KR102065273B1 - Method and apparatus for automatically generating building element information by meaning for user-customized three-dimensional modeling service - Google Patents

Abstract

Disclosed are a method for automatically generating a meaningful building element information for a user-customized 3D modeling providing service and a device thereof. The method for automatically generating the meaningful building element information comprises: a step of calculating a normal vector of polygons constituting a building in a 3D building modeling file; a step of normalizing the calculated normal vector; a step of storing the normalized normal vector in a database; a step of obtaining a gap angle with a normal vector stored in a database based on a three-dimensional Z-axis orthogonal to the floor of the building with respect to each polygon of the building; and a step of determining a roof of the building if the gap angle is greater than or equal to a predetermined angle and determining the wall of the building if the angle exceeds a predetermined angle.

Description

METHOD AND APPARATUS FOR AUTOMATICALLY GENERATING BUILDING ELEMENT INFORMATION BY MEANING FOR USER-CUSTOMIZED THREE-DIMENSIONAL MODELING SERVICE}

The present invention relates to a method and apparatus for automatically generating semantic building element information for a customized 3D modeling providing service.

According to the Building Law, the definition of roof is the structure corresponding to the cover or cover installed on the top of the building to avoid rain, snow, dew, etc. The definition of the wall is the surface structure that is divided by blocking the interior or exterior between buildings vertically. to be. Based on this, the LOD (Level of Detail) is automatically generated by separating the roof and the wall, and building semantics information can be generated by separating the data that grouped the existing roof and the wall into one.

However, since the existing 3D building modeling format does not support building elements by semantics, there is a limit in automatically generating LOD or generating building semantic information in providing a 3D building modeling service using such a format. Existing three-dimensional building modeling formats include bil, poi, dat, and xdo. bil is the height map data of the tile area, poi is the point of interest (POI) symbol data of the tile area, or billboard symbol data of the tile area, and dat is a real 3D model symbol of the tile area. Is index data, and xdo is shape information data of an actual three-dimensional model.

In particular, national or institutional 3D building modeling data for spatial and cultural contents is difficult to use and efficient in applications requiring large processing due to its large data capacity when used in private industries such as games, tourism, movies, and performances. There is a problem that greatly falls.

Patent Registration No. 10-1031661 (2011.04.20) Registered Patent Publication No. 10-1465481 (2014.11.20)

The present invention has been made to solve the above-mentioned problems of the prior art, and the semantic building elements such as roofs, walls, windows, doors, etc., which are not supported by the existing 3D modeling format, are automatically adapted to the OGC standard. By providing a plan that can be created by the user, it is possible to adjust the level of detail (LOD) according to the user's request and provide a customized 3D modeling service that can provide the data lightly through mesh simplification. An object of the present invention is to provide a method and apparatus for automatically generating building element information for each meaning.

Another object of the present invention is to automatically classify and store information such as roofs, walls, windows, doors, etc., which are not included in the original 3D building modeling file, and store them in the database, and the user can select the LOD level and file format (COLLADA, OBJ, GML). The present invention provides a method and apparatus for automatically generating semantic building element information for a user-customizable 3D modeling providing service, by selecting desired data.

According to an aspect of the present invention for solving the above technical problem, the method for automatically generating building element information by semantics according to an aspect of the present invention is a method for automatically generating building element information for each semantic user by providing a 3D modeling service. Calculating a normal vector of polygons constituting the polygon; Normalizing the calculated normal vector; Storing the normalized normal vector in a database; Obtaining an angle between the polygons of the building and the normal vector stored in the database on the basis of a three-dimensional Z axis orthogonal to the floor of the building; And determining the roof of the building when the angle is greater than or equal to a predetermined angle, and determining the wall of the building when the angle exceeds a certain angle.

In one embodiment, the reference angle for discriminating the roof and the wall is 82 ° by the angle between the floor and the normal vector.

According to an aspect of the present invention for solving the above technical problem, the method for automatically generating building element information for each semantic is a method for automatically generating building element information for each semantic for a 3D modeling providing service. Calculating a normal vector of polygons; Normalizing the calculated normal vector; Storing the normalized normal vector in a database; Obtaining an angle between the polygons of the building and the normal vector stored in the database on the basis of a three-dimensional Z axis orthogonal to the floor of the building; Determining the roof angle of the building when the angle is greater than or equal to a predetermined angle, and determining the wall of the building when the angle exceeds a predetermined angle; Designating a boundary for inserting a roof object and a wall object in the building information of the building; And generating a grouped three-dimensional model file by inputting data into the element corresponding to the roof and the element corresponding to the wall in the three-dimensional building modeling file.

