Design recommendation system for building wall components

Abstract

The present invention provides a design recommendation system for building wall components. The design recommendation system for building wall components provides a structure in which design proposals for building wall components, such as interior and exterior wall panels, windows, and doors, are provided to designers or 3D modeling personnel through an API that implements a deep learning-based design big data analysis process and a generative AI-based design recommendation generation process, and a final 3D shape design is produced through revisions and editing on the basis of the design proposals, such that the time and effort required to develop design proposals are reduced while both product stability and quality are enhanced by providing design optimized proposals tailored to a target product. Also, the design recommendation system for building wall components provides a structure in which information input and output is performed on the basis of input and output of images / drawings or input of text prompt commands within a single design recommendation workspace window, either through a design recommendation application installed on a computing device, such as PCs, laptops, and tablet PCs, or through a design recommendation webpage operated via an online server, and a design screen window of a design program is automatically activated when a design proposal displayed on the design recommendation workspace window is selected, thereby maximizing user convenience and work efficiency.

Description

Building wall composition object design recommendation system

[0001] The present invention relates to a system for recommending design of building wall components, and more specifically, to a system for recommending design of building wall components, such as interior and exterior wall panel materials, windows, and doors, which constitute a building, by providing design drafts for building wall components, such as interior and exterior wall panel materials, windows, and doors, to design experts (designers) or 3D modeling design personnel through an API that implements a deep learning-based design big data analysis process and a generative artificial intelligence-based recommended design generation process, and by providing a structure in which a final 3D shape design is produced through tasks such as supplementation and editing based on the design draft, the time and manpower consumed in the design draft work are minimized, and not only can the stability and quality of the product be expected to be improved as a design draft optimized for the target product is provided, but also, by providing a structure in which input / output of images / drawings or input of text prompt commands is performed within a single design recommendation work screen window of a design recommendation web page operated through a design recommendation application installed on a computing terminal (PC, laptop, tablet PC, etc.) or an online server, and in which the design screen window of a design program is automatically activated when a design draft output on the design recommendation work screen window is selected, the user's work convenience and work efficiency are improved. This is about a recommendation system for designing building wall composition objects that can be maximized.

[0002] The scope of application of industrial pattern design is wide, from the exterior walls of buildings to the packaging of small items. However, the quality of pattern design does not go beyond the personal taste of the designer and the intuitive ideas of the industry producing the product. For this reason, there are difficulties in producing creative patterns.

[0003] Meanwhile, the design industry is mostly comprised of freelancers, and the main design fields of companies utilizing design are visual design (33.96%), service / experience design (19.87%), spatial design (14.06%), and product design (12.88%). However, the reality is that companies utilizing design believe that designers are "essential but not a key element" and "not utilized systematically." In terms of actual on-site personnel, designers only reach the level of associates, assistant managers, and deputy managers / section chiefs among all designers, and most are college graduates by education level and those in their 20s and 30s by career experience. The most common reason for the difficulty in recruiting design personnel is that "their expectations, such as wages and working hours, do not match", making recruiting design personnel very difficult.

[0004] In this situation, not only is there no systematic database of products to be manufactured, but it's also extremely difficult to commission or find designs that meet the various requirements for implementing designs for various products, such as product characteristics, product size, and customer design preferences. To overcome these challenges, a system that reflects the technology needed to recommend designs that are both systematic and customer-focused is needed.

[0005] [Prior Art Literature]

[0006] (Patent Document 1) Republic of Korea Patent Gazette No. 10-2533499, "Artificial Intelligence-Based Customized Interior Design Recommendation Platform."

[0007] (Patent Document 2) Republic of Korea Patent Publication No. 10-2020-0055813, "Automatic Industrial Pattern Design Generation System Using Deep Learning"

[0008] (Patent Document 3) Republic of Korea Patent Publication No. 10-2023-0146345, "Design Recommendation and Final Draft Provision System for Products to Be Released."

