Three-dimensional model generation method, device, equipment and storage medium

Abstract

The present application relates to the technical field of engineering drawing, and particularly relates to a three-dimensional model generation method, device, equipment and storage medium. The method loads a two-dimensional drawing through a building three-dimensional modeling software, identifies layer names of all layers in the two-dimensional drawing, and determines a component type of a component corresponding to each layer according to each layer name; determines a corresponding drawing identification mode and a corresponding information analysis mode according to the component type; extracts a drawing feature group corresponding to the component from a corresponding layer according to the drawing identification mode; analyzes the drawing feature group according to the information analysis mode to obtain modeling information corresponding to the component; and performs three-dimensional modeling based on the modeling information corresponding to each component to obtain a three-dimensional model corresponding to the two-dimensional drawing, thereby solving the problem of low accuracy in converting a two-dimensional drawing into a three-dimensional model in the prior art.

Description

Technical Field

[0001] This invention relates to the field of engineering drawing technology, and in particular to a method, apparatus, device and storage medium for generating three-dimensional models. Background Technology

[0002] Architectural design has long relied on two-dimensional drawings as its primary medium. However, two-dimensional drawings convey less information and are less intuitive than three-dimensional models. The shift from two-dimensional to three-dimensional working modes is an unstoppable trend, making the "two-to-three" conversion technology—bridging the gap between two and three dimensions—extremely important.

[0003] Existing 3D model generation solutions typically require manual creation of 3D models by referring to 2D drawings, while supplementing the corresponding 3D model information. Some functions can automatically generate 3D models from 2D drawings, but these functions often produce poor results, with frequent instances of incorrect generation and missing structures.

[0004] In summary, existing technologies suffer from low accuracy in converting two-dimensional drawings into three-dimensional models. Summary of the Invention

[0005] The main objective of this application is to provide a method, apparatus, device, and storage medium for generating three-dimensional models, in order to solve the problem of low accuracy in converting two-dimensional drawings into three-dimensional models in the prior art.

[0006] The first aspect of this invention provides a method for generating a three-dimensional model, the method comprising: loading a two-dimensional drawing using architectural three-dimensional modeling software; identifying the layer names of all layers in the two-dimensional drawing; and determining the component type of each component corresponding to each layer based on the layer names; querying a corresponding drawing recognition method and a corresponding information parsing method from a component recognition library in the architectural three-dimensional modeling software based on the component type; extracting a drawing feature group corresponding to the component from the corresponding layer based on the drawing recognition method; parsing the drawing feature group according to the information parsing method to obtain modeling information corresponding to the component; and performing three-dimensional modeling based on the modeling information corresponding to each component to obtain a three-dimensional model corresponding to the two-dimensional drawing.

[0007] Optionally, in a first implementation of the first aspect of the present invention, the drawing feature group includes at least line segment combinations and feature points. The line segment combination is a combination of line segment elements, and the line segment elements include line segments, curves with endpoints, and closed curves. Extracting the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method includes: extracting geometric information from the layer using the architectural 3D modeling software, wherein the geometric information includes points and lines; filtering out lines with two points as endpoints from the lines to obtain line segments and curves with endpoints; filtering out closed curves from the lines; and extracting... The positional relationship between all the line segment elements; determining the corresponding positional conditions from a preset positional condition dataset according to the drawing recognition method; determining whether the positional relationship between the line segment elements satisfies the positional conditions, and if so, extracting the corresponding line segment elements; extracting the intersection points and inflection points in the line segment combination according to the positional relationship between the line segment elements; calculating the geometric center of the line segment combination; determining the corresponding selection conditions from a preset feature selection dataset according to the drawing recognition method; determining whether the intersection points, inflection points, and geometric centers satisfy the selection conditions, and if so, recording them as feature points.

[0008] Optionally, in a second implementation of the first aspect of the present invention, the drawing feature group further includes a graphic element type and text annotation, wherein the text annotation includes an annotation object and annotation content. The step of extracting the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method further includes: extracting text information and graphic element types corresponding to all graphic elements in the layer using the architectural 3D modeling software, wherein the text information includes characters, identifiers, and the positions of the characters and identifiers in the layer; searching in the layer for the line segment combination or feature point closest to the position; using the closest line segment combination or feature point as the annotation object, and the corresponding character and the corresponding identifier as the annotation content, to generate the text annotation.

[0009] Optionally, in a third implementation of the first aspect of the present invention, the modeling information includes at least shape information and position information. The step of parsing the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component includes: determining a corresponding shape information family in a preset shape information dataset according to the corresponding information parsing method, and searching the shape information family based on the line segment combination to obtain the matched shape information; determining a position information generation method according to the corresponding information parsing method, and generating corresponding position information based on the shape information and the corresponding feature points according to the position information generation method.

[0010] Optionally, in a fourth implementation of the first aspect of the present invention, the modeling information further includes dimension information and cross-sectional information. The step of parsing the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component further includes: determining a dimension parameter conversion method according to the corresponding information parsing method, and converting the annotation content into parameter values ​​corresponding to the annotation object according to the dimension parameter conversion method; generating dimension information based on the annotation object and the corresponding parameter values; determining a corresponding cross-sectional generation method family in a preset cross-sectional generation method dataset according to the corresponding information parsing method; determining the primitive type extracted from the location information, and determining the corresponding cross-sectional generation method in the cross-sectional generation method family according to the primitive type; and generating corresponding cross-sectional information based on the primitive and parameter values ​​according to the cross-sectional generation method.

[0011] Optionally, in a fifth implementation of the first aspect of the present invention, the step of performing three-dimensional modeling based on the modeling information corresponding to each of the components to obtain a three-dimensional model corresponding to the two-dimensional drawing includes: determining the three-dimensional graphics corresponding to each component in a preset three-dimensional graphics dataset according to the shape information; determining the positional relationship between all the three-dimensional graphics according to the position information; and generating a three-dimensional model corresponding to the two-dimensional drawing based on the positional relationship between the three-dimensional graphics.

[0012] Optionally, in a sixth implementation of the first aspect of the present invention, generating a three-dimensional model corresponding to the two-dimensional drawing based on the positional relationship between the three-dimensional graphics includes: configuring the three-dimensional graphics according to the size information and the cross-sectional information to obtain a component model; determining the positional relationship between all the component models according to the positional information; and generating a three-dimensional model corresponding to the two-dimensional drawing based on the positional relationship between the component models.

[0013] A second aspect of the present invention provides a three-dimensional model generation apparatus, comprising: a determining module, configured to load two-dimensional drawings using architectural three-dimensional modeling software, identify the layer names of all layers in the two-dimensional drawings, and determine the component type of each component corresponding to each layer based on the layer names; a querying module, configured to query a corresponding drawing recognition method and a corresponding information parsing method from a component recognition library in the architectural three-dimensional modeling software based on the component type; a first extraction module, configured to extract the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method; a first parsing module, configured to parse the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component; and a first modeling module, configured to perform three-dimensional modeling based on the modeling information corresponding to each component to obtain a three-dimensional model corresponding to the two-dimensional drawings.

