KR20240105644A - Cad data automatic labeling device and method for landscape design - Google Patents

Abstract

The present invention relates to an automatic CAD data labeling device and method for landscape design, the device comprising: a CAD data receiver that receives CAD data of a landscape design drawing from a user terminal; a layer extraction unit that extracts layers for each purpose from the CAD data; a labeling performing unit that extracts objects and outlines for each layer for each purpose, identifies an area, and performs labeling to assign a color for each purpose to the identified area; and a learning data configuration unit that merges the labeled data for each purpose to form training data for an artificial intelligence model for automating landscape design.

Description

CAD data automatic labeling device and method for landscape design {CAD DATA AUTOMATIC LABELING DEVICE AND METHOD FOR LANDSCAPE DESIGN}

The present invention relates to landscape design automation technology, and more specifically, to a CAD data automatic labeling device and method for landscape design that can automatically convert CAD data of landscape design drawings into labeling data for artificial intelligence learning.

Among the various criteria that determine the value of an apartment, the value of the floor plan and landscaping are factors that become more important as time passes. Although the importance of high-quality landscape design is increasing, the design process still involves a lot of inefficient manual work and repetitive processes.

All landscaping work is carried out according to blueprints. The process of drawing and expressing a design is called drafting. Drafting is the process of using drafting tools to express the designer's intentions with lines, symbols, letters, etc. on drafting paper. Since drawings are necessary for construction, the expression must be concise and accurate, and must be written so that anyone can easily understand. Recently, computer aided design & drafting (CAD) has been actively used in landscape design.

However, in landscape design, the process of creating landscape drawings takes a lot of time and money because it requires manual labor. For this reason, there are limits to frequent revisions or producing multiple plans to find the optimal landscape design. Additionally, there are limitations to modifying the landscape engineering design to an improved design at the construction site. Additionally, we have no choice but to propose limited landscaping construction plans to customers.

Therefore, if the overall process of landscape design can be automated, the landscape drawing creation process can be processed more efficiently.

Korean Patent No. 10-0668159 (2007.01.05)

One embodiment of the present invention seeks to provide an automatic CAD data labeling device and method for landscape design that can automatically convert CAD data of landscape design drawings into labeling data for artificial intelligence learning.

One embodiment of the present invention provides a CAD data automatic labeling device and method for landscape design that can reduce costs and time by automating tasks that were previously handled manually, increase the profitability of landscape design, and improve the quality of landscaping. I want to.

Among embodiments, an automatic CAD data labeling device for landscape design includes a CAD data receiving unit that receives CAD data of a landscape design drawing from a user terminal; a layer extraction unit that extracts layers for each purpose from the CAD data; a labeling performing unit that extracts objects and outlines for each layer for each purpose, identifies an area, and performs labeling by assigning a color for each purpose to the identified area; and a learning data configuration unit that merges the labeled data for each purpose to form training data for an artificial intelligence model for automating landscape design.

The CAD data receiver may be composed of drawing layers for each purpose in the construction area as the CAD data, divided by lines and colors.

The layer extraction unit may parse the CAD data and extract layers for each purpose based on the layer name and color.

The layer extraction unit may extract layers of buildings, pedestrian paths, driveways, and green space in the construction area based on the layer name, and may extract layers of fire roads, pedestrian paths, and landscape roads based on the layer color.

The labeling unit extracts objects and outlines for each layer for each purpose through image tracking, identifies the area, and performs labeling by assigning a color to the identified area according to the use.

The labeling performing unit may recognize objects through an artificial intelligence object detection model of maskRCNN or CNN.

The learning data configuration unit may merge the data labeled for each purpose according to the priority according to the purpose and generate artificial intelligence learning data.

Among embodiments, a method for automatically labeling CAD data for landscape design includes receiving CAD data of a landscape design drawing from a user terminal; Extracting a layer for each purpose from the CAD data; Extracting objects and outlines for each layer for each purpose, identifying an area, and labeling the identified area with a color for each purpose; and merging the labeled data for each purpose to form training data for an artificial intelligence model for automating landscape design.

The disclosed technology can have the following effects. However, since it does not mean that a specific embodiment must include all of the following effects or only the following effects, the scope of rights of the disclosed technology should not be understood as being limited thereby.

The apparatus and method for automatically labeling CAD data for landscape design according to an embodiment of the present invention can automatically convert CAD data of landscape design drawings into labeling data for artificial intelligence learning.

