This method is executed by a computer device. As Figure 4 shown, the method for identifying inner and outer walls based on three-dimensional space may include the following steps:

[0095] Step 410: Construct a three-dimensional space node tree according to the three-dimensional model of the building components. The three-dimensional space node tree is used to describe the three-dimensional space in the three-dimensional model.

[0096] The three-dimensional model of a building component is used to describe various types of information of the building component, such as geometric information, professional attributes, and status information.

[0097] The three-dimensional model of a building component can be divided into different three-dimensional spaces. In the embodiments of the present application, a three-dimensional space node tree is used to describe the three-dimensional spaces in the three-dimensional model. In the three-dimensional space node tree, there is a root node and several leaf nodes, and the leaf nodes of the three-dimensional space node tree can represent the divided three-dimensional spaces in the three-dimensional model.

[0098] Optionally, the three-dimensional space node tree is a balanced binary tree. A balanced binary tree is a binary sorting tree in which the height difference between the left subtree and the right subtree of each node is at most equal to 1.

[0099] Step 420: Based on the three-dimensional space node tree, identify the space types of different three-dimensional spaces in the three-dimensional model. The space types include: component space, indoor space, and outdoor space.

[0100] After the computer device completes the expression of the three-dimensional spaces of the three-dimensional model of the building component through the space node tree, it identifies the space types of the divided three-dimensional spaces and correspondingly updates the types of the corresponding leaf nodes in the three-dimensional space node tree. Next, the meanings of different space types are described:

[0101] · Component space

[0102] The space formed by the components in the three-dimensional model.

[0103] · Indoor space

[0104] In the three-dimensional model, the space belonging to the indoor environment.

[0105] · Outdoor space

[0106] In the three-dimensional model, the space belonging to the outdoor environment.

[0107] Exemplarily, with reference to Figure 5 , it shows the recognition effect of the space type. Figure 5 On the left side of Figure 5 is shown the three-dimensional model of the building component, Figure 5 and on the right side is the space after recognition, including: component space, indoor space, and outdoor space. It should be understood that the outdoor space also includes the space outside the model of the three-dimensional model, which is not shown in the legend in

[0108] Step 430: Based on the space type of the three-dimensional space contacted by the target component in the building component, identify the wall type of the target component. The wall types include: interior wall, exterior wall.

[0109] According to the concept introductions of the interior wall and the exterior wall in the above text, the main basis for distinguishing the interior wall and the exterior wall is: whether it is in contact with outdoor air. In the 3D model of building components, there is no expression of air. Therefore, in the embodiments of the present application, space is used to replace the expression of air, and the space types include: component space, indoor space, and outdoor space. According to the contact relationship between the target component and these spaces, the interior wall and the exterior wall are identified.

[0110] For example: when the 3D space contacted by the target component in the building component includes the outdoor space, the wall type of the target component is the exterior wall; when the 3D space contacted by the target component in the building component does not include the outdoor space, the wall type of the target component is the interior wall.

[0111] In summary, the interior and exterior wall identification method based on 3D space provided in this embodiment constructs a 3D space node tree according to the 3D model of the building component. The 3D space node tree can express the 3D space in the 3D model, and then identify the space type of the 3D space. According to the contact relationship between the component and the space type, the wall type of the component is identified. Compared with the method of identifying the interior and exterior walls based on the 2D outer contour of the 3D model, the information of the 3D space is introduced for the identification of the interior and exterior walls, which is beneficial to improving the accuracy of the identification.

[0112] In an exemplary embodiment, the bounding box is used to divide the 3D space in the 3D model, and the nodes in the 3D space node tree correspond to the bounding boxes, thereby constructing the 3D space node tree.

[0113] Figure 6 It is a flowchart of a method for identifying interior and exterior walls based on 3D space shown in an exemplary embodiment. This method is executed by a computer device. As Figure 6 shown, the method for identifying interior and exterior walls based on 3D space may include the following steps:

[0114] Step 610: According to the 3D model, divide bounding boxes for the components in the building component respectively, and at least one bounding box corresponds to one component.

[0115] The computer device obtains the geometric information of the 3D model, and based on the geometric information, divides bounding boxes for the components in the building component respectively.

[0116] For example: when the component is a regular component, one bounding box corresponds to one component; when the component is an irregular component, multiple bounding boxes correspond to one component.

[0117] Among them, the regular component refers to a straight wall with horizontal and vertical properties, and the irregular component refers to a wall with non-horizontal and vertical properties. For example: the irregular components include: inclined walls, arc walls, and irregular floor slabs.

[0118] Optionally, the bounding boxes divided in the embodiments of the present application are axis-aligned bounding boxes (AABBs).

[0119] Step 620: Use the divided bounding boxes to construct a three-dimensional space node tree.

