KR20200029145A - Method and Apparatus for indoor space modeling with function, behavior and structure for intelligent indoor service - Google Patents

Abstract

Disclosed are a method and an apparatus for modeling an indoor space including a meaning and a function of a space for an intelligent indoor space service. According to the present invention, the method for modeling an indoor space including a meaning and a function of a space for an intelligent indoor space service comprises: a processing step of extracting three-dimensional data for an indoor space; a step of segmenting the three-dimensional data into a plurality of cells; a step of checking a physical constraint and a semantic constraint for the plurality of cells to divide a space by dividing a space in which a person can move and a space in which a person cannot move; a step of dividing an object cell in which an object exists and an empty cell in which nothing exists, and merging the divided space and the plurality of cells; a step of generating final cells by merging and assigning an ID and information for each cell; a step of generating an area which is an upper semantic space by merging the final cells based on structure and function information of the final cells; a step of generating a space merged in accordance with function and structure information of the generated area; and a step of using the final cells, the area, and the space to generate an edge, and use the edge to generate a function-behavior-structure Map (FBSMAP).

Description

{Method and Apparatus for indoor space modeling with function, behavior and structure for intelligent indoor service}

The present invention relates to an indoor space modeling method and apparatus including functions and meanings of space for an intelligent indoor space service.

Recently, indoor environment-related services have attracted attention. Examples are intelligent indoor services such as assisted living (secretary service), indoor navigation, and augmented reality-based avatar telepresence. There are many technically necessary elements to provide intelligent indoor service. One of the important factors is to acquire digital data about the indoor space.

In the past, indoor space services mainly consisted of positioning services such as navigating services based on precise positioning, but AI-based user-assisted services are expected to be central in the upcoming fourth industrial revolution. For example, the AI secretary understands the user's movements or behaviors, and provides appropriate personalization service accordingly.In voice guidance, information is provided, such as IoT devices located in indoor spaces (e.g. lighting, power Etc.) can be expected to be properly controlled.

In order for such a system to work well, physical coordinates of where the IoT devices located in the indoor space exist are required, and furthermore, it must be provided so that the AI secretary can understand the clear meaning of the space. This is because it is difficult for the AI secretary to understand where the indoor space is the resting area, and where it is the active space or the living room.

Therefore, there is a need for a method for logical modeling including the functions and meanings of spaces, beyond the physical modeling method of past spaces, which is required for future intelligent services.

The technical problem to be achieved by the present invention is to provide a method and apparatus for logical modeling including the function and meaning of a space, further from the physical modeling method of the past space required for the future intelligent service. In more detail, we propose a new indoor space modeling method, FBSMAP (Function-Behavior-Structure Map), which focuses on spatial functions in a complex indoor environment so that the system can better understand the meaning of space and provide it to services.

In one aspect, the indoor space modeling method including the function and meaning of space for an intelligent indoor space service proposed in the present invention includes a processing step of extracting 3D data for an indoor space, and subdividing the 3D data into a plurality of cells , Identifying physical constraints and semantic constraints on the plurality of cells, dividing spaces by dividing spaces that can be moved by people and non-movable spaces, and object-cells and nothing where objects exist Distinguishing the empty cells (Empty-Cells), performing the merging between the divided spaces and the plurality of cells, and performing the merging to generate a final cell and give information and ID for each cell Step, based on the structure and function information of the final cell, the upper semantic space is zero through merging between the final cells. Generating an area, generating a merged space according to the function and structure information of the generated area, and generating an edge using the final cell, the area and the space, and using the edge And generating a function-behavior-structure map (FBSMAP).

The step of dividing a space by dividing a space that can be moved by a person and a space that cannot be moved by checking physical constraints and semantic constraints for the plurality of cells is a boundary box for an obstacle using the size value of the obstacle in the indoor space Create a (Bounding Box), check the intersection of the grid and the bounding box of the plurality of cells to check the exact location and association with respect to the obstacle, check the physical constraints, and whether there is access to the movable spaces Identify semantic constraints that determine

In the step of generating the final cell by performing the merging and giving information and ID for each cell, in the case of an object-cell, there is structure and behavior information according to an object, and according to an object A final cell containing structure and operation information is generated.