The automatic CAD data labeling device and method for landscape design according to an embodiment of the present invention can reduce costs and time by automating tasks that were previously handled manually, increase the profitability of landscape design, and improve the quality of landscaping. .

1 is a diagram illustrating an automatic CAD data labeling system according to the present invention.
FIG. 2 is a diagram explaining the system configuration of the automatic labeling device of FIG. 1.
FIG. 3 is a diagram explaining the functional configuration of the automatic labeling device of FIG. 1.
Figure 4 is a flowchart explaining the automatic labeling method of CAD data for landscape design according to the present invention.
5 to 7 are diagrams illustrating an example of the output of each step of the CAD data automatic labeling process.

Since the description of the present invention is only an example for structural or functional explanation, the scope of the present invention should not be construed as limited by the examples described in the text. In other words, since the embodiment can be modified in various ways and can have various forms, the scope of rights of the present invention should be understood to include equivalents that can realize the technical idea. In addition, the purpose or effect presented in the present invention does not mean that a specific embodiment must include all or only such effects, so the scope of the present invention should not be understood as limited thereby.

Meanwhile, the meaning of the terms described in this application should be understood as follows.

Terms such as “first” and “second” are used to distinguish one component from another component, and the scope of rights should not be limited by these terms. For example, a first component may be named a second component, and similarly, the second component may also be named a first component.

When a component is referred to as being “connected” to another component, it should be understood that it may be directly connected to the other component, but that other components may exist in between. On the other hand, when a component is said to be “directly connected” to another component, it should be understood that there are no other components in between. Meanwhile, other expressions that describe the relationship between components, such as “between” and “immediately between” or “neighboring to” and “directly neighboring to”, should be interpreted similarly.

Singular expressions should be understood to include plural expressions unless the context clearly indicates otherwise, and terms such as “comprise” or “have” refer to implemented features, numbers, steps, operations, components, parts, or them. It is intended to specify the existence of a combination, and should be understood as not excluding in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

For each step, identification codes (e.g., a, b, c, etc.) are used for convenience of explanation. The identification codes do not explain the order of each step, and each step clearly follows a specific order in context. Unless specified, events may occur differently from the specified order. That is, each step may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the opposite order.

The present invention can be implemented as computer-readable code on a computer-readable recording medium, and the computer-readable recording medium includes all types of recording devices that store data that can be read by a computer system. . Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage devices. Additionally, the computer-readable recording medium can be distributed across computer systems connected to a network, so that computer-readable code can be stored and executed in a distributed manner.

All terms used herein, unless otherwise defined, have the same meaning as commonly understood by a person of ordinary skill in the field to which the present invention pertains. Terms defined in commonly used dictionaries should be interpreted as consistent with the meaning they have in the context of the related technology, and cannot be interpreted as having an ideal or excessively formal meaning unless clearly defined in the present application.

1 is a diagram illustrating an automatic CAD data labeling system according to the present invention.

Referring to FIG. 1, the CAD data automatic labeling system 100 may include a user terminal 110, an automatic labeling device 130, a database 150, and an external system 170.

The user terminal 110 may correspond to a terminal device operated by a user. In an embodiment of the present invention, a user may be understood as one or more users, and a plurality of users may be divided into one or more user groups. Each of one or more users may correspond to one or more user terminals 110 . That is, the first user is the first user terminal, the second user is the second user terminal,... , the nth (where n is a natural number) user may correspond to the nth user terminal.

Additionally, the user terminal 110 is a device constituting the CAD data automatic labeling system 100 and may correspond to a computing device capable of uploading or directly inputting CAD original data. For example, the user terminal 110 may be implemented as a smartphone, laptop, or computer that can be operated by being connected to the automatic labeling device 130, but is not necessarily limited thereto, and may also be implemented as a variety of devices such as a tablet PC. .

The user terminal 110 may be connected to the automatic labeling device 130 through a network, and a plurality of user terminals 110 may be connected to the automatic labeling device 130 at the same time. In addition, the user terminal 110 can connect to the CAD data automatic labeling system 100 and install and run a dedicated program or application that can use related services.

The automatic labeling device 130 may be implemented as a server corresponding to a computer or program that can automatically generate labeling data for artificial intelligence learning from CAD original data of landscape design drawings. The automatic labeling device 130 may be connected to the user terminal 110 through a wired network or a wireless network such as Bluetooth or WiFi, and may transmit and receive data with the user terminal 110 through the network. Additionally, the automatic labeling device 130 may operate in conjunction with the external system 170 to provide additional functions.