[0120] After the bounding boxes are respectively divided for the components in the building components, the computer device uses these divided bounding boxes to construct a three-dimensional space node tree. At this time, since the bounding boxes are used to represent the three-dimensional space in the three-dimensional model and the three-dimensional space node tree is constructed using the bounding boxes, the nodes of the three-dimensional space node tree correspond to the bounding boxes.

[0121] Exemplarily, with reference to Figure 7 , the bounding boxes are respectively divided for the components in the building components. In Figure 7 , each component is expressed using a single bounding box, and thus bounding boxes 1 to 9 are obtained. These bounding boxes are constructed into a three-dimensional space node tree with the property of a balanced binary tree according to certain division rules. The leaf nodes of the three-dimensional space node tree correspond to one of the bounding boxes 1 to 9.

[0122] Step 630: Based on the three-dimensional space node tree, identify the space types of different three-dimensional spaces in the three-dimensional model. The space types include: component space, indoor space, and outdoor space.

[0123] Step 640: Based on the space type of the three-dimensional space contacted by the target component in the building components, identify the wall type of the target component. The wall types include: interior wall, exterior wall.

[0124] In a possible implementation manner, step 640 includes: when there is a bounding box in the target component that is in contact with the outdoor space, identify the wall type of the target component as an exterior wall; when there is no bounding box in the target component that is in contact with the outdoor space, identify the wall type of the target component as an interior wall.

[0125] Exemplarily, with reference to Figure 8 , in the figure, there are walls 1, 2, and 3 marked with arrows. Among them, wall 2 is adjacent to the indoor space and the component space, and wall 2 is identified as an interior wall; wall 1 is adjacent to the indoor space, the component space, and the outdoor space (the space outside the model of the three-dimensional model, not shown in the legend), and wall 1 is identified as an exterior wall; wall 3 is adjacent to the indoor space, the component space, and the outdoor space, and wall 3 is identified as an exterior wall.

[0126] In summary, the method for identifying internal and external walls based on a three-dimensional space provided in this embodiment constructs a three-dimensional space node tree according to the three-dimensional model of building components. The three-dimensional space node tree can represent the three-dimensional space in the three-dimensional model, and then identify the space type of the three-dimensional space. According to the contact relationship between the components and the space type, the wall type of the components is identified. Compared with the method of identifying internal and external walls based on the two-dimensional outer contour of the three-dimensional model, the information of the three-dimensional space is introduced for identifying internal and external walls, which is beneficial to improving the accuracy of identification.

[0127] At the same time, the method for identifying internal and external walls based on a three-dimensional space provided in this embodiment uses bounding boxes to divide the three-dimensional space in the three-dimensional model. The nodes in the three-dimensional space node tree correspond to the bounding boxes, so as to construct a three-dimensional space node tree for representing the three-dimensional space, ensuring the accuracy of the space representation of the three-dimensional space node tree.

[0128] In an exemplary embodiment, a three-dimensional space node tree is constructed by means of recursive processing.

[0129] Figure 9 is a flowchart of a method for constructing a three-dimensional space node tree shown according to an exemplary embodiment. This method is executed by a computer device. As Figure 9 shown, the method for constructing the three-dimensional space node tree (i.e., step 620 in the above embodiment) may include the following steps:

[0130] Step 910: Obtain the bounding box reference values of each divided bounding box in three dimensions to form three first coordinate sequences. The first coordinate sequences correspond to the dimensions one by one. The sequence values in each first coordinate sequence include: the bounding box reference values of all bounding boxes in one dimension. The bounding box reference value includes: the maximum dimension value and the minimum dimension value of the bounding box.

[0131] The computer device takes out the divided bounding boxes of all components and takes out the maximum dimension values and minimum dimension values of all bounding boxes in three dimensions, so as to form three first coordinate sequences.

[0132] Optionally, the sequence values in each first coordinate sequence are de-duplicated and sorted.

[0133] Exemplarily, all bounding boxes include: bounding box 1 and bounding box 2. According to the maximum dimension value and the minimum dimension value of each bounding box in each dimension, it is obtained that the bounding box reference values of bounding box 1 include: (X1, X2), (Y1, Y2), (Z1, Z2); the bounding box reference values of bounding box 2 include: (X1, X3), (Y1, Y2), (Z3, Z4). Then, based on the bounding box reference values of the above-mentioned respective bounding boxes, the first coordinate sequence in the X-axis dimension is (X1, X2, X3), the first coordinate sequence in the Y-axis dimension is (Y1, Y2), and the first coordinate sequence in the Z-axis dimension is (Z1, Z2, Z3, Z4).

[0134] Step 920: Based on the sequence reference values in each first coordinate sequence, obtain a first merged bounding box, and use the first merged bounding box as the root node of the three-dimensional space node tree, where the sequence reference values include: the maximum sequence value and the minimum sequence value in the first coordinate sequence.