The step of generating an area that is an upper semantic space through merging between the final cells based on the structure and function information of the final cell is based on the physical constraints and the spatial characteristics and functions of the obstacle determined by the distance and direction of the obstacle. An area is generated based on the determined semantic constraint, and the object-cell reflecting the physical constraint is merged into the area by the semantic constraint.

Generating an edge using the final cell, the region and space, and generating a function-behavior-structure map (FBSMAP) using the edge include empty-cell and object-cell among the final cells. By connecting the center points of (Object-Cell), an edge that is movement information that a person can move is generated.

In another aspect, the indoor space modeling apparatus including the function and meaning of space for an intelligent indoor space service proposed by the present invention is a processing unit for extracting 3D data for an indoor space, and subdividing the 3D data into a plurality of cells Then, by checking physical constraints and semantic constraints on the plurality of cells, a space dividing unit for dividing a space by dividing a space in which a person can move from a non-movable space, and an object-cell in which an object exists ) And an empty cell (Empty-Cell) that does not exist, a cell merging unit performing merging between the divided spaces and the plurality of cells, and performing the merging to generate a final cell and for each cell. Final cell generation unit that gives information and ID, and merges between final cells based on the structure and function information of the final cell A region generating unit for generating an area that is a higher-order semantic space, a space generating unit for generating a merged space according to the function and structure information of the generated region, and the final cell, the region, and the space It includes an FBSMAP generator that generates an edge and generates a function-behavior-structure map (FBSMAP) using the edge.

According to embodiments of the present invention, an ambiguous indoor spatial representation can be established by performing spatial modeling based on spatial features of an indoor space, and the quality of the indoor space Objectives can be established for analysis. In addition, correlation and similarity between spaces can be acquired as quantitative indicators, and various advanced intelligent services based on indoor spaces such as avatar telepresence, artificial intelligence-based intelligent smart home services, and indoor location-based services can be provided. It can be used as basic data.

1 is a view for explaining indoor space terminology according to an embodiment of the present invention.
2 is a flowchart illustrating an indoor space modeling method including functions and meanings of spaces for an intelligent indoor space service according to an embodiment of the present invention.
3 is a view for explaining a FBS expression method according to an embodiment of the present invention.
4 is a view for explaining a processing process for an indoor space according to an embodiment of the present invention.
5 is a view for explaining a process of subdividing the indoor space into a plurality of cells according to an embodiment of the present invention.
6 is a view for explaining a process of checking the physical constraints and semantic constraints according to an embodiment of the present invention.
7 is a view for explaining a merge process between a plurality of cells according to an embodiment of the present invention.
8 is a view for explaining a final cell generation process according to an embodiment of the present invention.
9 is a view for explaining a region generation process according to an embodiment of the present invention.
10 is a view showing an area generated according to an embodiment of the present invention.
11 is a view for explaining a process for creating a space according to an embodiment of the present invention.
12 is a view for explaining an edge generation process according to an embodiment of the present invention.
13 is a view for explaining the structure of an FBSMAP according to an embodiment of the present invention.
14 is a diagram showing the structure of an indoor space modeling apparatus including functions and meanings of spaces for an intelligent indoor space service according to an embodiment of the present invention.
15 and 16 are diagrams showing FBSMAP results according to an embodiment of the present invention.

Intelligent indoor service has many factors to consider for effective performance. One of the factors is the understanding of indoor spatial information. In order to provide optimal indoor service, it is necessary to understand not only the user but also the clear meaning of space in a complex indoor environment. As the demand for indoor intelligent services increases, a system for generating the corresponding information becomes more important. The existing modeling method is insufficient for the system to effectively recognize the meaning of space and use it for services.

First, existing indoor space modeling methods understand space as an explicit meaning. For example, functional characteristics may be different for each living room in a home, and the purpose may be different due to the arrangement of various indoor objects in the space. That is, the space of the living room may have various characteristics and purposes for every family, but it is expressed in one meaning only.

Second, the functional characteristics of the indoor space are not effectively expressed so that the system can understand it. In a space, one furniture can have various functionalities, and when combined with other furniture, it can have various functionalities. For example, the functional characteristics may be different when there is only one chair and when there is a desk and a chair together.