[0014] Optionally, in a first implementation of the second aspect of the present invention, the first extraction module includes: an extraction unit, configured to extract geometric information from the layer using the architectural 3D modeling software, wherein the geometric information includes points and lines; a first extraction unit, configured to filter out lines with two points as endpoints from the lines to obtain line segments and curves with endpoints; and to filter out closed curves from the lines; a second extraction unit, configured to extract the positional relationships between all the line segment elements; a third extraction unit, configured to determine corresponding positional conditions from a preset positional condition dataset according to the drawing recognition method; and to determine whether the positional relationships between the line segment elements satisfy the positional conditions, and if so, to extract the corresponding line segment elements; a fourth extraction unit, configured to extract the intersection points and inflection points in the line segment combination according to the positional relationships between the line segment elements; a calculation unit, configured to calculate the geometric center of the line segment combination; and a selection unit, configured to determine corresponding selection conditions from a preset feature selection dataset according to the drawing recognition method; and to determine whether the intersection points, inflection points, and geometric centers satisfy the selection conditions, and if so, to record them as feature points.

[0015] Optionally, in a second implementation of the second aspect of the present invention, the apparatus further includes a second extraction module, configured to extract text information and the corresponding element types of all elements in the layer using the architectural 3D modeling software, wherein the text information includes characters, identifiers, and the positions of the characters and identifiers in the layer; search the layer for the line segment combination or feature point closest to the position; use the closest line segment combination or feature point as the annotation object, and use the corresponding character and the corresponding identifier as the annotation content to generate the text annotation.

[0016] Optionally, in a third implementation of the second aspect of the present invention, the first parsing module includes: a matching unit, configured to determine a corresponding shape information family in a preset shape information dataset according to the corresponding information parsing method, and search the shape information family based on the line segment combination to obtain the matched shape information; and a first generation unit, configured to determine a position information generation method according to the corresponding information parsing method, and generate corresponding position information based on the shape information and the corresponding feature site according to the position information generation method.

[0017] Optionally, in a fourth implementation of the second aspect of the present invention, the apparatus further includes a second parsing module, configured to determine a size parameter conversion method according to the corresponding information parsing method, and convert the annotation content into parameter values ​​corresponding to the annotation object according to the size parameter conversion method, and generate size information based on the annotation object and the corresponding parameter values; determine a corresponding cross-section generation method family in a preset cross-section generation method dataset according to the corresponding information parsing method; determine the primitive type extracted from the location information, and determine the corresponding cross-section generation method in the cross-section generation method family according to the primitive type; and generate corresponding cross-section information based on the primitive and parameter values ​​according to the cross-section generation method.

[0018] Optionally, in a fifth implementation of the second aspect of the present invention, the first modeling module includes: a first determining unit, configured to determine the three-dimensional graphics corresponding to each component in a preset three-dimensional graphics dataset based on the shape information; a second determining unit, configured to determine the positional relationship between all the three-dimensional graphics based on the positional relationship between the three-dimensional graphics; and a second generating unit, configured to generate a three-dimensional model corresponding to the two-dimensional drawing based on the three-dimensional graphics according to the positional relationship between the three-dimensional graphics.

[0019] Optionally, in a sixth implementation of the second aspect of the present invention, the device further includes a second modeling module, configured to configure the three-dimensional graphic according to the size information and the cross-sectional information to obtain a component model; determine the positional relationship between all the component models according to the position information; and generate a three-dimensional model corresponding to the two-dimensional drawing based on the positional relationship between the component models.

[0020] A third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the computer device to perform the various steps of the above-described three-dimensional model generation method.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the various steps of the above-described three-dimensional model generation method. Specifically, the method involves loading two-dimensional drawings using architectural three-dimensional modeling software, identifying the layer names of all layers in the two-dimensional drawings, and determining the component type of each component corresponding to each layer based on the layer names; determining the corresponding drawing recognition method and information parsing method based on the component type; extracting the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method; parsing the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component; and performing three-dimensional modeling based on the modeling information corresponding to each component to obtain a three-dimensional model corresponding to the two-dimensional drawings. This solves the problem of low accuracy in converting two-dimensional drawings into three-dimensional models in the prior art.

[0022] In the technical solution of this invention, the method specifically involves loading two-dimensional drawings using architectural 3D modeling software, identifying the layer names of all layers in the two-dimensional drawings, and determining the component type of each component corresponding to each layer based on the layer names; determining the corresponding drawing recognition method and information parsing method based on the component type; extracting the drawing feature groups corresponding to the component from the corresponding layers according to the drawing recognition method; parsing the drawing feature groups according to the information parsing method to obtain the modeling information corresponding to the component; and performing 3D modeling based on the modeling information corresponding to each component to obtain a 3D model corresponding to the two-dimensional drawings. During the recognition process, the corresponding drawing recognition method and information parsing method are determined according to the different component types, identifying line segment combinations, feature points, primitive types, and text annotations in the layers, and converting them into shape information, position information, size information, and cross-sectional information corresponding to the component. The shape information, position information, size information, and cross-sectional information are then used for 3D modeling, improving accuracy and thus solving the problem of low accuracy in converting two-dimensional drawings into 3D models in the prior art. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the first embodiment of the three-dimensional model generation method in this invention;

[0024] Figure 2 This is a schematic diagram of a second embodiment of the three-dimensional model generation method in this invention;

[0025] Figure 3 This is a schematic diagram of the third embodiment of the three-dimensional model generation method in this invention;

[0026] Figure 4This is a schematic diagram of one embodiment of the three-dimensional model generation device in this invention;

[0027] Figure 5 This is a schematic diagram of another embodiment of the three-dimensional model generation device in this invention;

[0028] Figure 6 This is a schematic diagram of one embodiment of the computer device in this invention. Detailed Implementation

[0029] To address the low accuracy issue in converting 2D drawings into 3D models in existing technologies, this application provides a 3D model generation method, apparatus, device, and storage medium. The method loads 2D drawings using architectural 3D modeling software, identifies the layer names of all layers in the 2D drawings, and determines the component type corresponding to each layer based on the layer names. It then determines the corresponding drawing recognition method and information parsing method based on the component type. According to the drawing recognition method, it extracts the drawing feature groups corresponding to the component from the corresponding layers. According to the information parsing method, it parses the drawing feature groups to obtain the modeling information corresponding to the component. Based on the modeling information corresponding to each component, it performs 3D modeling to obtain a 3D model corresponding to the 2D drawing. During the recognition process, the corresponding drawing recognition method and information parsing method are determined according to the different component types. This identifies line segment combinations, feature points, primitive types, and text annotations in the layers and converts them into shape information, position information, size information, and cross-sectional information corresponding to the component. The shape information, position information, size information, and cross-sectional information are then used for 3D modeling, improving accuracy and thus solving the problem of low accuracy in converting 2D drawings into 3D models in existing technologies.