[0135] The computer device calculates the bounding box after merging all the bounding boxes of the components, that is, the first merged bounding box, and the first merged bounding box covers all the bounding boxes and is used as the root node of the three-dimensional space node tree.

[0136] Exemplarily, the first coordinate sequence in the X-axis dimension is (X1, X2, X3), the first coordinate sequence in the Y-axis dimension is (Y1, Y2), and the first coordinate sequence in the Z-axis dimension is (Z1, Z2, Z3, Z4). Then, the maximum sequence value and the minimum sequence value in each first coordinate sequence are taken out, and it is obtained that the first merged bounding box is a bounding box in the shape of (X1, X3), (Y1, Y2), (Z1, Z4).

[0137] Step 930: Use the three first coordinate sequences, and perform recursive partitioning with the root node as the initial node until the first recursive termination condition is met, to obtain the three-dimensional space node tree.

[0138] Exemplarily, perform recursive partitioning with the root node as the initial node, and obtain: node 1 and node 2; determine whether node 1 meets the first recursive termination condition, if not, continue to perform recursive partitioning on node 1; determine whether node 2 meets the first recursive termination condition, if not, continue to perform recursive partitioning on node 2.

[0139] In a possible implementation manner, step 930 includes:

[0140] (1) Determine the target node and the target bounding box corresponding to the target node, where the target node is the node to be recursively partitioned currently.

[0141] Among them, the initial node of the recursive partitioning is the root node.

[0142] (2) From the three first coordinate sequences, the sequence values contained within the target bounding box are respectively taken out to form three sub-coordinate sequences.

[0143] (3) Take the coordinate axis of the target sub-coordinate sequence as the division direction, take the median of the sequence values in the target sub-coordinate sequence, divide the target bounding box, and form two child nodes of the target node. The target sub-coordinate sequence is the one with the largest number of sequence values among the three sub-coordinate sequences.

[0144] Exemplarily, the three sub-coordinate sequences are: the sub-coordinate sequence in the X-axis dimension is (X1, X2, X3), the sub-coordinate sequence in the Y-axis dimension is (Y1, Y2), and the sub-coordinate sequence in the Z-axis dimension is (Z1, Z2). Therefore, the length of the sub-sequence in the X-axis dimension is 3, the length of the sub-sequence in the Y-axis dimension is 2, and the length of the sub-sequence in the Z-axis dimension is 2. Then the target sub-coordinate sequence is the sub-coordinate sequence in the X-axis dimension.

[0145] Take the X-axis as the division direction, take the median X2 to divide the bounding box, and obtain the two divided bounding boxes as the child nodes of the current node. Among them, bounding box 1 is a bounding box in the shape of (X1, X2), (Y1, Y2), (Z1, Z2), and bounding box 2 is a bounding box in the shape of (X2, X3), (Y1, Y2), (Z1, Z2).

[0146] (4) Repeat the above steps until all nodes meet the first recursive termination condition to obtain a three-dimensional space node tree.

[0147] The computer device continuously performs recursive partitioning until all nodes meet the first recursive termination condition.

[0148] Optionally, the first recursive termination condition includes: the number of sequence values contained within the target bounding box in each first coordinate sequence is less than or equal to 2.

[0149] It should be understood that when the recursion terminates, there are only two cases. One is that the bounding box of the current node is contained within the bounding box of a certain component, and in this case, the type label of the leaf node is the component space. The other is that the bounding box of the current node is completely outside the bounding boxes of all components, and in this case, the type label of the leaf node is empty.

[0150] In summary, the method for constructing a three-dimensional space node tree provided in this embodiment uses a recursive processing method to construct a three-dimensional space node tree, ensuring the reliability of the construction of the three-dimensional space node tree.

[0151] The leaf nodes in the three-dimensional space node tree include: leaf nodes of the component space type and leaf nodes of the empty type. The leaf nodes of the component space type correspond to the component space, and the leaf nodes of the empty type correspond to the indoor space or the outdoor space.

[0152] The leaf node of the component space type means that in the three-dimensional model, the bounding box corresponding to this leaf node is completely contained inside the bounding box of a certain component. The leaf node of the empty type means that in the three-dimensional model, the bounding box corresponding to this leaf node is completely outside the bounding boxes of all components.

[0153] In an exemplary embodiment, the leaf nodes of the empty type in the three-dimensional space node tree are traversed to distinguish the indoor space and the outdoor space in the three-dimensional space.

[0154] Figure 10 is a flowchart of a method for identifying a space type shown according to an exemplary embodiment. This method is executed by a computer device. As Figure 10 shown, the method for identifying the space type (i.e., step 630 in the above embodiment) may include the following steps:

[0155] Step 1010: Take out the first merged bounding box corresponding to the root node in the three-dimensional space node tree.