The present invention proposes a new indoor space modeling method, FBSMAP (Function-Behavior-Structure Map), which focuses on spatial functions in a complex indoor environment so that the system can better understand the meaning of space and provide it to services. Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a view for explaining indoor space terminology according to an embodiment of the present invention.

1, terms for classifying indoor spaces are defined in the present invention. Furniture and objects exist in the interior space, and are classified as shown in FIG. 1 according to the size and functionality of the space area in order to express it as a subdivided semantic space. For this classification, refer to Norberg-Schulz's 'existential space' theory and OGC's international standard for indoor spatial information, IndoorGML.

Existential space lays the theoretical foundation for the semantic of interior space. IndoorGML is an international standard for modeling room-based indoor space, especially indoorGML, which uses the concept of cell space, which uses architectural elements such as roofs, walls, and ceilings indoors. It is not used as an element that defines space, but is simplified by focusing on the space that can be expressed from each architectural element. Examples of typical spaces are rooms, hallways, and stairs.

The present invention is intended to be expressed as a subdivided semantic space in consideration of both furniture and indoor objects. To this end, it is classified as follows according to the size and functionality of the domain with reference to Norberg-Schulz's 'existential space' theory and OGC's international standard for indoor space information, IndoorGML.

In the present invention, the cell shown in Fig. 1 (a), the center & edge shown in Fig. 1 (b), the area shown in Fig. 1 (c), shown in Fig. 1 (d) Redefine it as an area (space).

The cell in FIG. 1 (a) refers to a minimum unit space area having a minimum meaning as a space range in which a person exists. Basically, it is divided into a grid, but it can be composed of polygons depending on the situation.

In the case of the center of FIG. 1 (b), the center point is defined in a spatial area having a minimum meaning, and in the case of an edge, a path that a person can move is represented, and is expressed as connection information between the center and the center. .

The area (Area) of FIG. 1 (c) means a set of cell space areas capable of consistent behavior through aggregation between cells. For example, a set of a space in which a TV exists and a space in which a sofa exists can be defined as a resting place that anyone can understand.

In the case of the space of FIG. 1 (d), it is defined as a semantic space through a combination between regions at a higher level than the region. For example, it can be expressed and understood as a living room or kitchen.

In this way, the meaning of a more specific space can be given and the hierarchical composition is possible through the arrangement of terms for the indoor space. This systematic information can be easily used for storage and sharing.

2 is a flowchart illustrating an indoor space modeling method including functions and meanings of spaces for an intelligent indoor space service according to an embodiment of the present invention.

The proposed indoor space modeling method including the function and meaning of space for an intelligent indoor space service includes a processing step of extracting 3D data for an indoor space (210), subdividing the 3D data into a plurality of cells (220), and Checking physical constraints and semantic constraints on a plurality of cells to divide a space that can be moved by a person and a space that cannot be moved, thereby dividing the space (230), and an object-cell (Object-Cell) in which the object exists Separating the empty-empty empty cells (Empty-Cell), performing a merge between the divided space and the plurality of cells (240), and performing the merge to generate a final cell and information about each cell And assigning an ID (250), based on the structure and function information of the final cell, creating an area, which is an upper semantic space, through merging between the final cells. Step 260, generating a merged space according to the function and structure information of the generated area (270) and generating an edge using the final cell, the area and the space, and using the edge And generating a function-behavior-structure map (FBSMAP) (280).

In step 210, processing is performed to extract 3D data for the indoor space.

In step 220, the 3D data is subdivided into a plurality of cells.

In step 230, physical spaces and semantic constraints are checked for the plurality of cells to divide a space that can be moved and a space that can be moved by a person to divide the space. A boundary box for an obstacle is generated using the size value of the obstacle in the indoor space. Check the intersection of the grid of the plurality of cells and the bounding box to check the exact location and association with respect to the obstacle, check the physical constraints, and determine the semantic constraints to determine whether there is access to the movable spaces. To confirm.

In step 240, an object-cell (Object-Cell) in which an object exists and an empty-cell (Empty-Cell) in which nothing exists are distinguished, and merging between the divided space and the plurality of cells is performed.

In step 250, the merge is performed to generate a final cell and information and an ID for each cell are provided. In the case of an object-cell, since structure and behavior information exists according to an object, a final cell including structure and operation information is generated according to an object.