[0030] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1The first embodiment of the three-dimensional model generation method in this invention includes the following implementation steps:

[0032] 101. Load two-dimensional drawings using architectural 3D modeling software, identify the layer names of all layers in the two-dimensional drawings, and determine the component type of each component corresponding to each layer based on the layer name;

[0033] In this step, the architectural 3D modeling software includes at least Revit, which is the name of a series of software from Autodesk.

[0034] In this step, the two-dimensional drawings include at least CAD (Computer Aided Design) drawings.

[0035] 102. Based on the component type, query the corresponding drawing recognition method and the corresponding information parsing method from the component recognition library in the architectural 3D modeling software;

[0036] In this step, the component types include at least columns, walls, beams, and slabs;

[0037] In this step, the component identification library includes at least the drawing identification method and the corresponding information parsing method for the column, the wall, the beam and the slab.

[0038] 103. Based on the drawing recognition method, extract the drawing feature groups corresponding to the components from the corresponding layers;

[0039] In this step, the drawing feature group includes line segment combinations and feature points. The line segment combination is a combination of line segment elements, and the line segment elements include line segments, curves with endpoints, and closed curves.

[0040] This step can be implemented in the following way:

[0041] The point and line primitives in the layer are read using the architectural 3D modeling software to obtain points and lines;

[0042] The continuity features of the line graph elements are extracted using the architectural 3D modeling software.

[0043] Based on the continuity feature, lines with the two points as endpoints are selected from the lines to obtain the line segments and the curves with endpoints;

[0044] Based on the continuity characteristic, the closed curves are selected from the lines;

[0045] The position coordinates of the point primitives and line primitives are extracted using the architectural 3D modeling software, and the positional relationships between all the line segment elements are determined based on the position coordinates.

[0046] The corresponding location conditions are determined from a preset location condition dataset according to the drawing recognition method.

[0047] Determine whether the positional relationship between the line segment elements satisfies the positional condition. If it does, extract the corresponding line segment elements. For example, if the positional condition is intersection, extract the line segment elements that have an intersection relationship.

[0048] Based on the positional relationship between the line segment elements, extract the intersection points and inflection points in the line segment combination;

[0049] Extract the position coordinates of all points in the line segment combination, and calculate the geometric center of the line segment combination based on the position coordinates of all points;

[0050] Based on the drawing recognition method, the corresponding selection conditions are determined from the preset feature selection dataset;

[0051] Determine whether the intersection point, the inflection point, and the geometric center satisfy the selection conditions. If they do, they are recorded as feature points. For example, if the selection condition is that the point belongs to the intersection point and the inflection point and is closest to the geometric center, then select the point among the intersection point and the inflection point that is closest to the geometric center.

[0052] 104. Based on the information parsing method, parse the feature groups of the drawing to obtain the modeling information corresponding to the components;

[0053] In this step, the modeling information includes shape information and position information, wherein the shape information includes a shape diagram and the position information includes positioning points and positioning range;

[0054] This step can be implemented in the following way:

[0055] Based on the corresponding information parsing method, the corresponding shape information family is determined in the preset shape information dataset. For example, if the information parsing method is the information parsing method corresponding to a column, then the corresponding shape information family is determined to be the column shape information family.

[0056] The line segment combination is compared with all shape diagrams in the shape information family to obtain the comparison result;

[0057] Extract the shape image with the highest similarity from the comparison results;

[0058] The location information generation method is determined according to the corresponding information parsing method. For example, if the information parsing method is the information parsing method corresponding to the beam, then the location information generation method is determined to be the location information generation method corresponding to the beam.

[0059] According to the location information generation method, a corresponding positioning point is generated based on the feature point, and a corresponding location range is generated based on the positioning point and the shape diagram. For example, if the location information generation method is the location information generation method for a beam, then the feature point is converted into the positioning point of the beam, and the shape diagram is converted into the location range of the beam based on the positioning point. The location information corresponding to the beam is generated based on the location range of the beam and the positioning point of the beam.

[0060] 105. Perform 3D modeling based on the modeling information corresponding to each component to obtain a 3D model corresponding to the 2D drawings.

[0061] This step can be implemented in the following way:

[0062] The modeling information corresponding to each component is parsed to obtain the shape diagram, the positioning point, and the position range corresponding to each component;

[0063] Using the shape image as an index, a search is performed in a preset 3D graphic dataset, and the matching 3D graphic is obtained;

[0064] The positioning points are converted into corresponding three-dimensional coordinates in a preset three-dimensional modeling space;

[0065] Using the three-dimensional coordinates as a reference, the position range is projected onto the three-dimensional modeling space to obtain the projection range;

[0066] The three-dimensional graphic is output to the three-dimensional modeling space according to the projection range to obtain the three-dimensional model.

[0067] By implementing the above method, a two-dimensional drawing is loaded using architectural 3D modeling software. The layer names of all layers in the two-dimensional drawing are identified, and the component type corresponding to each layer is determined based on the layer names. A corresponding drawing recognition method and information parsing method are determined based on the component type. According to the drawing recognition method, the drawing feature group corresponding to the component is extracted from the corresponding layer. According to the information parsing method, the drawing feature group is parsed to obtain the modeling information corresponding to the component. 3D modeling is performed based on the modeling information corresponding to each component to obtain a 3D model corresponding to the two-dimensional drawing. Through the drawing recognition method corresponding to the component type, line segment combinations are identified by processing point and line primitives, and the corresponding shape diagram is searched in a preset dataset to obtain shape information. Based on the corresponding information parsing method, the selection method for feature points is determined, enabling the selection of feature points matching the component type and converting them into corresponding positioning points and location ranges to obtain location information. 3D modeling is performed based on the location and shape information. The shape diagram determines the 3D graphics, and the location information determines the position of each 3D graphic in the 3D modeling space. This improves the accuracy of modeling and solves the problem of low accuracy in converting two-dimensional drawings into 3D models in existing technologies.

[0068] Please see Figure 2 The second embodiment of the three-dimensional model generation method in this invention comprises the following steps:

[0069] 201. Load two-dimensional drawings using architectural 3D modeling software, identify the layer information of all layers in the two-dimensional drawings, and determine the component type of each component corresponding to each layer based on the layer information.

[0070] In this step, the process of determining the component type of each component corresponding to each layer based on the layer information includes:

[0071] Extract the layer identifier from the layer information;

[0072] Using the layer identifier as an index, a search is performed in the preset component type dataset to obtain the matching component type.

[0073] 202. Determine the corresponding drawing recognition method and information parsing method based on the component type;

[0074] This step can be implemented in the following way:

[0075] Based on the layer identifier, determine the corresponding drawing recognition method from the preset drawing recognition method table;

[0076] Based on the layer identifier, the corresponding information parsing method is determined from the preset information parsing method table.

[0077] 203. Based on the corresponding drawing recognition method, extract the line segment combinations and feature points corresponding to the components from the corresponding layers;

[0078] This step is basically the same as step 103 in the previous embodiment, so it will not be repeated here.