[0009] Accordingly, the present invention aims to provide a new type of design recommendation system for building wall components, which improves the problems of the prior art by 1) providing a design draft for a building wall object to a user of a design recommendation program module - a design expert (designer) or 3D modeling design personnel - through a deep learning-based design big data analysis process and a generative artificial intelligence-based recommended design creation process through a deep learning process management unit and a design recommendation management unit that can be implemented through an open API, and 2) providing a structure in which a design draft provided in the form of an image file such as a JPG is automatically loaded into a design program and then the user performs a design draft supplementation and editing work to produce a final 3D shape design, thereby minimizing the time and manpower consumed in the design draft work, and providing a design draft optimized for the target product, thereby enabling the stability and quality of the product to be improved.

[0010] In addition, the present invention provides a new type of building wall composition object design recommendation system in which input / output of images / drawings or input / output of information according to input of text prompt commands is performed in a single design recommendation work screen window of a design recommendation application installed on a computing terminal (PC, laptop, tablet PC, etc.) or a design recommendation web page operated through an online server, and a structure in which the design screen window of a design program is automatically activated when a design draft output in the design recommendation work screen window is selected, thereby maximizing the user's work convenience and work efficiency.

[0011] According to a feature of the present invention for achieving the above-described object, the present invention provides a design recommendation system for a building wall component object, characterized by comprising a deep learning process management unit (100) that collects information related to a building wall surface to generate building wall-big data, sets elements (design components) constituting the design of a building wall surface, and performs deep learning on the building wall-big data according to the design components to generate a design recommendation program module (10); a design recommendation management unit (200) that receives the design recommendation program module (10) from the deep learning process management unit (100), transmits recommended design request information input by a user to the design recommendation program module (10), and outputs a plurality of design drafts generated according to the operation of the design recommendation program module (10); and a user terminal unit (300) that is operated by a user, inputs the recommended design request information, and receives design drafts in conjunction with the design recommendation management unit (200).

[0012]

[0013] In the building wall component design recommendation system according to the present invention, the deep learning process management unit (100) comprises: a building wall image DB (110) in which building wall images including photographed images, design images, and graphic images of building walls are collected and stored in a database; an building wall object design information setting module (120) in which objects forming a building wall, elements constituting the design of each object (object design components), overall design characteristic information, and characteristic information for each object design component are set, wherein the objects include wall panels, windows, and doors, the object design components include shape, shape, color, material, and thickness, the overall design characteristic information includes at least one selected from style and concept, and the characteristic information for each object design component includes shape characteristic information, shape characteristic information, color characteristic information, material characteristic information, and thickness characteristic information; It may be configured to include a building wall-big data DB (130) in which building wall images of the building wall image DB (110) are classified based on objects, object design components, overall design characteristic information, and characteristic information for each object design component set in the building wall object design information setting module (120) and building wall-big data is stored in a database; a deep learning-based recommendation program generation module (140) that generates a design recommendation program module (10) by performing deep learning on the building wall-big data through a deep learning-based artificial intelligence learning algorithm (20).

[0014]

[0015] In the building wall component design recommendation system according to the present invention, the user terminal unit (300) receives, from the user, a recommendation design request information including 1) building specific information including an object to be provided with a design draft, 2) object specific information to be provided with a design draft, 3) new construction-remodeling design distinction information distinguishing whether it corresponds to one selected from new construction design and remodeling design, 4) keyword information linked to overall design characteristic information (overall design characteristic keyword information), 5) keyword information linked to characteristic information for each object design component (object design keyword information), and 6) keyword information linked to an object for which weighting is desired or an object design component for which weighting is desired (weight keyword information).

[0016] The above building specific information is entered as one or more pieces of information selected from the group of image information, design drawing information, and command text information.

[0017] The above object specific information, new construction-remodeling design distinction information, overall design characteristic keyword information, object design keyword information, and weight keyword information may be entered as command text information.