In step 260, an area, which is an upper semantic space, is generated through merging between final cells based on the structure and function information of the final cell. Object-cells that generate regions based on physical constraints determined by the distance and direction of obstacles and semantic constraints determined by spatial features and functions of obstacles, and object-cells that reflect physical constraints are domains by semantic constraints. Are merged into

In step 270, a merged space is generated according to the function and structure information of the generated region.

In step 280, an edge is generated using the final cell, the region and space, and a function-behavior-structure map (FBSMAP) is generated using the edge. Among the final cells, the center points of the empty-cell and the object-cell are connected to generate an edge that is movement information that can be moved by a person. The interior space modeling method including functions and meanings of spaces for the proposed intelligent interior space service will be described in more detail with reference to FIGS. 3 to 13.

3 is a view for explaining a FBS expression method according to an embodiment of the present invention.

The indoor space modeling method presented in the present invention is characterized by generating an FBSMAP through the following steps.

First, FBS is an abbreviation of Function-Behavior-Structure and is a method of constructing ontology. When designing things, objects are conceptualized into three categories of FBS. In order to secure semantic features of space, each unit space is analyzed as physical and semantic (behavior & function) features. When identifying the nature of a space, it is intended to use the property that its nature is determined by the objects present in the space, and for this, it is characterized by expressing objects (for example, furniture) existing in the indoor space with FBS. Is done.

In a series of examples, when there is one chair, which is one object in FIG. 3 (a), when this is expressed through FBS, a structure and behavior may be defined. The function for the chair is basically expressed through motion, so it is not expressed separately. For example, the structure may indicate the type, size, location, direction, and the motion may include sit, recline, and stand on. Can indicate action.

3 (b) can be expressed in the same way. When there is one desk, which is a single object, it can be defined through FBS to define the structure and behavior. For example, the structure can indicate the type, size, location, direction, and put on, recline, get under, etc. Can indicate the operation of.

When each defined object is combined in the same space, new functions can be created. For example, when FIG. 3 (a) and FIG. 3 (b) exist in one space, it can be expressed as a study space, a book reading space, or a sitting space, respectively. The present invention generates an indoor space map using an expression method according to FBS.

In the case of FIG. 3 (c), when there are two objects, a chair and a desk, when this is expressed through FBS, a structure and a function can be defined. For example, a structure may indicate a type, a number, and a size, and a function may indicate an operation such as study, read, or stand on. .

4 is a view for explaining a processing process for an indoor space according to an embodiment of the present invention.

In the processing step 210 of extracting 3D data for an indoor space, information generated in 3D about the environment inside the house is obtained, and all information is extracted. There are various studies that automatically generate reality as a house 3D model. In the present invention, it is assumed that an indoor house 3D model is generated through scanning using an RGB-D camera. Elements such as walls, ceilings, floors, windows, doors, etc. of the building can also be separated to obtain information, and all furniture is separated into new categories. 4 shows a process of transforming the physical world into a 3D data set in one embodiment.

FIG. 4 (a) shows a physical based rendering, FIG. 4 (b) shows a depth, and FIG. 4 (c) shows a volumetric ground truth. FIG. 4 (d) shows house 3D model data obtained by converting this physical world into a 3D data set.

5 is a view for explaining a process of subdividing the indoor space into a plurality of cells according to an embodiment of the present invention.

In step 220 of subdividing 3D data into a plurality of cells, a reference point must be identified in order to generate a cell. Using 3D data, the reference point is determined based on the point rising in the Z direction from the lowest point in the X and Y coordinates of the floor surface. A grid rectangle is created at the origin and in the X and Y directions to completely cover all floor surfaces. All cells located outside the bottom surface are removed in view of the vertical projection. Then, a plane intersecting the contour of the floor surface is selected and refined according to the shape of the floor surface. The grid according to the embodiment of the present invention is characterized in that it is generated at intervals of 0.6 m ² . This is the minimum space occupied by one person.

6 is a view for explaining a process of checking the physical constraints and semantic constraints according to an embodiment of the present invention.

In step 230 of dividing a space by dividing a space that can be moved by a person and a space that cannot be moved by checking physical constraints and semantic constraints for a plurality of cells, physical constraints and semantic constraints for cells Check to distinguish the space that can be moved by people from the space that cannot be moved. Physical constraints are determined by building elements and furniture. Particularly, in the 3D model, furniture information has a size value. Through this, a bounding box for each furniture is generated in the interior space.