[0079] 204. Based on the corresponding drawing recognition method, extract the corresponding element type and text annotation of the component from the corresponding layer;

[0080] In this step, the text annotation includes the annotation object and the annotation content;

[0081] This step can be implemented in the following way:

[0082] The text information and the corresponding element types of all elements in the layer are extracted using the architectural 3D modeling software. The text information includes characters, strings, and identifiers.

[0083] Extract the positions of the characters, strings, and identifiers in the layer from the layer information, and denote them as the first position set;

[0084] Extract the line segment combination and the position of the feature point in the layer from the layer information, and denote it as the second position set;

[0085] Search for the location point closest to the first location set from the second location set, and extract the line segment combination or feature point corresponding to the closest location point;

[0086] The text annotation is generated by taking the nearest line segment combination or feature point as the annotation object and the corresponding character, string and identifier as the annotation content.

[0087] In practical applications, the process of extracting text information and the corresponding element types of all elements from the layer using the architectural 3D modeling software includes:

[0088] Extract the positioning lines from the layer, and traverse the closed shape enclosed by the positioning lines to obtain all the primitives in the closed shape;

[0089] Obtain the primitive types corresponding to all primitives in the closed figure, and determine the corresponding primitive parsing method based on the primitive types corresponding to all primitives in the closed figure;

[0090] According to the graphic element parsing method, text information is read from the graphic elements.

[0091] 205. Based on the corresponding information parsing method, analyze the line segment combination and feature points to obtain the shape and position information of the component;

[0092] This step is basically the same as step 104 in the previous embodiment, so it will not be repeated here.

[0093] 206. Based on the corresponding information parsing method, parse the element type and text annotation to obtain the corresponding dimension information and cross-sectional information of the component;

[0094] This step can be implemented in the following way:

[0095] The size parameter conversion method is determined according to the corresponding information parsing method, and the annotation content is converted into the parameter value corresponding to the annotation object according to the size parameter conversion method. Size information is generated based on the annotation object and the corresponding parameter value. For example, the value of the annotation content is read and converted into the parameter value corresponding to the standard unit system. The annotation object and the corresponding parameter value are saved according to the correspondence relationship to obtain the size information.

[0096] Based on the corresponding information parsing method, determine the corresponding cross-section generation method family in the preset cross-section generation method dataset;

[0097] Determine the type of graphic element extracted from the location information, determine the corresponding cross-section generation method in the cross-section generation method family based on the graphic element type, and generate the corresponding cross-section information based on the graphic element and parameter value according to the cross-section generation method.

[0098] Further, the process of determining the primitive type extracted from the location information, determining the corresponding cross-section generation method in the cross-section generation method family based on the primitive type, and generating corresponding cross-section information based on the primitive and parameter values ​​according to the cross-section generation method includes:

[0099] If the primitive type is the primitive type corresponding to a curve graphic;

[0100] Extract cross-sectional feature points from the primitives, such as the center of a circle;

[0101] Text annotations are extracted from the graphic elements, and corresponding cross-sectional information is generated based on the cross-sectional feature points and the text annotations. For example, if the annotation object of the text annotation is a circle, a circle is constructed according to the center of the circle to obtain the cross-section, and the corresponding annotation content is read as the radius of the circle to obtain the cross-sectional information.

[0102] If the primitive type is the primitive type corresponding to a polygon;

[0103] Extract inflection points from the primitives;

[0104] Based on the inflection point, the corresponding line segment is queried in the line segment combination;

[0105] The corresponding parameter value is retrieved from the size information based on the line segment.

[0106] A cross section is generated based on the inflection point and the line segment corresponding to the inflection point, and the cross section size is generated based on the corresponding parameter value. For example, the line segment corresponding to the inflection point is used as the side, and the parameter value corresponding to the side is used as the side length to generate the corresponding polygon, thus obtaining the cross section.

[0107] The cross-sectional information is generated based on the cross-section and cross-sectional dimensions.

[0108] 207. Generate drawing feature groups based on line segment combinations, feature points, graphic element types, and text annotations, and generate modeling information based on shape information, location information, size information, and cross-sectional information;

[0109] This step can be implemented in the following way:

[0110] The line segment combination, feature point, graphic element type, and text annotation are analyzed to obtain the components corresponding to the line segment combination, feature point, graphic element type, and text annotation respectively;

[0111] The line segment combination, feature point, graphic element type, and text annotation corresponding to the same component are saved according to the correspondence to obtain the drawing feature group;

[0112] The shape information, position information, size information, and cross-sectional information are parsed to obtain the components corresponding to the shape information, position information, size information, and cross-sectional information, respectively.

[0113] The shape information, position information, size information, and cross-sectional information corresponding to the same component are saved according to their correspondence to obtain the modeling information.

[0114] 208. Output the drawing feature groups and modeling information using the preset method;

[0115] This step can be implemented in the following way:

[0116] The drawing feature group is output in a preset recognition preview interface. For example, a recognition preview image is generated based on the line segment combination, the feature point, the graphic element type, and the text annotation, and then output through the recognition preview interface.

[0117] The modeling information is used to output the results in a preset analysis preview interface. For example, a recognition analysis preview image is generated based on the shape information, position information, size information, and cross-sectional information, and then output through the analysis preview interface.

[0118] In practical applications, this step also includes:

[0119] Read the user's interaction commands;

[0120] The corresponding line segment combination, feature point, graphic element type, and text label are modified according to the interactive instructions;

[0121] The corresponding shape information, position information, size information, and cross-sectional information are modified according to the interaction instructions.

[0122] 209. Perform 3D modeling based on the modeling information corresponding to each component to obtain a 3D model corresponding to the 2D drawings.

[0123] In this step, the modeling information includes the shape information, the position information, the size information, and the cross-sectional information, wherein the shape information includes a shape diagram, the position information includes position points, the size information includes parameter values, and the cross-sectional information includes a cross-section and cross-sectional dimensions;

[0124] This step can be implemented in the following way:

[0125] The modeling information corresponding to each component is parsed to obtain the shape diagram, the location, the parameter value, the cross section, and the cross section size corresponding to each component.

[0126] Based on the shape diagram and cross-section, the corresponding 3D graphics are generated using the architectural 3D modeling software.

[0127] Based on the parameter values ​​and the cross-sectional dimensions, the three-dimensional graphic is configured to obtain a component model. For example, according to a preset scale, the parameter values ​​and the cross-sectional dimensions are converted into the modeling dimensions corresponding to the three-dimensional graphic, and the component model is generated based on the three-dimensional graphic according to the modeling dimensions.

[0128] The positioning points are converted into corresponding three-dimensional coordinates in a preset three-dimensional modeling space;

[0129] Based on the three-dimensional coordinates, the positional relationship between each component model is determined according to the parameter values ​​and the cross-sectional dimensions, wherein the positional relationship between each component model includes at least contact and non-contact;

[0130] Based on the three-dimensional coordinates and the positional relationship between the component models, the component models are output in the three-dimensional modeling space to obtain the corresponding three-dimensional models.