[0018]

[0019] In the building wall composition object design recommendation system according to the present invention, the design recommendation management unit (200) comprises: an image / drawing input management module (210) that manages an image / drawing input / output window (2) into which image information and design drawing information corresponding to the building specific information are input; a text prompt management module (220) that manages a command input window (3) into which command text information, object specific information, new construction-remodeling design distinction information, design overall characteristic keyword information, object design keyword information, and weighted keyword information corresponding to the building specific information are input; a building image extraction module (230) that extracts a building image corresponding to the building specific information transmitted from the text prompt management module (220) through a real-time online search or from the building wall image DB (110); An object blur processing module (240) that receives a building image from the image / drawing input management module (210) and the building image extraction module (230), and blurs a corresponding object portion within the building image according to object specific information transmitted from the text prompt management module (220); A blur processing-building image output module (250) that outputs the object-blurred building image transmitted from the object blur processing module (240) to the image / drawing input / output window (2); A design draft output module (260) that inputs the object specific information, new construction-remodeling design classification information, design overall characteristic keyword information, object design keyword information, and weighted keyword information transmitted from the text prompt management module (220) to the design recommendation program module (10), receives a plurality of design drafts generated from the design recommendation program module (10), and outputs them to the image / drawing input / output window (2);It may be configured to include a design draft storage management module (260') that stores a plurality of design drafts generated from the above design recommendation program module (10), and classifies and lists them based on objects, object design components, overall design characteristics, characteristics of each object design component, etc., and creates a database; a design draft selection management module (270) that manages the input of user selection information for selecting one of the plurality of design drafts; and a design draft application-building image output module (280) that removes object blurring from a building image output to the image / drawing input / output window (2), applies the design draft selected by the input of user selection information to the corresponding object part, and outputs the resulting building image with the design draft applied to the image / drawing input / output window (2).

[0020]

[0021] The building wall surface composition object design recommendation system according to the present invention further includes a design program unit (400) installed in the user terminal unit (300) and performing 3D modeling design of building wall surface object design;

[0022] The above design recommendation management unit (200) further includes a design program activation module (290) that automatically activates the design screen window (4) of the design program unit (400) when a building image with design draft applied, which is output to the image / drawing input / output window (2) by the design draft application-building image output module (280), is selected by the user, thereby enabling 3D modeling design work to be continuously performed for the design draft finally selected by the user.

[0023]

[0024] In the building wall composition object design recommendation system according to the present invention, the image / drawing input management module (210), text prompt management module (220), building image extraction module (230), object blur processing module (240), blur processing-building image output module (250), design draft output module (260), design draft selection management module (270), draft application-building image output module (280), and design program activation module (290) of the design recommendation management unit (200) are,

[0025] It may be a component for driving any one selected from among a design recommendation application (700) installed on a computing terminal (500) that may include a PC, a laptop, a smart pad, and a smartphone; and a design recommendation web page (800) provided through an online server (600) connected to an online network.

[0026]

[0027] In the building wall composition object design recommendation system according to the present invention, the design recommendation application (700) and the design recommendation web page (800) are formed by dividing the image / drawing input / output window (2) and the command input window (3) within a single design recommendation work screen window (1).

[0028] Input of image / design drawing information, input of command text information / object specific information / new construction-remodeling design classification information / design overall characteristic keyword information / object design keyword information / weight keyword information, output of object-blurred building image, output of multiple design drafts, input of user selection information for multiple design drafts, output of building image with design draft applied can be performed within a single design recommendation work screen window (1).

[0029] The design recommendation program module provided by the architectural wall component design recommendation system of the present invention enables manufacturers and designers to quickly and accurately simulate and manufacture designs with desired shapes, shortening product development cycles, reducing costs, and improving product quality and performance. Furthermore, it is expected to enhance product stability and quality by proactively identifying unforeseen issues.

[0030] Figure 1 is a basic configuration block diagram of a building wall composition object design recommendation system according to the present invention;

[0031] FIG. 2 is a detailed block diagram of a deep learning process management unit according to an embodiment of the present invention;

[0032] FIG. 3 is a detailed block diagram of a design recommendation management unit according to an embodiment of the present invention;

[0033] FIG. 4 is a conceptual diagram showing a design program activation structure of a building wall composition object design recommendation system according to an embodiment of the present invention;

[0034] FIG. 5 is a drawing showing a modeling tool selection configuration of a building wall surface composition object design recommendation system according to an embodiment of the present invention;

[0035] FIG. 6 is a conceptual diagram showing that a building wall composition object design recommendation system according to an embodiment of the present invention is implemented as an application or web page.