By checking the intersection of the created cell grid and the bounding box, you can check the exact location of the furniture and its association. The semantic constraint is to determine whether or not to have access to the movable spaces, and the accessible space may differ depending on the type of the user. This includes the user's own personal information and walking type.

7 is a view for explaining a merge process between a plurality of cells according to an embodiment of the present invention.

In step 240 of separating the object-cell (Object-Cell) in which an object exists and the empty-cell (Empty-Cell) in which nothing exists, and performing merging between the divided space and the plurality of cells, the previous two Inter-cell merging is performed on the divided spaces performed in the step. The object-cell (Object-Cell) 710 in which the object (Object) is present and the empty-cell (Empty-Cell) 720 in which nothing is present are separated, and the object-cell (Object-Cell) is merged.

Object-cells can be secured through the furniture's bounding box. The merge is performed considering the object border and the cell size. In this process, the size of the data is reduced by removing unnecessary partitions. FIG. 7 (a) shows a merged image between a divided space and a plurality of cells, and FIG. 7 (b) shows an object-cell and an empty-cell.

8 is a view for explaining a final cell generation process according to an embodiment of the present invention.

In the step 250 of generating a final cell by merging and giving information and ID for each cell, the final cell is determined through the previous three steps, and information and ID are assigned to each cell. In the case of an object-cell, there are structures and actions depending on the object, so the final cell includes actions and structures depending on the object.

9 is a view for explaining a region generation process according to an embodiment of the present invention.

Based on the structure and function information of the final cell, in step 260 of generating an area, which is an upper semantic space, through merging between the final cells based on the structure and function information, based on the structure and function information of the object-cell, The region, which is the upper semantic space, is generated by merging between cells.

The criteria for domain creation are divided into two categories. First, based on the physical constraints determined by the distance and direction of the obstacle and the semantic constraints determined by the spatial characteristics and functions of the obstacle, an object-cell reflecting the physical constraints is applied to the semantic constraints. Is merged into an area. In other words, the object-cells reflect physical and spatial characteristics such as the distance and direction of furniture, and are merged into areas by semantic criteria. Fig. 9 (a) shows the cell space, and Fig. 9 (b) shows the area space.

10 is a view showing an area generated according to an embodiment of the present invention.

As illustrated in FIG. 10, a structure, operation, and structure for cell 2, a function for cell 1, and a function for area 1 may be represented. In addition, the structure for three, the operation, the structure for cell 4, the operation, and the function for area 2 can be represented.

11 is a view for explaining a process for creating a space according to an embodiment of the present invention.

In the step 270 of generating a merged space according to the function and structure information of the generated area, the space is merged according to the function and structure of the area. The basic principles are similar to the previous one, but have a higher level of meaning.

The space is created based on the function list, location and meaning in a given indoor environment. Since the spatial concept of area and space is a higher-level layer connected by layers compared to cells, the properties of the initially defined cell are inherited. Fig. 10 (a) shows the area space, and Fig. 10 (b) shows the space.

12 is a view for explaining an edge generation process according to an embodiment of the present invention.

Finally, in the step 280 of generating an edge using the final cell, the region and the space, and generating an FBSMAP (Function-Behavior-Structure Map) using the edge, the center point of the empty cell and the object cell (280) Center) to create an edge. Edge can be used as mobile information that can be moved by a person, and at the same time, it can be used as various analysis data for the space.

13 is a view for explaining the structure of an FBSMAP according to an embodiment of the present invention.

The cell 1310, the region 1320, and the space 1330 are shown in 3D in FIG. 13 (a), and the structure of the FBSMAP proposed in the present invention is shown in FIG. 13 (b).

Cells exist in the lowest layer in the space, and the cells have a structure and behavior. Cells are divided into object-cells and empty-cells in detail. A region may be generated by merging adjacent cells according to the S and B of the cells, and through this, a function that is upper semantic information may be defined. Areas can also be created by merging adjacent areas in the same way. As shown in FIG. 13 (b), FBSMAP can be DB to be used as spatial information for intelligent indoor space service.