[0156] Step 1020: Perform an outward thickening process on the outer surface of the first merged bounding box to obtain six outer bounding boxes, and use the six outer bounding boxes as the initial traversal queue.

[0157] The first merged bounding box has six outer surfaces. By thickening these six outer surfaces outward (assuming a thickening of 100 mm, the specific value does not affect the result), six outer bounding boxes can be obtained. Add these six outer bounding boxes to the traversal queue.

[0158] Step 1030: Use the traversal queue to perform a traversal process on the three-dimensional space node tree to identify the leaf nodes of the outdoor space type among the leaf nodes of the empty type.

[0159] Take out a bounding box from the traversal queue, determine the bounding boxes adjacent to the taken-out bounding box, and find the leaf nodes corresponding to the adjacent bounding boxes in the three-dimensional space node tree and with the type being empty. If found, update the type of the leaf node from empty to outdoor space, and add the bounding box of this leaf node to the traversal queue for subsequent traversal processing.

[0160] In a possible implementation manner, step 1030 includes:

[0161] (1) Take out a bounding box from the traversal queue and determine the bounding boxes adjacent to the taken-out bounding box in the three-dimensional model.

[0162] Among them, the taken-out bounding box can be an outer bounding box or the bounding box corresponding to the leaf node in the three-dimensional space node tree that is subsequently added to the traversal queue.

[0163] (2) Update the leaf nodes of the empty type corresponding to the bounding boxes adjacent to the retrieved bounding box in the three-dimensional space node tree to the leaf nodes of the outdoor space type.

[0164] (3) Add the bounding boxes corresponding to the leaf nodes whose types are updated in the three-dimensional space node tree to the traversal queue.

[0165] (4) Repeat the above steps until the traversal queue is empty.

[0166] After the traversal queue is empty, the remaining leaf nodes of the empty type that have not been updated can be marked as the leaf nodes of the indoor type.

[0167] In summary, the method for identifying the space type provided in this embodiment traverses the leaf nodes of the empty type in the three-dimensional space node tree, so as to distinguish the indoor space and the outdoor space in the three-dimensional space, which is beneficial to the efficient identification of the indoor space and the outdoor space.

[0168] In an exemplary embodiment, a recursive processing method is adopted to divide the bounding boxes for irregular components.

[0169] Figure 11 It is a flowchart of a method for dividing the bounding boxes of components shown according to an exemplary embodiment. This method is executed by a computer device. As Figure 11 shown, the method for dividing the bounding boxes of the components (i.e., step 610 in the above embodiment) may include the following steps:

[0170] Step 1110: For the target component in the building component, according to the three-dimensional model, retrieve the feature points of the target component.

[0171] Optionally, the feature points include: the endpoints of all edges of the component shape, and the interval points. Among them, the interval points refer to the points selected at a certain precision interval for the hypotenuse or the arc edge.

[0172] Step 1120: Obtain the coordinate values of the feature points in three dimensions as sequence values to form three second coordinate sequences, and the second coordinate sequences correspond to the dimensions one by one.

[0173] The computer device obtains the dimension values of all feature points in three dimensions, so as to form three second coordinate sequences.

[0174] Optionally, the sequence values in each second coordinate sequence are de-duplicated and sorted.

[0175] Exemplarily, there are A feature points, A is a positive integer: (X1, Y1, Z1), (X2, Y2, Z2), ……, (X A , Y A , Z A) Then, based on the coordinate values of the above feature points, the second coordinate sequence in the X-axis dimension is obtained as (X1, X2, ……, X A ) The second coordinate sequence in the Y-axis dimension is (Y1, Y2, ……, Y A ) The second coordinate sequence in the Z-axis dimension is (Z1, Z2, ……, Z A ).

[0176] Step 1130: Based on the sequence reference values in each second coordinate sequence, obtain a second merged bounding box, where the sequence reference values include: the maximum sequence value and the minimum sequence value in the second coordinate sequence.

[0177] The computer device calculates the bounding box of the component shape, that is, the second merged bounding box, and the second merged bounding box covers the component.

[0178] Exemplarily, the second coordinate sequence in the X-axis dimension is (X1, X2, X3), the second coordinate sequence in the Y-axis dimension is (Y1, Y2), and the second coordinate sequence in the Z-axis dimension is (Z1, Z2, Z3, Z4). Then, the maximum sequence value and the minimum sequence value in each second coordinate sequence are taken out, and the second merged bounding box obtained is a bounding box in the shape of (X1, X3), (Y1, Y2), (Z1, Z4).

[0179] Step 1140: Use the three second coordinate sequences to recursively divide the second merged bounding box until the second recursive termination condition is met, and obtain all the bounding boxes corresponding to the target component.