[0036] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings. In the drawings and detailed description, descriptions and references to structures and operations readily apparent to those skilled in the art have been simplified or omitted. In particular, in the drawings and detailed descriptions, detailed descriptions and illustrations of specific technical structures and operations of elements not directly related to the technical features of the present invention have been omitted, and only technical structures related to the present invention have been briefly illustrated or described.

[0037]

[0038] The building wall composition object design recommendation system according to an embodiment of the present invention is configured to include a deep learning process management unit (100), a design recommendation management unit (200), and a user terminal unit (300), as shown in FIG. 1.

[0039]

[0040] The deep learning process management unit (100) is a module that collects information related to a building wall surface to create building wall-big data, sets elements (design components) that constitute the design of a building wall surface, and performs deep learning on the building wall-big data according to the design components to create a design recommendation program module (10).

[0041] Here, the deep learning process management unit (100) according to the embodiment of the present invention is configured to include a building wall image DB (110), a building wall object design information setting module (120), a building wall-big data DB (130), and a deep learning-based recommendation program creation module (140), as shown in FIG. 2.

[0042] The building wall image DB (110) is a DB in which building wall images including photographed images, design images, and graphic images of building walls are collected and stored in a database.

[0043] The building wall object design information setting module (120) is a module in which objects forming a building wall, elements forming the design of each object (object design components), overall design characteristic information, and characteristic information for each object design component are set. Here, objects include wall panels, windows, doors, etc., object design components include shape, form, color, material, and thickness, overall design characteristic information includes style, concept, etc., and characteristic information for each object design component includes shape characteristic information, shape characteristic information, color characteristic information, material characteristic information, and thickness characteristic information.

[0044] The building wall-big data DB (130) is a DB in which building wall-big data is created by classifying building wall images of the building wall image DB (110) based on objects, object design components, overall design characteristic information, and characteristic information for each object design component set in the building wall object design information setting module (120).

[0045] The deep learning-based recommendation program generation module (140) is a module that generates a design recommendation program module (10) by performing deep learning on building wall-big data using a deep learning-based artificial intelligence learning algorithm (20). The deep learning-based recommendation program generation module (140) can be implemented through an open API.

[0046]

[0047] The design recommendation management unit (200) receives a design recommendation program module (10) from a deep learning process management unit (100), transmits recommended design request information input by a user to the design recommendation program module (10), and outputs multiple design drafts generated according to the operation of the design recommendation program module (10).

[0048]

[0049]

[0050] *The design recommendation management unit (200) according to the embodiment of the present invention is configured to include an image / drawing input management module (210), a text prompt management module (220), a building image extraction module (230), an object blur processing module (240), a blur processing-building image output module (250), a design draft output module (260), a design draft storage management module (260'), a design draft selection management module (270), and a draft application-building image output module (280), as shown in FIG. 3.

[0051] The image / drawing input management module (210) is a module that manages the image / drawing input / output window (2) into which image information and design drawing information corresponding to building-specific information are input. Building-specific information may include photos of the location where a building is located, photos of a building, photos of a building wall, photos of objects constituting a building wall, etc. In the case of design drawings, the design drawings of the building / building wall / building wall constituent objects may be used as building-specific information.

[0052] The text prompt management module (220) is a module that manages a command input window (3) into which command text information corresponding to building-specific information, object-specific information, new-remodeling design distinction information, overall design characteristic keyword information, object design keyword information, and weighted keyword information are entered. The text prompt management module (220) can receive and process information in natural language. Here, the building-specific information may be building location information, building name, etc., the object-specific information may be an object name, and the new-remodeling design distinction information may be a word / phrase / sentence that distinguishes whether the object for which a design draft is created is new or remodeled. The overall design characteristic keyword information may be a word / phrase / sentence that defines a style, concept, etc., and the object design keyword may be a word / phrase / sentence that defines shape, appearance, color, material, and thickness, which are object design components, and a word / phrase / sentence that defines shape characteristic information, shape characteristic information, color characteristic information, material characteristic information, and thickness characteristic information. Weighted keyword information can be words / phrases / sentences that define elements that are emphasized or key points when creating a design draft, for example, it can be expressed as the sentence "Make the windows stand out more."