14 is a diagram showing the structure of an indoor space modeling apparatus including functions and meanings of spaces for an intelligent indoor space service according to an embodiment of the present invention.

The proposed indoor space modeling device including the function and meaning of space for the intelligent indoor space service includes a processing unit 1410, a space division unit 1420, a cell merging unit 1430, a final cell generation unit 1440, and area generation It includes a unit 1450, a space generator 1460, and an FBSMAP generator 1470.

The processing unit 1410 extracts 3D data for the indoor space.

The space dividing unit 1420 divides 3D data into a plurality of cells, and checks physical constraints and semantic constraints on the plurality of cells to divide a space that can be moved by a person and a space that cannot be moved, thereby dividing the space. The space dividing unit 1420 generates a bounding box for an obstacle using the size value of the obstacle in the indoor space. Check the intersection of the grid of the plurality of cells and the bounding box to check the exact location and association with respect to the obstacle, check the physical constraints, and determine the semantic constraints to determine whether there is access to the movable spaces. To confirm.

The cell merging unit 1430 distinguishes between an object-cell (Object-Cell) in which an object exists and an empty-cell (Empty-Cell) in which nothing exists, and performs merging between the divided space and the plurality of cells.

The final cell generator 1440 performs merging to create a final cell and provides information and IDs for each cell. In the case of an object-cell, since the final cell generator 1440 has structure and behavior information according to an object, the final cell generator 1440 generates a final cell that includes structure and operation information, respectively.

The region generating unit 1450 generates an area, which is an upper semantic space, by merging the final cells based on the structure and function information of the final cell. The region generating unit 1450 generates regions based on physical constraints determined by the distance and direction of obstacles and semantic constraints determined by spatial features and functions of the obstacles, and the object-cell reflecting the physical constraints Merged into domains by semantic constraints.

The space generation unit 1460 generates a merged space according to the function and structure information of the generated area.

The FBSMAP generation unit 1470 generates an edge using the final cell, the region and space, and an FBSMAP (Function-Behavior-Structure Map) using the edge. The FBSMAP generation unit 1470 connects the center points of the empty cell and the empty cell (Object-Cell) among the final cells to generate an edge that is movement information that can be moved by a person.

15 and 16 are diagrams showing FBSMAP results according to an embodiment of the present invention.

15 (a) and 15 (b) are FBSMAP results according to an embodiment of the present invention, and space 1510, area 1520, object-cell 1530, M0 cell 1540, edge ( 1550) and the center point (1560). 16 is a result of FBSMAP according to another embodiment of the present invention, and spaces 1610, regions 1620, object-cells 1630, M0 cells 1640, edges 1650 and center points 1660. Shown.

According to an embodiment of the present invention, an ambiguous indoor spatial representation can be established by performing space modeling based on spatial features of an indoor space.

The method of cornering the interior space based on interior and furniture has not been clearly resolved. According to an embodiment of the present invention, an objectification method for qualitative analysis of an indoor space can be established, and correlations and similarities between spaces can be obtained as quantitative indicators. In addition, it can be used as basic data to provide various advanced intelligent services based on indoor space, such as avatar telepresence, artificial intelligence-based intelligent smart home service, and indoor location-based service.

The device described above may be implemented with hardware components, software components, and / or combinations of hardware components and software components. For example, the devices and components described in the embodiments include, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors (micro signal processors), microcomputers, field programmable arrays (FPAs), It may be implemented using one or more general purpose computers or special purpose computers, such as a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. The processing device may run an operating system (OS) and one or more software applications running on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, a processing device may be described as one being used, but a person having ordinary skill in the art, the processing device may include a plurality of processing elements and / or a plurality of types of processing elements. It can be seen that may include. For example, the processing device may include a plurality of processors or a processor and a controller. In addition, other processing configurations, such as parallel processors, are possible.

The software may include a computer program, code, instruction, or a combination of one or more of these, and configure the processing device to operate as desired, or process independently or collectively You can command the device. Software and / or data may be interpreted by a processing device, or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. Can be embodied in The software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer-readable medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiments or may be known and usable by those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, and magnetic media such as floptical disks. -Hardware devices specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language code that can be executed by a computer using an interpreter, etc., as well as machine language codes produced by a compiler.