The database 150 may correspond to a storage device that stores various information required during the operation of the automatic labeling device 130. The database 150 can store CAD data about landscape design drawings collected from various sources, store information about landscape design drawing analysis, data labeling automation algorithms, or store information about object labeling, etc., and must do so. Without being limited, the automatic labeling device 130 may store information collected or processed in various forms during the process of automatically labeling CAD data for landscape design.

The external system 170 may correspond to a computing device or server that can provide additional functions in the process of automating the creation of landscape design drawings in conjunction with the automatic labeling device 130. For example, the external system 170 may include a payment system, a messenger system, an authentication system, and a cloud system.

FIG. 2 is a diagram explaining the system configuration of the automatic labeling device of FIG. 1.

Referring to FIG. 2, the automatic labeling device 130 may be implemented including a processor 210, a memory 230, a user input/output unit 250, and a network input/output unit 270.

The processor 210 can execute a procedure that processes each step in the process of operating the automatic labeling device 130, and can manage the memory 230 that is read or written throughout the process. ) can schedule the synchronization time between volatile and non-volatile memory. The processor 210 can control the overall operation of the automatic labeling device 130 and is electrically connected to the memory 230, the user input/output unit 250, and the network input/output unit 270 to control the data flow between them. You can. The processor 210 may be implemented as a central processing unit (CPU) of the automatic labeling device 130.

The memory 230 may be implemented as a non-volatile memory such as a solid state drive (SSD) or a hard disk drive (HDD) and may include an auxiliary memory used to store all data required for the automatic labeling device 130, It may include a main memory implemented as volatile memory such as RAM (Random Access Memory).

The user input/output unit 250 may include an environment for receiving user input and an environment for outputting specific information to the user. For example, the user input/output unit 250 may include an input device including an adapter such as a touch pad, touch screen, on-screen keyboard, or pointing device, and an output device including an adapter such as a monitor or touch screen. In one embodiment, the user input/output unit 250 may correspond to a computing device connected through a remote connection, in which case the automatic labeling device 130 may be performed as an independent server.

The network input/output unit 270 includes an environment for connecting with external devices or systems through a network, for example, Local Area Network (LAN), Metropolitan Area Network (MAN), Wide Area Network (WAN), and VAN ( It may include an adapter for communication such as a Value Added Network).

FIG. 3 is a diagram explaining the functional configuration of the automatic labeling device of FIG. 1.

Referring to FIG. 3, the automatic labeling device 130 may include a CAD data receiving unit 310, a layer extracting unit 330, a labeling performing unit 350, a learning data configuration unit 370, and a control unit 390. there is.

The CAD data receiving unit 310 may receive CAD data related to landscape design from the user terminal 110. In one embodiment, the CAD data receiving unit 310 may receive a file related to a landscape design drawing created through a CAD program from the user terminal 110. Here, the CAD data may correspond to original data related to CAD drawings created at the order receiving stage or at the design stage. That is, the user can upload a file related to the landscape design drawing through the user terminal 110. To this end, the CAD data receiving unit 310 can provide the user terminal 110 with an interface for uploading files, and the user terminal 110 can display the interface on the screen through a dedicated program or application.

In one embodiment, the CAD data receiving unit 310 may receive CAD data in the DWG format file extension used in AutoCAD. Here, AutoCAD is basically used as an example, but it is not necessarily limited to this, and of course, various programs capable of landscape design can be used. A landscape designer uses a CAD program on the user terminal 110 to create a 2D design drawing using construction data as elements such as points, polylines, blocks, and text. Data can be transmitted to the CAD data receiving unit 310. At this time, the CAD data may include all information entered by the user. Here, the CAD data is a 2D landscape design drawing and may include site information for a specific area, including an apartment complex. Target site information may include data on site drawings, building layout drawings, entrance and exit locations, and green space areas. A site drawing is a drawing that includes site information in the relevant area and may include information on site location, area, use, etc. A building layout drawing is a drawing of buildings placed on a site and may include information about the location and width of roads adjacent to the site, the distance between buildings, etc. The location of the entrance may include an entrance at the boundary of the area, an entrance for each building, etc. The green space area may correspond to the total area of green space existing within the area, and can be expressed as the ratio of the green space area to the land area.