In one embodiment, the method for automatically generating building element information by semantics may further include calculating a quadratic error metric (QEM) of each edge of the building from the grouped three-dimensional model file and deleting edges having the smallest error. can do.

In one embodiment, the deleting may include searching for the lowest vertices of the lowest height of the building, setting the error of the lowest vertices high, and calculating the QEM.

Means for automatically generating building element information for each meaning according to another aspect of the present invention for solving the technical problem is a device for automatically generating building element information for each user for a 3D modeling providing service customized, building in the 3D building modeling file A normal vector calculation unit for calculating a normal vector of polygons forming a circle; A normalizer for normalizing the calculated normal vector; A storage unit for storing the normalized normal vector in a database; An angle angle calculation unit for each polygon of the building to obtain an angle between the polygon and the normal vector stored in the database based on a three-dimensional Z-axis orthogonal to the floor of the building; A discriminating unit for discriminating the roof of the building if the inter-angle is greater than or equal to a predetermined angle, and determining the wall of the building if the inter-angle is greater than a predetermined angle; A designation unit for designating a boundary for inserting a roof object and a wall object in the building information of the building; And a generation unit configured to generate a grouped three-dimensional model file by inputting data to the element corresponding to the roof and the element corresponding to the wall in the three-dimensional building modeling file.

According to an embodiment, the apparatus for automatically generating building element information for each meaning may include: a setting unit configured to search for the lowest vertices of the lowest height of the building and to set the error of the lowest vertices high; And a deleting unit for deleting edges having the smallest error by calculating a quadrature error metric (QEM) of each edge of the building in the grouped three-dimensional model file.

When using the method and apparatus for automatically generating building element information for each semantic for the above-described user-customized 3D modeling providing service, the building element information for each semantic is automatically generated by adopting grouping by data semantics using a normal vector. By doing so, it has the advantage of automatically generating semantic building elements such as roofs, walls, windows, doors, etc., which are not supported by most 3D modeling formats, in accordance with the Open Geospatial Consortium (OGC) standard.

In addition, according to the present invention, the LOD (Level of Detail) can be adjusted according to the request of the user, so that the user can customize LOD conversion, which can be customized, and provide an efficient three-dimensional modeling map information accordingly. have.

In addition, according to the present invention, it is possible to quickly service the 3D modeling map information with an optimized data capacity customized for the user terminal according to the environment of the user terminal by reducing the data by utilizing mesh simplification.

As described above, according to the present invention, even if most of the existing 3D modeling production files do not have semantic data for each element of a building or are not supported by the format itself, By using the automatic generation of semantic building element information by semantics, the aforementioned data (existing 3D modeling production files) are classified by semantics and converted into international standard data to give users three-dimensional modeling maps containing semantic building element information. You can provide information or provide custom data by extracting only specific building elements.

In addition, users can easily analyze, modify, and utilize 3D models through semantic data. In addition, through mesh simplification or mesh simplification, there is no big difference in the human eye, but the size of data can be significantly reduced, thereby providing a relatively high volume of 3D modeling data to the user at high speed and smoothly. There are advantages to it.

FIG. 1 is a block diagram illustrating a basic configuration that may be employed in an apparatus for automatically generating building element information for each meaning for a user-customizable 3D modeling providing service according to an exemplary embodiment of the present invention.
2 is a flowchart illustrating a method for automatically generating building element information for each semantic user for 3D modeling providing service according to an embodiment of the present invention.
FIG. 3 is a schematic diagram for explaining a normal vector employed in the method of FIG. 2.
4 and 5 are exemplary diagrams for describing a main process of a method for automatically generating building element information for each meaning by the method of FIG. 2.
6 and 7 are exemplary diagrams for describing building element information for each semantic generated by the method of FIG. 2.
8 is an exemplary diagram for describing a main process of a method for automatically generating building element information for each meaning according to another embodiment of the present invention.
FIG. 9 is a block diagram of an apparatus for automatically generating building element information for each semantic character for a 3D modeling providing service according to another embodiment of the present invention. Referring to FIG.