As described above, although the embodiments have been described by a limited embodiment and drawings, those skilled in the art can make various modifications and variations from the above description. For example, the described techniques are performed in a different order than the described method, and / or the components of the described system, structure, device, circuit, etc. are combined or combined in a different form from the described method, or other components Alternatively, even if replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (10)

A processing step of extracting 3D data for the interior space;
Subdividing the 3D data into a plurality of cells;
Dividing a space by identifying physical and semantic constraints on the plurality of cells and dividing a space in which a person can move from a non-movable space;
Separating an object-cell (Object-Cell) in which an object exists and an empty-cell (Empty-Cell) in which nothing exists, and merging the divided space and the plurality of cells;
Generating a final cell by performing the merging and giving information and an ID for each cell;
Generating an area that is an upper semantic space through merging between final cells based on the structure and function information of the final cell;
Generating a merged space according to the function and structure information of the generated region; And
Generating an edge using the final cell, the region and space, and generating a function-behavior-structure map (FBSMAP) using the edge
Indoor space modeling method comprising a. According to claim 1,
The step of dividing a space by dividing a space that can be moved by a person and a space that cannot be moved by checking physical constraints and semantic constraints on the plurality of cells is performed.
Physical constraints by creating a bounding box for an obstacle using the size value of the obstacle in the indoor space, and checking the intersection of the grid of the plurality of cells and the boundary box to confirm the correct position and association with respect to the obstacle Checking conditions and checking semantic constraints to determine if you have access to mobile spaces
Interior space modeling method. According to claim 1,
The step of generating the final cells by performing the merging and giving information and IDs for each cell is:
In the case of an object-cell, since structure and behavior information exists according to an object, a final cell including structure and behavior information is generated according to an object.
Interior space modeling method. According to claim 1,
The step of generating an area that is an upper semantic space through merging between final cells based on the structure and function information of the final cell is:
Object-cells that generate regions based on physical constraints determined by the distance and direction of obstacles and semantic constraints determined by spatial features and functions of obstacles, and object-cells that reflect physical constraints are domains by semantic constraints. To be merged into
Interior space modeling method. According to claim 1,
The step of generating an edge using the final cell, the region and the space, and generating an FBSMAP (Function-Behavior-Structure Map) using the edge,
Among the final cells, the center points of the empty-cell and the object-cell are connected to generate an edge, which is movement information that can be moved by a person.
Interior space modeling method. A processing unit that extracts 3D data for the indoor space;
A space dividing unit that divides the 3D data into a plurality of cells and identifies physical and semantic constraints on the plurality of cells, thereby dividing a space by dividing a space in which a person can move from a non-movable space;
A cell merging unit which distinguishes between an object-cell where an object exists and an empty-cell, and performs merging between the divided space and the plurality of cells;
A final cell generator for generating final cells by performing the merging and giving information and IDs for each cell;
An area generating unit generating an area, which is a higher semantic space, through merging between final cells based on the structure and function information of the final cell;
A space generation unit generating a merged space according to the function and structure information of the generated area; And
An FBSMAP generator that generates an edge using the final cell, the region, and space, and an FBSMAP (Function-Behavior-Structure Map) using the edge.
Indoor space modeling device comprising a. The method of claim 6,
The space division unit,
Physical constraints by creating a bounding box for an obstacle using the size value of the obstacle in the indoor space, and checking the intersection of the grid of the plurality of cells and the boundary box to confirm the correct position and association with respect to the obstacle Checking conditions and checking semantic constraints to determine if you have access to mobile spaces
Indoor space modeling device. The method of claim 6,
The final cell generation unit,
In the case of an object-cell, since structure and behavior information exists according to an object, a final cell including structure and behavior information is generated according to an object.
Indoor space modeling device. The method of claim 6,
The region generating unit,
Object-cells that generate regions based on physical constraints determined by the distance and direction of obstacles and semantic constraints determined by spatial features and functions of obstacles, and object-cells that reflect physical constraints are domains by semantic constraints. To be merged into
Indoor space modeling device. The method of claim 6,
The FBSMAP generation unit,
Among the final cells, the center points of the empty-cell and the object-cell are connected to generate an edge, which is movement information that can be moved by a person.
Indoor space modeling device.

Images