[0053] The building image extraction module (230) is a module in which building images corresponding to building specific information transmitted from the text prompt management module (220) are extracted through real-time online search or from the building wall image DB (110).

[0054] The object blur processing module (240) receives a building image from the image / drawing input management module (210) and the building image extraction module (230), and is a module that blurs the corresponding object portion within the building image according to object specific information transmitted from the text prompt management module (220).

[0055] The blur processing-architectural image output module (250) is a module that outputs an object-blurred architectural image transmitted from the object blur processing module (240) to the image / drawing input / output window (2).

[0056] The design draft output module (260) is a module that inputs object specific information, new construction-remodeling design classification information, overall design characteristic keyword information, object design keyword information, and weight keyword information transmitted from the text prompt management module (220) into the design recommendation program module (10), receives multiple design drafts generated from the design recommendation program module (10), and outputs them to the image / drawing input / output window (2).

[0057] The design draft storage management module (260') is a module that stores multiple design drafts generated from the design recommendation program module (10), classifies them based on objects, object design components, overall design characteristics, and characteristics of each object design component, and then lists them and creates a database.

[0058] The design draft selection management module (270) manages the input of user selection information for selecting one of multiple design drafts. Selection of a design draft can be performed by mouse click, etc.

[0059] The design application-building image output module (280) is a module that removes object blurring from the building image output to the image / drawing input / output window (2), applies the design draft selected by inputting user selection information to the corresponding object part, and outputs the resulting building image with the design draft applied to the image / drawing input / output window (2).

[0060]

[0061] The user terminal unit (300) is operated by the user, and inputs the recommended design request information and receives the design draft by linking with the design recommendation management unit (200). The user terminal unit (300) receives the recommended design request information from the user, which includes 1) building specific information including the object to receive the design draft, 2) object specific information to receive the design draft, 3) new construction-remodeling design distinction information that distinguishes whether it corresponds to one selected from new construction design and remodeling design, 4) keyword information linked to overall design characteristic information (overall design characteristic keyword information), 5) keyword information linked to characteristic information for each object design component (object design keyword information), and 6) keyword information linked to the object for which weighting is desired or the object design component for which weighting is desired (weight keyword information). Building specific information is entered as one or more pieces of information selected from the group of image information, design drawing information, and command text information, and object specific information, new construction-remodeling design distinction information, overall design characteristic keyword information, object design keyword information, and weighted keyword information are entered as command text information.

[0062]

[0063] Meanwhile, the building wall surface composition object design recommendation system according to an embodiment of the present invention can be linked with a design program unit (400) as shown in FIG. 4. The design program unit (400) is installed in a user terminal unit (300) and performs 3D modeling design of building wall surface object design.

[0064] Here, the design recommendation management unit (200) according to the embodiment of the present invention is provided with a design program activation module (290). When a user selects a building image with a design draft applied, which is output to the image / drawing input / output window (2) by the design draft application-building image output module (280), the design program activation module (290) automatically activates the design screen window (4) of the design program unit (400). Through this, 3D modeling design work for the design draft finally selected by the user can be continuously performed.

[0065] Meanwhile, the design program activation module (290) can enable selection of the type of design program to be automatically activated through a separate modeling tool selection menu window (1a) that is output together when a building image with a design draft applied is output in the image / drawing input / output window (2) as shown in Fig. 5. That is, when the program name indicated in the modeling tool selection menu window (1a) is selected, the corresponding program is automatically activated. The types of design programs may include 2D / 3D model design software or applications such as Photoshop, Illustrator, Solidworks, and Inventor.