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the present invention. In describing each drawing, like reference numerals are used for like elements.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, in the following description of the present invention, a description of a known function or configuration will be omitted to clarify the gist of the present invention.

FIG. 1 is a block diagram illustrating a basic configuration that may be employed in an apparatus for automatically generating building element information for each meaning for a user-customizable 3D modeling providing service according to an exemplary embodiment of the present invention. 2 is a flowchart illustrating a method for automatically generating building element information for each semantic user for 3D modeling providing service according to an embodiment of the present invention.

Referring to FIG. 1, the apparatus 100 for automatically generating building element information for each meaning according to the present embodiment may be customized to a client terminal 300 or a web server 400 through a network 200 including a portal and a service platform. 3D modeling providing service can be provided.

To this end, the apparatus 100 for automatically generating building element information for each meaning may include a concierge and a navigator, and may include units for route determination, presentation, and the like. The presentation unit processes signals and data for representing route vectors, route directions, and maps.

In addition, the apparatus 100 for automatically generating building element information for each meaning may include a geocoder for converting an address into a coordinate, a reverse geocoder for converting a coordinate into an address, a gateway, a directory, and the like. It may be further provided.

In addition, the apparatus 100 for automatically generating building element information for each meaning 100 performs grouping on the basis of data semantics using a normal vector, and performs user-specific segmentation level of detail (LOD) transformation to convert the lightweight data to the client terminal 300 or the like. It may be provided with a means for providing to another web server 400 or a component for performing a function corresponding to the means (see FIG. 9).

The client terminal 300 may include a mobile terminal or a desktop computer, and a wireless application protocol (WAP), a java platform, a micro edition (J2ME), a CE platform, and the like may be used. The web server 400 may include a wireless application server (WAS).

A method of automatically generating building element information for each semantic that may be employed by the apparatus for automatically generating building element information for each meaning described above will now be described with reference to FIG. 2.

First, the apparatus 100 for automatically generating building element information for each meaning calculates normal vectors of polygons constituting a building in the 3D building modeling file (S21).

Next, the calculated normal vector is normalized (S22).

Next, the normalized normal vector is stored in a database (S23).

Next, for each polygon of the building, the angle between the polygon and the corresponding normal vector stored in the database is calculated based on the three-dimensional Z-axis orthogonal to the floor of the building (S24).

Next, if the angle is greater than a predetermined angle is determined as the roof of the building, if the angle exceeds a certain angle is determined as the wall of the building (S25).

Next, a boundary for inserting the roof object and the wall object in the building information of the building is designated (S26).

Next, data is input to the element corresponding to the roof and the element corresponding to the wall in the 3D building modeling file to generate a grouped 3D model file (S27).

Next, the quadratic error metric (QEM) of each edge of the building can be calculated from the grouped three-dimensional model file before the three-dimensional model file is provided to the client terminal or web server through the network. See S28 in FIG. 8). At this time, before the QEM calculation or before the edge deletion, the lowest vertices of the lowest height of the building may be searched and the error of the lowest vertices may be set high.

Hereinafter, the steps of the above-described method for automatically generating building element information for each meaning will be described in more detail.

FIG. 3 is a schematic diagram for explaining a normal vector employed in the method of FIG. 2. 4 and 5 are exemplary diagrams for describing a main process of a method for automatically generating building element information for each meaning by the method of FIG. 2.

Group by data semantics using normal vector

In connection with the first step S21 described above, a normal vector is made to distinguish the roof and the wall of the building in the three-dimensional (3D) concept. To this end, as shown in FIG. 3, 3D building modeling acquires information on vertices of all polygons. In three dimensions, a vertex may be represented by X, Y, and Z coordinates, and at least three vertices may form a polygon. In this embodiment, each vertex may be calculated to calculate a normal vector value.

In addition, in relation to the second step S22, a normalization operation for calculating and normalizing a normal vector is performed for each polygon constituting the building. Through normalization, a unit vector having a unit of 1 is generated. This unit vector contributes to speeding up the angle calculation process on the processor or memory of the automatic generation of semantic building element information.

In addition, in relation to the third step S23, data for distinguishing buildings may be represented by using a unit vector UV for a normal vector stored in a database.