In one embodiment, the CAD data receiver 310 may be implemented as a layer of drawings for each purpose of the target site as CAD data. Here, CAD data can be implemented by dividing a plurality of layers, such as a land layer, a building layer, a green space layer, and a road layer, into lines and colors.

The layer extraction unit 330 can extract layers of the received CAD data for each purpose. A layer refers to creating a layer that binds specific objects. By separating specific floors, it is possible to easily delete or move only specific floors. Layers separated into multiple layers are viewed as one on the CAD drawing. In one embodiment, the layer extractor 330 may parse the received CAD data and extract layers for each purpose. The layer extraction unit 330 may decompose a plurality of drawing layers constituting CAD data based on layer information such as color and name of the layer and extract drawing layers for each purpose. At this time, the layer extractor 330 can select and designate a layer for each purpose, such as a construction area, building, green space, pedestrian path, or roadway, from the CAD data.

The labeling unit 350 may extract objects and outlines for the extracted layers for each purpose and perform individual labeling. In one embodiment, the labeling unit 350 may extract objects and outlines for layers for each purpose through image tracking. The labeling unit 350 may extract elements for each purpose by tracking the image for the layer for each purpose. The labeling unit 350 identifies the boundary of the land in the land layer, identifies the boundary of the building within the land in the building layer, specifies a color for each area, and shapes it to recognize areas and objects. In addition, the labeling unit 350 identifies the boundary of the green area within the site in the green area layer and the boundaries of the pedestrian path and roadway in the road layer, assigns a color to each area, and shapes it to recognize areas and objects separately. can do. The labeling performance unit 350 can improve the object recognition rate through maskRCNN and CNN artificial intelligence model learning.

maskRCNN is an object detection model that detects an area (bounding box) where an object is likely to be located, then predicts what category there will be in the detected area, and predicts whether the pixel in the detected area is the predicted object. In particular, maskRCNN is a model that can detect multiple objects in an image using a bounding box and even mask each object using instance segmentation. CNN (Convolutional Neural Network) is a model that can extract features from data and identify patterns in these features.

The labeling unit 350 may perform a labeling operation to identify areas and objects in each layer image and assign appropriate labels so that the CAD data can be used as learning data. Learning data may correspond to a set of layer images containing label information extracted from CAD data. In one embodiment, the labeling unit 350 may perform labeling to designate a color according to the purpose within the identified area and convert the raster image into a vector. Here, the labeled data may correspond to a raster image of a two-dimensional matrix with values between 0 and 255. White can have a value of 255, and black can have a value of 0.

The learning data configuration unit 370 can configure learning data by merging data labeled for each purpose according to the priority according to the purpose. Here, the learning data can be used to learn an artificial intelligence model for automating landscape design. The learning data configuration unit 370 can process labeled data for each purpose into various formats depending on the purpose of use and store them in the database 150 as artificial intelligence learning data. In one embodiment, the learning data configuration unit 370 may label each layer image for each purpose and then reorganize it into an array of desired dimensions.

The control unit 390 controls the overall operation of the automatic labeling device 130 and controls the control flow or Data flow can be managed.

Figure 4 is a flowchart explaining the automatic labeling method for landscape design according to the present invention.

Referring to FIG. 4, the automatic labeling device 130 may receive CAD data of a landscape design drawing from the user terminal 110 through the CAD data receiving unit 310 (step S410). For example, the user can upload original design drawing data in a dwg file through the user terminal 110, and the CAD data reception unit 310 can receive the file from the user terminal 110. .

Additionally, the automatic labeling device 130 can extract layers for each purpose from CAD data through the layer extractor 330 (step S430). At this time, the layer extraction unit 330 can extract layers for each purpose within the target area such as buildings, green spaces, and roads that exist in the CAD data through the lines or names of the layers.

The automatic labeling device 130 may extract objects and outlines for each layer for each purpose extracted through the labeling unit 350 and then perform individual labeling (step S450). For example, the labeling unit 350 can extract objects and outlines from layer images for each purpose and identify areas through artificial intelligence models such as maskRCNN or CNN. The labeling unit 350 may perform labeling by assigning a color for each purpose to the identified area.

Afterwards, the learning data configuration unit 370 may merge the individually labeled data to form training data for an artificial intelligence model for landscape design automation (step S470).

Therefore, the CAD data automatic labeling device 150 according to the present invention can construct artificial intelligence learning data for landscape design automation by graphically converting CAD data, which is a set of lines, into drawing data.