[0180] Exemplarily, the second merged bounding box is recursively divided to obtain: bounding box 1 and bounding box 2; it is judged whether bounding box 1 meets the second recursive termination condition. If not, continue to recursively divide bounding box 1; it is judged whether bounding box 2 meets the second recursive termination condition. If not, continue to recursively divide bounding box 2.

[0181] In a possible implementation, step 1140 includes:[[]]

[0182] (1) Determine the target bounding box to be recursively divided currently.

[0183] Among them, the initial target box for recursive division is the second merged bounding box.

[0184] (2) From the three second coordinate sequences, respectively take out the sequence values included in the target bounding box to form three sub-coordinate sequences.

[0185] (3) Take the coordinate axis of the target sub - coordinate sequence as the division direction, take the median of the sequence values in the target sub - coordinate sequence, and divide the target bounding box to obtain two divided bounding boxes. The target sub - coordinate sequence is the one with the largest number of sequence values among the three sub - coordinate sequences.

[0186] Exemplarily, the three sub - coordinate sequences are: the sub - coordinate sequence in the X - axis dimension is (X1, X2, X3), the sub - coordinate sequence in the Y - axis dimension is (Y1, Y2), and the sub - coordinate sequence in the Z - axis dimension is (Z1, Z2). Therefore, the length of the subsequence in the X - axis dimension is 3, the length of the subsequence in the Y - axis dimension is 2, and the length of the subsequence in the Z - axis dimension is 2. Then the target sub - coordinate sequence is the sub - coordinate sequence in the X - axis dimension.

[0187] Take the X - axis as the division direction, take the median X2 to divide the bounding box, and obtain two divided bounding boxes as the child nodes of the current node. Among them, bounding box 1 is a bounding box with the shape of (X1, X2), (Y1, Y2), (Z1, Z2), and bounding box 2 is a bounding box with the shape of (X2, X3), (Y1, Y2), (Z1, Z2).

[0188] (4) Repeat the above steps until all bounding boxes meet the second recursive termination condition, and obtain all the bounding boxes corresponding to the target component.

[0189] Optionally, the second recursive termination condition includes:

[0190] · The target bounding box to be recursively divided currently is completely inside the target component.

[0191] At this time, take this target bounding box as the bounding box corresponding to the target component.

[0192] · The target bounding box to be recursively divided currently is completely outside the target component.

[0193] At this time, do not take this target bounding box as the bounding box corresponding to the target component, that is, discard this target bounding box.

[0194] · The target bounding box to be recursively divided currently partially intersects with the target component, and the size of the target bounding box is less than the preset precision value.

[0195] At this time, take this target bounding box as the bounding box corresponding to the target component.

[0196] In summary, the method for dividing the bounding box of the component provided in this embodiment uses a recursive processing method to divide the bounding box for an irregular component, ensuring the reliability of the result of dividing the bounding box of the irregular component.

[0197] It should be noted that the above - mentioned method embodiments can be implemented separately or in combination, and the present application does not limit this.

[0198] Figure 13 It is a structural block diagram of a three-dimensional space-based interior and exterior wall recognition device shown according to an exemplary embodiment. The device includes:

[0199] A node tree construction module 1301, configured to construct a three-dimensional space node tree according to a three-dimensional model of a building component, where the three-dimensional space node tree is used to describe the three-dimensional space in the three-dimensional model;

[0200] A space type recognition module 1302, configured to recognize the space types of different three-dimensional spaces in the three-dimensional model based on the three-dimensional space node tree, where the space types include: component space, indoor space, and outdoor space;

[0201] A wall type recognition module 1303, configured to recognize the wall type of a target component in the building component based on the space type of the three-dimensional space contacted by the target component, where the wall types include: interior wall, exterior wall.

[0202] In a possible implementation manner, the node tree construction module 1301 includes: a bounding box division sub-module and a node tree construction sub-module;

[0203] The bounding box division sub-module is configured to divide bounding boxes for each component in the building component according to the three-dimensional model, and one component corresponds to at least one bounding box;

[0204] The node tree construction sub-module is configured to construct the three-dimensional space node tree by using the divided bounding boxes.

[0205] In a possible implementation manner, the node tree construction sub-module is further configured to:

[0206] Obtain the bounding box reference values of each divided bounding box in three dimensions to form three first coordinate sequences, where the first coordinate sequences correspond to the dimensions one by one, and the sequence values in each first coordinate sequence include: the bounding box reference values of all the bounding boxes in one dimension, and the bounding box reference value includes: the maximum dimension value and the minimum dimension value of the bounding box;

[0207] Based on the sequence reference values in each first coordinate sequence, obtain a first merged bounding box, and use the first merged bounding box as the root node of the three-dimensional space node tree, where the sequence reference value includes: the maximum sequence value and the minimum sequence value in the first coordinate sequence;

[0208] Use the three first coordinate sequences, and perform recursive division with the root node as the initial node until a first recursive termination condition is met to obtain the three-dimensional space node tree.