In addition, in relation to the fourth step S24 and the fifth step S25, the unit vector UV calculated using the database resources is applied to the polygon of the building based on the three-dimensional Z axis as shown in FIG. 5. It can be used to find the angle θ for. According to the present embodiment, the apparatus 100 for automatically generating building element information for each meaning determines that the roof angle is closer to 0, and the outer surface (wall) is closer to 90 degrees. In this embodiment, the reference angle dividing the roof and the wall is set to 82 ° from the building floor or the ground. That is, if the angle (90-θ) of the inter-angle angle is in the range of 0 to 82 °, it is determined as a wall and when it exceeds 82 °, it is determined as a roof.

6 and 7 are exemplary diagrams for describing building element information for each semantic generated by the method of FIG. 2.

Customized segmentation LOD conversion and data provision

In relation to the sixth step S26 described above, the roof and the wall may be extracted and then applied to the CityGML model based on eXtensible Markup Language (XML). For example, as shown in FIGS. 6 and 7, a boundary is specified to put roof and wall objects in the building information, and data is grouped by inputting data to specific elements such as 'RoofSurface' and 'WallSurface' elements. A 3D modeling file. The 3D modeling file may be CityGML, but is not limited thereto.

In FIG. 6, the building modeling 20 may be extracted as the wall modeling 21 and the roof modeling 22. The extracted roof object and wall object can be displayed on the screen by selecting the invisible or invisible option among the objects of the building in the element toobar implemented in the application of the automatic generation of semantic building element information. have.

FIG. 7A illustrates a state in which all the bulding, roofsurface, and wallsurface objects are selected in the element toolbar 111, and FIG. 7B illustrates the same element toolbar 112. In FIG. 7, (c) shows a state in which both building and wallsurface objects are selected and the roofsurface object is not visible. Shows all roofsurface objects selected and the wallsurface objects not selected.

In the present embodiment, building data may be semantically divided based on CityGML following the open geospatial consortium (OGC) standard. By dividing a building into a 'RoofSurface' object and a 'WallSurface' object, it is possible to build a unit-specific LOD that can provide three-dimensional modeling data customized to user needs.

For example, if the user requires LOD2, the roof and walls must be mandatory according to the OGC standard, so that 'BuildingObjectMember' can combine 'RoofSurface' and 'WallSurface' objects. Using the method of the present embodiment, unlike the conventional method, that is, two methods of modeling a building in LOD0 including only the floor surface, there is an advantage that the 3D modeling data can be converted into a user-specified LOD. Conventional building modeling methods include a footprint on a floor and a roof edge that is a result of horizontal projection of a building corresponding to the floor.

CityGML is an open source software-based standardization model that uses maps and spatial analysis to the interior space. It is a file format that can perform urban model, facility, complex spatial model by defining objects and objects such as 3D geometry, 3D topology, meaning and representation of building data. You can expect versatility.

8 is an exemplary diagram for describing a main process of a method for automatically generating building element information for each meaning according to another embodiment of the present invention.

According to the present embodiment, the method for automatically generating building element information by semantics may further include lightening data in addition to the method for automatically generating building element information by semantics of FIG. 2. That is, in the present embodiment, 3D modeling data is lightened by utilizing mesh simplification. Using mesh simplification, it is possible to reduce the number of polygons and lighten the data while substantially maintaining the data quality of 3D modeling.

For example, FIG. 8 shows the results of before and after applying mesh simplification for buildings. In FIG. 8, (a) is a three-dimensional modeling building 20 before applying mesh simplification, and (b) is a three-dimensional modeling building 20a after applying mesh simplification. Although it is not clear from the drawing, the building 20 before application has 396 vertices and 646 polygons as a result of the mesh simplification. It can be seen that the light weight is 470.

On the other hand, mesh simplification may be implemented by calculating the QEM (Quadric Error Metric) of each edge of the building to delete the edge with the least error. In this case, some building model files do not have a floor surface. In this case, if the mesh simplification is performed, the shape of the building may be greatly changed and the original shape may not be maintained. Therefore, in the present embodiment, the vertices of the lowest height of the building model may be searched and the error of the edge including the vertex may be set high to prevent the floor surface of the building from being deleted.

FIG. 9 is a block diagram of an apparatus for automatically generating building element information for each semantic character for a 3D modeling providing service according to another embodiment of the present invention. Referring to FIG.