[0066]

[0067] The image / drawing input management module (210), text prompt management module (220), building image extraction module (230), object blur processing module (240), blur processing-building image output module (250), design draft output module (260), design draft selection management module (270), draft application-building image output module (280), and design program activation module (290) of the design recommendation management unit (200) constituting the building wall composition object design recommendation system according to the embodiment of the present invention constitute components for operating the design recommendation application (700) or the design recommendation web page (800) as shown in FIG. 5.

[0068] The design recommendation application (700) is installed on a computing terminal (500) that may include a PC, laptop, smart pad, or smartphone, and the design recommendation webpage (800) is provided through an online server (600) connected to an online network.

[0069] Meanwhile, the design recommendation application (700) and the design recommendation web page (800) according to the embodiment of the present invention are formed so that 1) an image / drawing input / output window (2) and a command input window (3) are divided and formed within a single design recommendation work screen window (1), as shown in FIG. 6, and 2) input of image / design drawing information, input of command text information / object specific information / new construction-remodeling design classification information / design overall characteristic keyword information / object design keyword information / weight keyword information, output of an object-blurred building image, output of multiple design drafts, input of user selection information for multiple design drafts, and output of an image of a building with a design draft applied are performed within the single design recommendation work screen window (1).

[0070]

[0071] The building wall composition object design recommendation system according to the embodiment of the present invention configured as described above 1) provides a design draft for a building wall object to a user of the design recommendation program module - a design expert (designer) or 3D modeling design personnel - through a deep learning-based design big data analysis process and a generative artificial intelligence-based recommended design creation process through a deep learning process management unit and a design recommendation management unit that can be implemented through an open API, and 2) provides a structure in which a design draft provided in the form of an image file such as JPG is automatically loaded into a design program and then the user performs a design draft supplementation and editing work to produce a final 3D shape design, thereby minimizing the time and manpower consumed in the design draft work, and providing a design draft optimized for the target product, so that the stability and quality of the product can be expected to be improved.

[0072] In addition, the building wall composition object design recommendation system according to an embodiment of the present invention provides a structure in which information input / output is performed according to input of images / drawings or text prompt commands within a single design recommendation work screen window of a design recommendation application installed on a computing terminal (PC, laptop, tablet PC, etc.) or a design recommendation web page operated through an online server, and the design screen window of a design program is automatically activated when a design draft output in the design recommendation work screen window is selected, thereby maximizing the user's work convenience and work efficiency.

[0073]

[0074] Although the building wall component design recommendation system according to the embodiment of the present invention as described above has been illustrated in accordance with the above description and drawings, this is merely an example, and those skilled in the art will readily understand that various changes and modifications are possible within the scope that does not depart from the technical spirit of the present invention.

Claims

1. A deep learning process management unit (100) that collects information related to a building wall surface to create building wall-big data, sets elements (design components) that constitute the design of a building wall surface, and performs deep learning on the building wall-big data according to the design components to create a design recommendation program module (10); A design recommendation management unit (200) that receives the design recommendation program module (10) from the deep learning process management unit (100), transmits the recommended design request information input by the user to the design recommendation program module (10), and outputs a plurality of design drafts generated according to the operation of the design recommendation program module (10); A building wall composition object design recommendation system characterized by comprising a configuration including a user terminal unit (300) that is operated by a user, inputs the above-mentioned recommended design request information, and receives a design draft in conjunction with the above-mentioned design recommendation management unit (200).

2. In paragraph 1, The above deep learning process management unit (100) is Building wall image DB (110) in which building wall images including photographic images, design images, and graphic images of building walls are collected and stored in a database; An object forming a building wall, an element forming the design of each object (object design component), overall design characteristic information, and characteristic information for each object design component are set, wherein the object includes a wall panel, a window, and an entrance door, and the object design component includes shape, shape, color, material, and thickness, and the overall design characteristic information includes at least one selected from style and concept, and the characteristic information for each object design component includes shape characteristic information, shape characteristic information, color characteristic information, material characteristic information, and thickness characteristic information. A building wall object design information setting module (120); A building wall-big data DB (130) in which building wall-big data is databased by classifying building wall images of the building wall image DB (110) based on objects, object design components, overall design characteristic information, and characteristic information for each object design component set in the building wall object design information setting module (120); A building wall composition object design recommendation system characterized by comprising a configuration including a deep learning-based recommendation program generation module (140) that generates a design recommendation program module (10) by performing deep learning on the building wall-big data through a deep learning-based artificial intelligence learning algorithm (20).