[0209] In a possible implementation, the node tree construction sub-module is further configured to:

[0210] Determine a target node and a target bounding box corresponding to the target node, where the target node is the node to be recursively partitioned currently;

[0211] Respectively extract the sequence values included in the target bounding box from the three first coordinate sequences to form three sub-coordinate sequences;

[0212] Take the coordinate axis of the target sub-coordinate sequence as the partitioning direction, take the median of the sequence values in the target sub-coordinate sequence, and partition the target bounding box to form two child nodes of the target node, where the target sub-coordinate sequence is the one with the largest number of sequence values among the three sub-coordinate sequences;

[0213] Repeat the above steps until all nodes meet the first recursive termination condition, and obtain the three-dimensional space node tree;

[0214] Wherein, the initial node of the recursive partitioning is the root node.

[0215] In a possible implementation, the first recursive termination condition includes:

[0216] The number of sequence values included in each of the first coordinate sequences in the target bounding box is less than or equal to 2.

[0217] In a possible implementation, the leaf nodes in the three-dimensional space node tree include: leaf nodes of the component space type and leaf nodes of the empty type, where the leaf nodes of the component space type correspond to the component space, and the leaf nodes of the empty type correspond to the indoor space or the outdoor space;

[0218] The space type recognition module 1302 is further configured to:

[0219] Extract the first merged bounding box corresponding to the root node in the three-dimensional space node tree;

[0220] Perform an outward thickening process on the outer surface of the first merged bounding box to obtain six outer bounding boxes, and use the six outer bounding boxes as the initial traversal queue;

[0221] Use the traversal queue to perform a traversal process on the three-dimensional space node tree to identify the leaf nodes of the outdoor space type in the leaf nodes of the empty type.

[0222] In a possible implementation, the space type recognition module 1302 is further configured to:

[0223] Take out an enclosing box from the traversal queue, and determine the enclosing boxes adjacent to the taken-out enclosing box in the three-dimensional model;

[0224] In the three-dimensional space node tree, update the leaf nodes of the empty type corresponding to the enclosing boxes adjacent to the taken-out enclosing box to the leaf nodes of the outdoor space type;

[0225] Add the enclosing boxes corresponding to the leaf nodes with type updates in the three-dimensional space node tree to the traversal queue;

[0226] Repeat the above steps until the traversal queue is empty.

[0227] In a possible implementation manner, the enclosing box division sub-module is further configured to:

[0228] For a target component in the building component, take out the feature points of the target component according to the three-dimensional model;

[0229] Obtain the coordinate values of the feature points in three dimensions as sequence values, and form three second coordinate sequences, which correspond to the dimensions one by one;

[0230] Based on the sequence reference values in each of the second coordinate sequences, obtain a second combined enclosing box, where the sequence reference values include: the maximum sequence value and the minimum sequence value in the second coordinate sequence;

[0231] Use the three second coordinate sequences to recursively divide the second combined enclosing box until the second recursive termination condition is met, to obtain all the enclosing boxes corresponding to the target component.

[0232] In a possible implementation manner, the enclosing box division sub-module is further configured to:

[0233] Determine the target enclosing box to be recursively divided currently;

[0234] Respectively take out the sequence values included in the target enclosing box from the three second coordinate sequences to form three sub-coordinate sequences;

[0235] Take the coordinate axis of the target sub-coordinate sequence as the division direction, take the median of the sequence values in the target sub-coordinate sequence, and divide the target enclosing box to obtain two divided enclosing boxes, where the target sub-coordinate sequence is the one with the largest number of sequence values among the three sub-coordinate sequences;

[0236] Repeat the above steps until all the enclosing boxes meet the second recursive termination condition, to obtain all the enclosing boxes corresponding to the target component.

[0237] In a possible implementation, the second recursive termination condition includes:

[0238] The current target bounding box to be recursively partitioned is completely inside the target component;

[0239] Or,

[0240] The current target bounding box to be recursively partitioned is completely outside the target component;

[0241] Or,

[0242] The current target bounding box to be recursively partitioned partially intersects the target component, and the size of the bounding box is less than a preset precision value.

[0243] In a possible implementation, the wall type recognition module 1303 is further configured to:

[0244] When there is a bounding box in the target component that is in contact with the outdoor space, identify the wall type of the target component as the exterior wall;

[0245] When there is no bounding box in the target component that is in contact with the outdoor space, identify the wall type of the target component as the interior wall.

[0246] It should be noted that: For the above-described interior and exterior wall recognition device based on a three-dimensional space, only the above-mentioned division of each functional module is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device provided in the above embodiment and the method embodiment belong to the same concept. For the specific implementation process, please refer to the method embodiment and will not be elaborated here.