[0131] By implementing the above method, a two-dimensional drawing is loaded using architectural 3D modeling software. The layer names of all layers in the two-dimensional drawing are identified, and the component type corresponding to each layer is determined based on the layer names. The corresponding drawing recognition method and information parsing method are determined based on the component type. According to the drawing recognition method, the drawing feature group corresponding to the component is extracted from the corresponding layer. According to the information parsing method, the drawing feature group is parsed to obtain the modeling information corresponding to the component. 3D modeling is performed based on the modeling information corresponding to each component to obtain a 3D model corresponding to the two-dimensional drawing. The method involves loading a two-dimensional drawing using architectural 3D modeling software, identifying the layer information of all layers in the two-dimensional drawing, and determining the component type corresponding to each layer based on the layer information. The corresponding drawing recognition method and information parsing method are determined based on the component type. According to the corresponding drawing recognition method, the line segment combination, feature point, primitive type, and text annotation corresponding to the component are extracted from the corresponding layer. According to the corresponding information parsing method, the line segment combination and feature point are... The process involves parsing the drawing to obtain the shape and position information of the components. Based on the corresponding information parsing method, the element types and text annotations are parsed to obtain the corresponding dimension and cross-sectional information of the components. Drawing feature groups are generated based on line segment combinations, feature points, element types, and text annotations, and modeling information is generated based on the shape, position, dimension, and cross-sectional information. The drawing feature groups and modeling information are output using a preset method. Three-dimensional modeling is performed based on the modeling information corresponding to each component, resulting in a three-dimensional model corresponding to the two-dimensional drawing. By selecting the drawing recognition and information parsing method corresponding to the component type, shape and position information are obtained based on the recognized line segment combinations and feature points, and dimension and cross-sectional information are obtained based on the recognized element types and text annotations. This improves the accuracy of recognition and information extraction. Furthermore, modeling information is generated based on the shape, position, dimension, and cross-sectional information, and three-dimensional modeling is performed based on this modeling information. This improves the accuracy of converting two-dimensional drawings into three-dimensional models, thus solving the problem of low accuracy in converting two-dimensional drawings into three-dimensional models in existing technologies.

[0132] Please see Figure 3 The third embodiment of the three-dimensional model generation method in this invention comprises the following steps:

[0133] 301. Load CAD drawings into Revit software and identify the layer information of all layers in the CAD drawings;

[0134] In this step, the Revit software includes Autodesk Revit Architecture, Autodesk Revit MEP, and Autodesk Revit Structure, where Revit is the name of a series of software from Autodesk.

[0135] In this step, the layer information includes the layer location, layer type, and layer name.

[0136] 302. Determine the component type of each layer based on the layer information;

[0137] In this step, the component types include at least columns, walls, beams, and slabs;

[0138] This step can be implemented in the following way:

[0139] A layer index is generated based on the layer position, the layer type, and the layer name;

[0140] Search for the corresponding component type in the preset component type dataset based on the layer index.

[0141] In practical applications, the component type dataset includes the following fields:

[0142] In CAD, the layer name for a wall is: WallLayer (default: JG-shear wall);

[0143] In CAD, the layer name for a column is: ColumnLayer (default: JG-Structural Column);

[0144] In CAD, the layer name for beams is: BeamLayer (default: JG-Structural Beam);

[0145] The layer containing the beam section label (BL) (default: JG-beam number);

[0146] The layer containing the beam attribute labels is SX (default: JG-beam number);

[0147] The layer containing the exposed water and electricity trenches is SD (default: exposed water and electricity trenches);

[0148] The layer containing the board elevation label is FG (default: JG-board elevation);

[0149] The layer containing the plate section label is FT (default: JG-plate thickness);

[0150] The layer containing the opening symbol line is FH (default: JG-floor slab opening).

[0151] 303. Determine the corresponding drawing recognition method based on the component type, extract the drawing feature groups from the layer based on the drawing recognition method, and convert them into the modeling information corresponding to the component;

[0152] This step can be implemented in the following way:

[0153] If the component type is a column, then the drawing feature group corresponding to the column includes the line segment combination, the element type, and the feature point, wherein the feature point includes the center point;

[0154] If the component type is a column, the modeling information corresponding to the column includes at least the shape information, the position information, and the cross-sectional information, wherein the shape information includes a graphic and a graphic type, the position information includes a positioning point, and the cross-sectional information includes a cross-sectional graphic and a cross-sectional dimension;

[0155] The line segments with connections are extracted from the layer using the Revit software to obtain the line segment combination;

[0156] Based on the line segment combination, a search is performed in the preset column graphic type dataset, and the matching graphic type is obtained;

[0157] The corresponding graphic transformation method is determined according to the graphic type, and the line segment combination is transformed into the corresponding graphic according to the graphic transformation method. For example, if the graphic type is a rectangular column, the vertices and edges in the line segment combination are extracted to obtain the graphic.

[0158] Extract the center point of the graphic and record it as the positioning point corresponding to the component;

[0159] Extract the element types of all elements from the layer using the Revit software;

[0160] The corresponding cross-section type is determined based on the element type, and the corresponding cross-section recognition method is determined based on the cross-section type. The corresponding cross-section graphic and cross-section dimensions are then recognized based on the cross-section graphic method.

[0161] Further, the process of determining the corresponding cross-section type based on the primitive type, determining the corresponding cross-section recognition method based on the cross-section type, and recognizing the corresponding cross-section shape and cross-section size based on the cross-section shape method includes:

[0162] If the primitive type is CIRCLE, then the corresponding cross-section type is determined to be a circular cross-section, wherein the cross-section dimension corresponding to the circular cross-section is the radius;

[0163] Search for characters or strings in the primitives, and read the radius from the characters or strings;

[0164] If the primitive type is POLYLINE, then the corresponding cross-section type is determined to be a polygonal cross-section, wherein the cross-section size corresponding to the polygonal cross-section is the length parameter value of each side;

[0165] Extract all inflection points from the given primitives;

[0166] Extract the line segment with the inflection point as the endpoint from the primitive, and denote it as an edge;

[0167] Search for the character closest to the edge from the primitives, and read the length parameter value corresponding to the edge from the character closest to the edge.

[0168] If the component type is a wall, then the drawing feature group corresponding to the wall includes the line segment combination and the feature point, wherein the feature point includes the center line;

[0169] If the component type is a wall, the modeling information corresponding to the wall includes at least the shape information, the position information, and the cross-sectional information, wherein the shape information includes graphics, the position information includes positioning lines, and the cross-sectional information includes cross-sectional dimensions;

[0170] The Revit software is used to extract line segments with connections from the layer to obtain the line segment combination, which is denoted as the line segment combination corresponding to the wall.

[0171] Based on the line segment combination corresponding to the wall, a search is performed in the preset wall graphic dataset, and the matched graphic is obtained and recorded as the graphic corresponding to the wall.

[0172] Generate the corresponding center line based on the graphic corresponding to the wall;

[0173] The position of the center line in the layer is determined using the Revit software to obtain the positioning line;

[0174] Calculate the geometric length of the graphic corresponding to the wall in the direction perpendicular to the positioning line;

[0175] According to the preset length conversion method, the geometric length of the graphic corresponding to the wall is converted into the cross-sectional dimensions.