9, the apparatus 100 for automatically generating building element information for each meaning according to the present embodiment includes a control unit 110, a storage unit 120, and a communication unit 130. The controller 110 may be connected to the database system 140 and the input / output device 150. The automatic generation of building element information for each meaning 100 may provide a 3D modeling service to a user terminal, a client terminal, a web server, and the like connected through a network. At this time, the automatic generation of building element information for each meaning 100 may generate lightweight three-dimensional modeling data capable of adjusting LOD step by step by generating building elements for each meaning according to a standard such as OGC (open geospatial consortium).

To this end, the controller 110 may include a processor or a microprocessor having at least one central processing unit (CPU) or a core. The controller 110 may be referred to as a control device or a computing device, and may execute a program or software module stored in the storage 220 to perform a preset function such as automatic generation of building element information for each meaning.

The software module stored in the storage unit 120 and executed by the control unit 110 includes a normal vector calculation module 121, a normalization module 122, a storage module 123, a angle angle calculation module 124, and a determination module ( 125, boundary designation module 126, and file generation module 127. In addition, the software module may further include a setting module 128 and a deleting module 129. Such a software module may correspond to a component mounted on a controller or a processor corresponding thereto to perform a corresponding function. The component may include a normal vector calculator, a normalizer, a storage, a angle calculator, a determiner, a boundary designator, a file generator, a setter, and a deleter.

Specifically, the normal vector calculation module 121 calculates a normal vector of polygons constituting the building in the 3D building modeling file. In addition, the above-described normalization module 122 normalizes the normal vector calculated by the normal vector calculation module 121. The storage module 123 stores the normal vector normalized by the normalization module 122 in the database.

The angle calculation module 124 calculates the angle between the polygons of the building and the normal vector stored in the database on the basis of the three-dimensional Z axis orthogonal to the floor of the building. The determination module 125 determines the roof of the building if the angle is greater than or equal to a predetermined angle, and determines the wall of the building if the angle exceeds a certain angle.

The aforementioned angle may be calculated by the following equations (1) to (3).

In Equations 1 to 3, X, Y, and Z are three vertices of a polygon, Z 'is a normal vector, and θ represents an angle (between angles) between the Z axis and the normal vector.

The boundary designation module 126 specifies a boundary for inserting a roof object and a wall object in the building information of the building. The file generation module 127 generates a grouped three-dimensional model file by inputting data to the element corresponding to the roof and the element corresponding to the wall in the three-dimensional building modeling file.

The configuration module 128 retrieves the lowest vertices of the lowest height of the building and sets the error of the lowest vertices high. The deletion module 129 calculates a quadric error metric (QEM) of each edge of the building in the grouped three-dimensional model file to delete the edge with the smallest error.

The storage unit 120 includes an operating system, a program, and a command set for the basic operation of a device for automatically generating building element information by meaning, in addition to a software module for implementing a method of automatically generating building element information by meaning for a 3D modeling service. And so on.

The storage unit 120 may include a semiconductor memory such as non-volatile random access memory (NVRAM), a dynamic random access memory (DRAM), which is a typical volatile memory, a hard disk drive (HDD), and an optical storage device. , Flash memory, cloud storage systems, or a combination thereof.

The software module may be stored in a separate computer readable medium (recording medium) in the form of a program or software, loaded into a computing device as necessary, and then implemented to perform a corresponding function according to a command of the controller 110. Computer-readable media can be provided to include program instructions, data files, data structures, and the like, alone or in combination. The programs recorded on the computer readable medium may be those specially designed and configured for the present invention, or may include those known and available to those skilled in computer software.

In addition, the computer readable medium may include a hardware device specifically configured to store and execute program instructions, such as a ROM, a RAM, a flash memory, and the like. The program command may include high-level language code that can be executed in the controller 110 using an interpreter as well as machine code generated by a compiler.

The communication unit 130 may connect one or more communication protocols to connect the device 100 for automatically generating building element information according to meaning by means of a user terminal, a client terminal, a tourism company server, a game company server, a movie company server, and a performance company server. It may be provided with at least one wired and / or wireless sub-communication system for supporting.

The input / output device 150 may include a keyboard, a mouse, a microphone, a touch pad, a touch panel, a display device, a speaker, and the like. The input / output device 150 may be connected to the controller 110 by wire or wirelessly. In a broad concept, the input / output device 150 may be included in a device for automatically generating building element information for each meaning.