[0247] Please refer to Figure 14 , which is a schematic diagram of a computer device provided according to an exemplary embodiment of the present application. The computer device includes a memory and a processor. The memory is used to store a computer program. When the computer program is executed by the processor, the above-described semantic similarity calculation method is implemented.

[0248] Among them, the processor can be a Central Processing Unit (CPU). The processor can also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., or a combination of the above types of chips.

[0249] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the methods in the embodiments of the present invention. The processor executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory, that is, to implement the methods in the above method embodiments.

[0250] The memory can include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created by the processor, etc. In addition, the memory can include high-speed random access memory, and can also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely set relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above networks include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0251] In an exemplary embodiment, a computer-readable storage medium is also provided, which is used to store at least one computer program, and the at least one computer program is loaded and executed by the processor to implement all or part of the steps in the above method. For example, the computer-readable storage medium can be a Read-Only Memory (ROM), a Random Access Memory (RAM), a Compact Disc Read-Only Memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0252] Other embodiments of the present application will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The present application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include known common general knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and examples are only illustrative, and the true scope and spirit of the present application are pointed out by the following claims.

[0253] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A method for identifying interior and exterior walls based on three-dimensional space, characterized in that, The method includes: Constructing a three-dimensional space node tree based on the three-dimensional model of the building component, including: dividing bounding boxes for the components in the building component according to the three-dimensional model, where one component corresponds to at least one bounding box; using the divided bounding boxes to construct the three-dimensional space node tree, and the three-dimensional space node tree is used to describe the three-dimensional space in the three-dimensional model; Based on the three-dimensional space node tree, identifying the space types of different three-dimensional spaces in the three-dimensional model, where the space types include: component space, indoor space, and outdoor space; among them, after completing the expression of the three-dimensional space of the three-dimensional model of the building component through the three-dimensional space node tree, identifying the space types of the divided three-dimensional spaces and correspondingly updating the types of the corresponding leaf nodes in the three-dimensional space node tree; Based on the space type of the three-dimensional space contacted by the target component in the building component, identifying the wall type of the target component, where the wall types include: interior wall, exterior wall; The leaf nodes in the three-dimensional space node tree include: leaf nodes of the component space type and leaf nodes of the empty type, where the leaf nodes of the component space type correspond to the component space, and the leaf nodes of the empty type correspond to the indoor space or the outdoor space; The identifying the space types of different three-dimensional spaces in the three-dimensional model based on the three-dimensional space node tree includes: Taking out the first merged bounding box corresponding to the root node in the three-dimensional space node tree; Performing an outward thickening process on the outer surface of the first merged bounding box to obtain six outer bounding boxes, and using the six outer bounding boxes as the initial traversal queue; Using the traversal queue to perform a traversal process on the three-dimensional space node tree to identify the leaf nodes of the outdoor space type in the leaf nodes of the empty type.

2. The method according to claim 1, wherein The constructing the three-dimensional space node tree using the divided bounding boxes includes: Obtaining the bounding box reference values of each of the divided bounding boxes in three dimensions to form three first coordinate sequences, where the first coordinate sequences correspond to the dimensions one by one, and the sequence values in each first coordinate sequence include: the bounding box reference values of all the bounding boxes in one dimension, and the bounding box reference value includes: the maximum dimension value and the minimum dimension value of the bounding box; Based on the sequence reference values in each first coordinate sequence, obtaining a first merged bounding box, and using the first merged bounding box as the root node of the three-dimensional space node tree, where the sequence reference value includes: the maximum sequence value and the minimum sequence value in the first coordinate sequence; Using the three first coordinate sequences, starting from the root node as the initial node, performing recursive partitioning until a first recursive termination condition is met to obtain the three-dimensional space node tree.

3. The method according to claim 2, characterized in that The using the three first coordinate sequences, starting from the root node as the initial node, performing recursive partitioning until a recursive termination condition is met to obtain the three-dimensional space node tree includes: Determining a target node and the target bounding box corresponding to the target node, where the target node is the node to be recursively partitioned currently; From the three first coordinate sequences, respectively extract the sequence values included in the target bounding box to form three sub-coordinate sequences; Take the coordinate axis of the target sub-coordinate sequence as the division direction, take the median of the sequence values in the target sub-coordinate sequence, divide the target bounding box to form two child nodes of the target node, and the target sub-coordinate sequence is the one with the largest number of sequence values among the three sub-coordinate sequences; Repeat the above steps until all nodes meet the first recursive termination condition to obtain the three-dimensional space node tree; Among them, the initial node for recursive division is the root node.

4. The method according to claim 3, wherein The first recursive termination condition includes: The number of sequence values included in the target bounding box for each of the first coordinate sequences is less than or equal to 2.