[0176] In practical applications, before converting the geometric length of the wall-corresponding graphic into cross-sectional dimensions according to a preset length conversion method, this step further includes:

[0177] The Revit software is used to search for characters, strings, or identifiers in the layers to obtain search results;

[0178] The search results are used to determine whether they contain characters, strings, or identifiers.

[0179] If so, read the value of the character, string, or identifier and convert it into the corresponding length conversion method.

[0180] If the component type is a beam, then the drawing feature group corresponding to the column includes the text annotation, the graphic element type, and the feature point, wherein the text annotation includes beam section label and beam elevation label, the graphic element type includes text graphic element, and the feature point includes beam line, beam center line, and beam length range;

[0181] If the component type is a beam, then the modeling information corresponding to the column includes at least the location information and the cross-sectional information, wherein the location information includes the location range, lap structure, endpoints and elevation, and the cross-sectional information includes the cross-sectional dimensions;

[0182] The beam line, beam centerline, and beam length range corresponding to the beam are extracted from the layer using the Revit software.

[0183] The location range of the beam is determined based on the beam line, center line, and beam length range corresponding to the beam.

[0184] The overlapping structure within the specified location range is extracted using the Revit software.

[0185] Based on the overlapping structure, the corresponding endpoint generation method is determined from the preset endpoint generation method set, and based on the endpoint generation method, the endpoints corresponding to the beam are determined based on the beam line, center line and beam length range corresponding to the beam.

[0186] The Revit software searches for characters or strings in the layer, reads beam section labels and beam elevation labels from the characters or strings, converts the beam section labels into the corresponding beam section dimensions, and converts the beam elevation labels into the corresponding beam elevations.

[0187] Further, the process of determining the corresponding endpoint generation method from a preset set of endpoint generation methods based on the overlapping structure, and determining the endpoints corresponding to the beam based on the beam line, center line, and beam length range corresponding to the beam according to the endpoint determination method, includes:

[0188] If the lap structure is a column, then the endpoint determination method is determined to be the endpoint determination method corresponding to the column, and according to the endpoint determination method corresponding to the column, the projection point of the positioning point corresponding to the column in the beam line direction is determined to be the endpoint corresponding to the beam.

[0189] If the overlapping structure is the outer plane of a shear wall, then the endpoint determination method is determined to be the endpoint determination method corresponding to the outer plane of the shear wall, and according to the endpoint determination method corresponding to the outer plane of the shear wall, the intersection of the beam centerline or the extension of the beam centerline and the centerline corresponding to the shear wall is determined to be the endpoint of the beam.

[0190] If the overlapping structure is the inner plane of a shear wall, then the endpoint determination method is determined to be the endpoint determination method corresponding to the inner plane of the shear wall, and according to the endpoint determination method corresponding to the inner plane of the shear wall, the contact point between the beam centerline and the shear wall is determined to be the endpoint corresponding to the beam.

[0191] In practical applications, if multiple overlapping structures are included, the overlapping structures are determined according to the following priority, and the corresponding endpoint determination method is determined based on the overlapping structures:

[0192] First priority: the column;

[0193] Second priority: the outer plane of the shear wall;

[0194] Third priority: the inner plane of the shear wall;

[0195] For example, if beam LA overlaps with both the outer plane A and the inner plane B of the shear wall, then the method for determining the endpoint of beam LA is the same as the method for determining the endpoint of the outer plane of the shear wall.

[0196] Further, the step of searching for characters or strings in the layer using the Revit software, reading beam section labels and beam elevation labels from the characters or strings, converting the beam section labels into the corresponding beam section dimensions, and converting the beam elevation labels into the corresponding beam elevations includes:

[0197] The Revit software is used to extract strings containing "x" or "X" from the layer and convert them into the form of "beam width x beam height" to obtain the cross-sectional dimensions of the beam.

[0198] The Revit software is used to extract strings containing "(" and ")" from the layer and convert them into the form "(Hs elevation)" to obtain the elevation corresponding to the beam.

[0199] In practical applications, the process of searching for characters or strings in the layer using the Revit software, reading beam section labels and beam elevation labels from the characters or strings, converting the beam section labels into the corresponding beam section dimensions, and converting the beam elevation labels into the corresponding beam elevations, further includes:

[0200] Search for text elements in the layer using the Revit software;

[0201] Read the value of the first searched text element to obtain the beam section label or beam elevation label.

[0202] If the component type is a plate, then the drawing feature group corresponding to the column includes the text annotation, the graphic element type, and the feature point, wherein the text annotation includes text graphic elements, the graphic element type includes text graphic element types, and the feature point includes positioning lines;

[0203] If the component type is a plate, then the modeling information corresponding to the column includes at least the shape information, the position information, and the size information, wherein the shape information includes a closed shape, the position information includes a minimum closed area and a side line, and the size information includes an offset value and a thickness;

[0204] The positioning lines corresponding to the board are extracted from the layer using the Revit software to obtain the closed shape enclosed by the positioning lines corresponding to the board.

[0205] Search for text primitives within the closed graphic and parse the text primitive type corresponding to the text primitive;

[0206] Based on the text element type, read the size information from the text element;

[0207] The Revit software searches all layers for the component closest to the positioning line corresponding to the board, and extracts the smallest enclosed area enclosed by the nearest component.

[0208] Extract the boundary of the smallest closed region to obtain the edge line corresponding to the plate.

[0209] Further, the step of reading size information from the text primitive according to the text primitive type includes:

[0210] If the text primitive type is FG, then read the offset value from the text primitive, wherein the offset value includes at least the Z-axis offset value;

[0211] If the text primitive type is FT, then the thickness is read from the text primitive, wherein the thickness includes at least the thickness of the floor slab.

[0212] 304. Load the modeling information into Revit software and perform automatic modeling using Revit software.

[0213] In practical applications, the modeling information also includes an elevation system, constraint layers, and offset heights. The elevation system includes a structural elevation system and a building elevation system. The constraint layers include a top constraint layer and a bottom constraint layer. The offset heights include a top offset height and a bottom offset height.

[0214] In practical applications, the process of loading modeling information into Revit software also includes:

[0215] The elevation system, constraint layer, and offset height are obtained through a pre-defined user interface.