Figures 5 to 7 are diagrams illustrating an example of the output of each step of the CAD data automatic labeling process.

Referring to FIG. 5 , the automatic labeling device 130 may receive CAD data of a landscape design drawing from the user terminal 110. For example, CAD data may correspond to an original drawing file created through a CAD program such as AutoCAD. Referring to FIG. 5, CAD data consists of shapes with each building (apartment) closed on the drawing, and a separate building layer exists. In addition, the pedestrian and fire roads are combined into one layer, and the pedestrian and fire roads are only color-coded and have many open curves. The green space layer exists only in one of the two separate regions, and the green space is made up of a separate solid layer. There is a separate road layer for vehicles, but it contains many open curves.

Referring to FIG. 6, the automatic labeling device 130 can identify and recognize each area and object extracted from the received CAD data. The automatic leveling device 130 can extract layers for each purpose from CAD data through layer name and color. For example, layer names for buildings, green spaces, pedestrian paths, and driveways can be given in relation to their uses, such as BUILDING, GRASS, PAVING, and ROAD, respectively. Referring to Figure 6, the layers for each use extracted from the CAD data of Figure 5, (a) is the construction area, that is, the land, (b) is the building, (c) is the pedestrian path, (d) is the green space, and (e) is the roadway. layers can be extracted. The automatic labeling device 130 can identify areas and objects by deleting unnecessary parts of each element and extracting outlines for each extracted layer for each purpose.

Referring to FIG. 7, the automatic labeling device 130 performs labeling by assigning colors for each purpose to the identified areas and objects, and merges the use-labeled data for each part according to priority to create artificial intelligence learning data. can be created. FIG. 7 shows the results of merging the layers (a) to (e) of FIG. 6 after performing individual labeling.

Although the present invention has been described above with reference to preferred embodiments, those skilled in the art may make various modifications and changes to the present invention without departing from the spirit and scope of the present invention as set forth in the claims below. You will understand that you can do it.

100: CAD data automatic labeling system
110: User terminal 130: Automatic labeling device
150: database 170: external system
210: Processor 230: Memory
250: user input/output unit 270: network input/output unit
310: CAD data receiving unit 330: layer extraction unit
350: Labeling performance unit 370: Learning data configuration unit
390: Control unit

Claims (8)

A CAD data receiving unit that receives CAD data of landscape design drawings from a user terminal;
a layer extraction unit that extracts layers for each purpose from the CAD data;
a labeling performing unit that extracts objects and outlines for each layer for each purpose, identifies an area, and performs labeling to assign a color for each purpose to the identified area; and
A CAD data automatic labeling device for landscape design, including a learning data configuration unit that merges the labeled data for each purpose to form learning data for an artificial intelligence model for landscape design automation.
The method of claim 1, wherein the CAD data receiving unit
A CAD data automatic labeling device for landscape design, characterized in that drawing layers for each purpose in the construction area are divided into lines and colors as the CAD data.
The method of claim 1, wherein the layer extractor
An automatic CAD data labeling device for landscape design, characterized in that it parses the CAD data and extracts layers for each purpose based on layer name and color.
The method of claim 3, wherein the layer extractor
Automatic labeling of CAD data for landscape design, characterized by extracting each layer of buildings, pedestrian paths, driveways, and green space in the construction area based on the layer name, and extracting each layer of fire roads, pedestrian paths, and landscape roads based on layer color. Device.
The method of claim 1, wherein the labeling performing unit
Automatic CAD data for landscape design, characterized by extracting objects and outlines for each layer for each purpose through image tracking, identifying the area, and performing labeling by assigning a color according to the use within the identified area. Labeling device.
The method of claim 5, wherein the labeling performing unit
CAD data automatic labeling device for landscape design, characterized by object recognition through maskRCNN or CNN artificial intelligence object detection model.
The method of claim 1, wherein the learning data configuration unit
An automatic CAD data labeling device for landscape design, characterized in that it merges data labeled for each purpose according to the priority of each purpose and generates it as artificial intelligence learning data.
Receiving CAD data of a landscape design drawing from a user terminal;
Extracting a layer for each purpose from the CAD data;
Extracting objects and outlines for each layer for each purpose, identifying an area, and performing labeling to assign a color for each purpose to the identified area; and
A method of automatically labeling CAD data for landscape design, comprising the step of merging the labeled data for each purpose and forming training data for an artificial intelligence model for landscape design automation.