In the above-described embodiment, the controller 110 corresponds to a web server or a web application server for a 3D modeling providing service, and the storage 120 corresponds to a database (DB) system for storing and managing data files. Can be. In that case, a data server may be disposed in the database system and the web application server so as to handle transmission and reception of signals and data therebetween.

In addition, according to the implementation, the control unit may be implemented as a web service server, and the storage unit 120 may be implemented as a DB server included in a data system connected to the web service server.

On the other hand, when using the method and apparatus for automatically generating building element information by semantics for the 3D modeling providing service according to the above-described embodiment, information on roofs, walls, windows, doors, etc., which are not included in the original 3D building modeling file They can be automatically classified and stored in a database, and users can select the LOD level and file format (COLLADA, OBJ, GML, etc.) to be provided with the desired 3D building modeling data. Here, the file format is not limited to the above-described COLLADA, OBJ, and GML formats, and the file format may be widely used to efficiently use 3D building modeling provided by a national institution or a public institution in industries such as game, tourism, film, and performance. Of course, any one or more of the formats that are highly compatible can be selected and used.

Although described above with reference to a preferred embodiment of the present invention, those skilled in the art will be variously modified and changed within the scope of the invention without departing from the spirit and scope of the invention described in the claims below I can understand that you can.

Claims (7)

delete delete As a method of automatically generating building element information by semantics for a 3D modeling providing service,
Calculating a normal vector of polygons constituting the building in the three-dimensional building modeling file;
Normalizing the calculated normal vector;
Storing the normalized normal vector in a database;
Obtaining an angle between the polygons of the building and the normalized normal vector based on a three-dimensional Z-axis orthogonal to the floor of the building;
Determining the roof angle of the building when the angle is greater than or equal to a predetermined angle, and determining the wall of the building when the angle exceeds a predetermined angle;
Designating a boundary for inserting a roof object and a wall object in the building information of the building; And
The element corresponding to the roof and the element corresponding to the wall in the 3D building modeling file. Element Information Converts 3D modeling data into grouped three-dimensional model files by converting 3D modeling data into user-defined granular LODs to enter data and semantically divide objects in building data based on CityGML following OGC (open geospatial consortium) standards. Generating;
Lightening three-dimensional modeling data by using mesh simplification by calculating quadric error metric (QEM) of each edge of the building from the grouped three-dimensional model file and deleting edges having the least error. Including but not limited to,
The lightweighting may include searching for the vertices of the lowest height of the building model in the grouped 3D model file, setting the error of the edge including the vertex relatively high, and setting the error and the QEM of the vertex relatively higher. Comparing calculated values to prevent the floor surface of the building model from being deleted from the three-dimensional modeling data, automatic building element information for each meaning. delete delete As a device for automatically generating building element information by semantics for a 3D modeling providing service,
A normal vector calculator for calculating normal vectors of polygons constituting the building in the 3D building modeling file;
A normalizer for normalizing the calculated normal vector;
A storage unit for storing the normalized normal vector in a database;
An angle angle calculation unit for each polygon of the building to obtain an angle between the normalized normal vector and a reference to a three-dimensional Z-axis orthogonal to the floor of the building;
A discriminating unit for discriminating the roof of the building if the inter-angle is greater than or equal to a predetermined angle, and determining the wall of the building if the inter-angle is greater than a predetermined angle;
A designation unit for designating a boundary for inserting a roof object and a wall object in the building information of the building; And
In the 3D building modeling file, element information data is input to elements corresponding to the roof and elements corresponding to the wall, and means to distinguish objects of the building data based on CityGML according to the open geospatial consortium (OGC) standard. A generation unit for generating a grouped three-dimensional model file by converting the 3D modeling data into a LOD segmented into a user-specific subdivision;
A setting unit for searching for vertices of a building model of the building and setting an error of an edge including the vertex; And
In order to reduce the three-dimensional modeling data by using mesh simplification in such a manner that the quadratic error metric (QEM) of each edge of the building model is calculated from the grouped three-dimensional model file and the edge having the smallest error is deleted. Including;
The deleting unit prevents the floor surface of the building model from being deleted from the 3D modeling data based on a QEM calculation value of the lowest height vertex of the building model and an error set relatively high with respect to the lowest height vertex. Automatic generation device for semantic building element information. delete

Images