[0216] By implementing the above method, a 2D drawing is loaded using architectural 3D modeling software. The layer names of all layers in the 2D drawing are identified, and the component type corresponding to each layer is determined based on the layer names. The corresponding drawing recognition method and information parsing method are determined based on the component type. According to the drawing recognition method, the drawing feature group corresponding to the component is extracted from the corresponding layer. According to the information parsing method, the drawing feature group is parsed to obtain the modeling information corresponding to the component. 3D modeling is performed based on the modeling information corresponding to each component to obtain a 3D model corresponding to the 2D drawing. In summary, CAD drawings are loaded using Revit software, and the layer information of all layers in the CAD drawing is identified. Based on the drawing... Layer information determines the component type of each layer; based on the component type, the corresponding drawing recognition method is determined; drawing feature groups are extracted from the layer based on the drawing recognition method and converted into modeling information corresponding to the component; the modeling information is loaded into Revit software, and automatic modeling is performed through Revit software; the corresponding drawing recognition method is determined by the component types such as beams, columns, slabs, and walls; after recognizing the drawing feature groups, they are converted into corresponding modeling information, such as recognizing feature points and line segment combinations and converting them into corresponding position and shape information. This can improve the accuracy of information recognition and conversion, thereby improving the accuracy of converting two-dimensional drawings into three-dimensional models, thus solving the problem of low accuracy in converting two-dimensional drawings into three-dimensional models in the existing technology.

[0217] The three-dimensional model generation method in the embodiments of the present invention has been described above. The three-dimensional model generation apparatus in the embodiments of the present invention will be described below. Please refer to... Figure 4 One embodiment of the three-dimensional model generation apparatus in this invention includes:

[0218] The determination module 401 is used to load two-dimensional drawings through the architectural three-dimensional modeling software, identify the layer names of all layers in the two-dimensional drawings, and determine the component type of the component corresponding to each layer based on the layer names.

[0219] The query module 402 is used to query the corresponding drawing recognition method and the corresponding information parsing method from the component recognition library in the architectural 3D modeling software according to the component type.

[0220] The first extraction module 403 is used to extract the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method;

[0221] The first parsing module 404 is used to parse the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component;

[0222] The first modeling module 405 is used to perform three-dimensional modeling based on the modeling information corresponding to each component, so as to obtain a three-dimensional model corresponding to the two-dimensional drawing.

[0223] By implementing the above-mentioned device, a two-dimensional drawing is loaded using architectural 3D modeling software. The layer names of all layers in the two-dimensional drawing are identified, and the component type corresponding to each layer is determined based on the layer names. A corresponding drawing recognition method and information parsing method are determined based on the component type. According to the drawing recognition method, the drawing feature group corresponding to the component is extracted from the corresponding layer. According to the information parsing method, the drawing feature group is parsed to obtain the modeling information corresponding to the component. Based on the modeling information corresponding to each component, 3D modeling is performed to obtain a 3D model corresponding to the two-dimensional drawing. This solves the problem of low accuracy in converting two-dimensional drawings into 3D models in the prior art.

[0224] Please see Figure 5 Another embodiment of the three-dimensional model generation device in this invention includes:

[0225] The determination module 401 is used to load two-dimensional drawings through the architectural three-dimensional modeling software, identify the layer names of all layers in the two-dimensional drawings, and determine the component type of the component corresponding to each layer based on the layer names.

[0226] The query module 402 is used to query the corresponding drawing recognition method and the corresponding information parsing method from the component recognition library in the architectural 3D modeling software according to the component type.

[0227] The first extraction module 403 is used to extract the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method;

[0228] The second extraction module 406 is used to extract text information and the corresponding element types of all elements in the layer using the architectural 3D modeling software. The text information includes characters, identifiers, and the positions of the characters and identifiers in the layer. The module searches the layer for the line segment combination or feature point that is closest to the position. The module uses the closest line segment combination or feature point as the annotation object and the corresponding character and identifier as the annotation content to generate the text annotation.

[0229] The first parsing module 404 is used to parse the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component;

[0230] The second parsing module 407 is used to determine the size parameter conversion method according to the corresponding information parsing method, and convert the annotation content into parameter values ​​corresponding to the annotation object according to the size parameter conversion method, and generate size information based on the annotation object and the corresponding parameter values; determine the corresponding cross-section generation method family in a preset cross-section generation method dataset according to the corresponding information parsing method; determine the primitive type extracted from the location information, and determine the corresponding cross-section generation method in the cross-section generation method family according to the primitive type; and generate corresponding cross-section information based on the primitive and parameter values ​​according to the cross-section generation method.

[0231] The first modeling module 405 is used to perform three-dimensional modeling based on the modeling information corresponding to each component, and obtain a three-dimensional model corresponding to the two-dimensional drawing.

[0232] The second modeling module 408 is used to configure the three-dimensional graphic according to the size information and the cross-sectional information to obtain a component model; determine the positional relationship between all the component models according to the position information; and generate a three-dimensional model corresponding to the two-dimensional drawing based on the positional relationship between the component models.

[0233] In this embodiment, the first extraction module 403 includes:

[0234] The first extraction unit 4031 is used to filter out lines from the lines that have two points as endpoints, to obtain line segments and curves with endpoints; and to filter out closed curves from the lines.

[0235] The second extraction unit 4032 is used to extract the positional relationships between all the line segment elements;

[0236] The third extraction unit 4033 is used to determine the corresponding position conditions from the preset position condition dataset according to the drawing recognition method; determine whether the positional relationship between the line segment elements satisfies the position conditions; if it does, extract the corresponding line segment elements.

[0237] The fourth extraction unit 4034 is used to extract the intersection points and inflection points in the line segment combination according to the positional relationship between the line segment elements;

[0238] The calculation unit 4035 is used to calculate the geometric center of the line segment combination;

[0239] The selection unit 4036 is used to determine the corresponding selection conditions from the preset feature selection dataset according to the drawing recognition method; and to determine whether the intersection, the inflection point and the geometric center satisfy the selection conditions. If they satisfy the conditions, they are recorded as feature points.

[0240] In this embodiment, the first parsing module 404 includes:

[0241] The matching unit 4041 is used to determine the corresponding shape information family in the preset shape information dataset according to the corresponding information parsing method, and to search the shape information family based on the line segment combination to obtain the matched shape information;

[0242] The first generation unit 4042 is used to determine the location information generation method according to the corresponding information parsing method, and generate corresponding location information based on the shape information and the corresponding feature sites according to the location information generation method.

[0243] In this embodiment, the first modeling module 405 includes:

[0244] The first determining unit 4051 is used to determine the three-dimensional graphics corresponding to each component in a preset three-dimensional graphics dataset based on the shape information.

[0245] The second determining unit 4052 is used to determine the positional relationship between all the three-dimensional graphics based on the position information;

[0246] The second generation unit 4053 is used to generate a three-dimensional model corresponding to the two-dimensional drawing based on the positional relationship between the three-dimensional graphics.

[0247] By implementing the above-mentioned device, a two-dimensional drawing is loaded using architectural 3D modeling software. The layer names of all layers in the two-dimensional drawing are identified, and the component type corresponding to each layer is determined based on the layer names. A corresponding drawing recognition method and information parsing method are determined based on the component type. According to the drawing recognition method, the drawing feature group corresponding to the component is extracted from the corresponding layer. According to the information parsing method, the drawing feature group is parsed to obtain the modeling information corresponding to the component. Based on the modeling information corresponding to each component, 3D modeling is performed to obtain a 3D model corresponding to the two-dimensional drawing. This solves the problem of low accuracy in converting two-dimensional drawings into 3D models in the prior art.