5. The method according to claim 1, wherein The process of traversing the three-dimensional space node tree using the traversal queue to identify the leaf nodes of the outdoor space type among the leaf nodes of the empty type includes: Take out a bounding box from the traversal queue and determine the bounding boxes adjacent to the taken-out bounding box in the three-dimensional model; Update the leaf nodes of the empty type corresponding to the bounding boxes adjacent to the taken-out bounding box in the three-dimensional space node tree to leaf nodes of the outdoor space type; Add the bounding boxes corresponding to the leaf nodes whose types are updated in the three-dimensional space node tree to the traversal queue; Repeat the above steps until the traversal queue is empty.

6. The method according to claim 1, wherein The process of dividing bounding boxes for each component in the building component according to the three-dimensional model includes: For the target component in the building component, extract the feature points of the target component according to the three-dimensional model; Obtain the coordinate values of the feature points in three dimensions as sequence values to form three second coordinate sequences, and the second coordinate sequences correspond to the dimensions one by one; Based on the sequence reference values in each of the second coordinate sequences, obtain a second merged bounding box, and the sequence reference values include: the maximum sequence value and the minimum sequence value in the second coordinate sequence; Use the three second coordinate sequences to recursively divide the second merged bounding box until the second recursive termination condition is met to obtain all the bounding boxes corresponding to the target component.

7. The method according to claim 6, characterized in that The process of using the three second coordinate sequences to recursively divide the second merged bounding box until the second recursive termination condition is met to obtain all the bounding boxes corresponding to the target component includes: Determine the target bounding box to be recursively divided currently; From the three second coordinate sequences, respectively extract the sequence values included in the target bounding box to form three sub-coordinate sequences; Take the coordinate axis of the target sub-coordinate sequence as the division direction, take the median of the sequence values in the target sub-coordinate sequence, divide the target bounding box to obtain two divided bounding boxes, and the target sub-coordinate sequence is the one with the largest number of sequence values among the three sub-coordinate sequences; Repeat the above steps until all bounding boxes meet the second recursive termination condition to obtain all the bounding boxes corresponding to the target component.

8. The method according to claim 7, wherein The second recursive termination condition includes: The target bounding box to be recursively partitioned is completely inside the target component; Or, The target bounding box to be recursively partitioned is completely outside the target component; Or, The target bounding box to be recursively partitioned intersects the target component partially, and the size of the target bounding box is smaller than a preset precision value.

9. The method according to claim 1, wherein Identifying the wall type of the target component based on the spatial type of the three-dimensional space contacted by the target component in the building component includes: When there is a bounding box in the target component that contacts the outdoor space, identifying the wall type of the target component as the exterior wall; When there is no bounding box in the target component that contacts the outdoor space, identifying the wall type of the target component as the interior wall.

10. An interior and exterior wall recognition device based on a three-dimensional space, characterized in that, The device includes: A node tree construction module, configured to construct a three-dimensional space node tree according to the three-dimensional model of the building component, including: partitioning bounding boxes for the components in the building component respectively according to the three-dimensional model, where one component corresponds to at least one bounding box; using the partitioned bounding boxes to construct the three-dimensional space node tree, and the three-dimensional space node tree is used to describe the three-dimensional space in the three-dimensional model; A spatial type identification module, configured to identify the spatial types of different three-dimensional spaces in the three-dimensional model based on the three-dimensional space node tree, where the spatial types include: component space, indoor space, and outdoor space; wherein, after completing the expression of the three-dimensional space of the three-dimensional model of the building component through the three-dimensional space node tree, identifying the spatial types of the partitioned three-dimensional spaces, and correspondingly updating the types of the corresponding leaf nodes in the three-dimensional space node tree; the leaf nodes in the three-dimensional space node tree include: leaf nodes of component space type, leaf nodes of empty type, the leaf nodes of component space type correspond to the component space, and the leaf nodes of empty type correspond to the indoor space or the outdoor space; identifying the spatial types of different three-dimensional spaces in the three-dimensional model based on the three-dimensional space node tree includes: taking out the first merged bounding box corresponding to the root node in the three-dimensional space node tree; performing an outward thickening process on the outer surface of the first merged bounding box to obtain six outer bounding boxes, and using the six outer bounding boxes as the initial traversal queue; using the traversal queue to perform a traversal process on the three-dimensional space node tree to identify the leaf nodes of outdoor space type in the leaf nodes of empty type; A wall type identification module, configured to identify the wall type of the target component based on the spatial type of the three-dimensional space contacted by the target component in the building component, where the wall types include: interior wall, exterior wall.

11. A computer device, characterized in that, The computer device includes a processor and a memory, and at least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the at least one instruction, at least one program, the code set or the instruction set is loaded and executed by the processor to implement the three-dimensional space-based interior and exterior wall identification method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement the three-dimensional space-based interior and exterior wall recognition method according to any one of claims 1 to 9.

Images