[0248] Please see Figure 6 The following is a detailed description of one embodiment of the computer device in this invention from the perspective of hardware processing.

[0249] Figure 6 This is a schematic diagram of the structure of a computer device 600 provided in an embodiment of the present invention. The computer device 600 can vary significantly due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 610 (e.g., one or more processors) and a memory 620, and one or more storage media 630 (e.g., one or more mass storage devices) for storing application programs 633 or data 632. The memory 620 and storage media 630 can be temporary or persistent storage. The program stored in the storage media 630 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the computer device 600. Furthermore, the processor 610 may be configured to communicate with the storage media 630 and execute the series of instruction operations in the storage media 630 on the computer device 600.

[0250] Computer device 600 may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input / output interfaces 660, and / or one or more operating systems 631, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 6 The computer device structure shown does not constitute a limitation on the computer device provided in this application. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0251] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the above-described three-dimensional model generation method.

[0252] In practical applications, the methods described above can be implemented based on artificial intelligence (AI) technology. AI is the theory, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results. Specifically, it can be executed on a server. The server can be a standalone server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and AI platforms.

[0253] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0254] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0255] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating a three-dimensional model, characterized in that, The three-dimensional model generation method is based on architectural three-dimensional modeling software, and the three-dimensional model generation method includes: The system loads two-dimensional drawings using architectural 3D modeling software, identifies the layer names of all layers in the two-dimensional drawings, and determines the component type of each component corresponding to each layer based on the layer names. Based on the component type, the corresponding drawing recognition method and the corresponding information parsing method are queried from the component recognition library in the architectural 3D modeling software; According to the drawing recognition method, the drawing feature group corresponding to the component is extracted from the corresponding layer. The drawing feature group includes at least line segment combinations and feature points. According to the information parsing method, the drawing feature group is parsed to obtain the modeling information corresponding to the component. The modeling information includes at least shape information and position information. Three-dimensional modeling is performed based on the modeling information corresponding to each component to obtain a three-dimensional model corresponding to the two-dimensional drawing. The step of parsing the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component includes: determining the corresponding shape information family in the preset shape information dataset according to the corresponding information parsing method, and searching the shape information family based on the line segment combination to obtain the matched shape information; determining the position information generation method according to the corresponding information parsing method, and generating the corresponding position information based on the shape information and the corresponding feature points according to the position information generation method.

2. The three-dimensional model generation method according to claim 1, characterized in that, The line segment combination is a combination of line segment elements, which include line segments, curves with endpoints, and closed curves. The step of extracting the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method includes: The geometric information in the layer is extracted using the architectural 3D modeling software, wherein the geometric information includes points and lines; From the lines, select the lines with the two points as endpoints to obtain line segments and curves with endpoints; Select closed curves from the lines; Extract the positional relationships between all the line segment elements; The corresponding location conditions are determined from a preset location condition dataset according to the drawing recognition method. Determine whether the positional relationship between the line segment elements satisfies the positional condition; if it does, extract the corresponding line segment elements. Based on the positional relationship between the line segment elements, extract the intersection points and inflection points in the line segment combination; Calculate the geometric center of the line segment combination; Based on the drawing recognition method, the corresponding selection conditions are determined from the preset feature selection dataset; Determine whether the intersection point, the inflection point, and the geometric center satisfy the selection conditions. If they do, record them as feature points.

3. The three-dimensional model generation method according to claim 2, characterized in that, The drawing feature group further includes element type and text annotation, wherein the text annotation includes annotation object and annotation content. The step of extracting the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method further includes: The text information and the corresponding element types of all elements in the layer are extracted using the architectural 3D modeling software. The text information includes characters, identifiers, and the positions of the characters and identifiers in the layer. Search the layer for the line segment combination or feature point that is closest to the location; The text annotation is generated by taking the nearest line segment combination or feature point as the annotation object and the corresponding character and the corresponding identifier as the annotation content.

4. The three-dimensional model generation method according to claim 3, characterized in that, The modeling information also includes dimensional information and cross-sectional information. The step of parsing the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component further includes: The size parameter conversion method is determined according to the corresponding information parsing method, and the annotation content is converted into the parameter value corresponding to the annotation object according to the size parameter conversion method. Size information is generated based on the annotation object and the corresponding parameter value. Based on the corresponding information parsing method, determine the corresponding cross-section generation method family in the preset cross-section generation method dataset; Based on the location information, the corresponding primitive type is extracted, and based on the primitive type, the corresponding cross-section generation method is determined in the cross-section generation method family; Based on the cross-section generation method, corresponding cross-section information is generated based on the graphic elements and parameter values.

5. The three-dimensional model generation method according to claim 4, characterized in that, The step of performing three-dimensional modeling based on the modeling information corresponding to each of the components to obtain a three-dimensional model corresponding to the two-dimensional drawing includes: Based on the shape information, determine the corresponding three-dimensional graphics of each component in a preset three-dimensional graphics dataset; Determine the positional relationships between all the three-dimensional graphics based on the position information; Based on the positional relationship between the three-dimensional graphics, a three-dimensional model corresponding to the two-dimensional drawing is generated.

6. The three-dimensional model generation method according to claim 5, characterized in that, The step of generating a 3D model corresponding to the 2D drawing based on the positional relationship between the 3D graphics includes: The three-dimensional graphic is configured based on the size information and the cross-sectional information to obtain the component model; Determine the positional relationships between all the component models based on the position information; Based on the positional relationship between the component models, a three-dimensional model corresponding to the two-dimensional drawing is generated.

7. A three-dimensional model generation device, characterized in that, The device includes: The determination module is used to load two-dimensional drawings through architectural 3D modeling software, identify the layer names of all layers in the two-dimensional drawings, and determine the component type of the component corresponding to each layer based on the layer names. The query module is used to query the corresponding drawing recognition method and the corresponding information parsing method from the component recognition library in the architectural 3D modeling software according to the component type; The extraction module is used to extract the drawing feature group corresponding to the component from the corresponding layer according to the drawing recognition method. The drawing feature group includes at least line segment combinations and feature points. The parsing module is used to parse the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component. The modeling information includes at least shape information and position information. The modeling module is used to perform three-dimensional modeling based on the modeling information corresponding to each component, and obtain a three-dimensional model corresponding to the two-dimensional drawing. The step of parsing the drawing feature group according to the information parsing method to obtain the modeling information corresponding to the component includes: determining the corresponding shape information family in the preset shape information dataset according to the corresponding information parsing method, and searching the shape information family based on the line segment combination to obtain the matched shape information; determining the position information generation method according to the corresponding information parsing method, and generating the corresponding position information based on the shape information and the corresponding feature points according to the position information generation method.

8. A computer device, characterized in that, include: A memory and at least one processor, wherein the memory stores instructions and the memory and the at least one processor are interconnected via a circuit; The at least one processor invokes the instructions in the memory to cause the computer device to perform the steps of the three-dimensional model generation method as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the three-dimensional model generation method as described in any one of claims 1-6.

Images