3. In paragraph 2, The above user terminal unit (300) is 1) Building specific information including the object to be provided with the design draft, 2) Object specific information to be provided with the design draft, 3) New construction-remodeling design classification information distinguishing whether it corresponds to one of the new construction design and the remodeling design, 4) Keyword information linked to overall design characteristic information (overall design characteristic keyword information), 5) Keyword information linked to characteristic information for each object design component (object design keyword information), 6) Recommendation design request information including keyword information linked to the object for which weight is to be applied or the object design component for which weight is to be applied (weight keyword information) is input by the user. The above building specific information is entered as one or more pieces of information selected from the group of image information, design drawing information, and command text information. A building wall composition object design recommendation system characterized in that the above object specific information, new construction-remodeling design distinction information, design overall characteristic keyword information, object design keyword information, and weight keyword information are input as command text information.

4. In paragraph 3, The above design recommendation management unit (200) is An image / drawing input management module (210) that manages an image / drawing input / output window (2) into which image information and design drawing information corresponding to the above building specific information are input; A text prompt management module (220) that manages a command input window (3) in which command text information corresponding to the above building specific information, object specific information, new construction-remodeling design distinction information, design overall characteristic keyword information, object design keyword information, and weight keyword information are entered; A building image extraction module (230) in which a building image corresponding to building specific information transmitted from the above text prompt management module (220) is extracted through real-time online search or extracted from the building wall image DB (110); An object blur processing module (240) that receives a building image from the image / drawing input management module (210) and the building image extraction module (230) and blurs a corresponding object portion within the building image according to object specific information transmitted from the text prompt management module (220); A blur processing-building image output module (250) that outputs an object blur processed building image transmitted from the object blur processing module (240) to an image / drawing input / output window (2); A design draft output module (260) that inputs the object specific information, new construction-remodeling design classification information, design overall characteristic keyword information, object design keyword information, and weight keyword information transmitted from the text prompt management module (220) into the design recommendation program module (10), receives a plurality of design drafts generated from the design recommendation program module (10), and outputs them to the image / drawing input / output window (2); A design draft storage management module (260') that stores multiple design drafts generated from the above design recommendation program module (10), classifies them based on objects, object design components, overall design characteristics, characteristics of each object design component, etc., and lists them in a database; A design draft selection management module (270) that manages the input of user selection information for selecting one of multiple design drafts; A building wall composition object design recommendation system characterized by comprising a configuration including a design application-building image output module (280) that removes object blurring from a building image output in an image / drawing input / output window (2), applies a design draft selected by inputting user selection information to a corresponding object part, and outputs a building image with a design draft applied thereto to the image / drawing input / output window (2).

5. In paragraph 4, It further includes a design program unit (400) installed in the user terminal unit (300) and performing 3D modeling design of a building wall object design; The above design recommendation management unit (200) is A building wall composition object design recommendation system characterized in that it further includes a design program activation module (290) that automatically activates the design screen window (4) of the design program unit (400) when a building image with design draft applied, which is output to the image / drawing input / output window (2) by the design draft application-building image output module (280), is selected by the user, thereby enabling 3D modeling design work to be continuously performed for the design draft finally selected by the user.

6. In paragraph 5, The image / drawing input management module (210), text prompt management module (220), building image extraction module (230), object blur processing module (240), blur processing-building image output module (250), design draft output module (260), design draft selection management module (270), draft application-building image output module (280), and design program activation module (290) of the above design recommendation management unit (200) are A design recommendation system for a building wall composition object, characterized in that it is a component for driving any one selected from among a design recommendation application (700) installed on a computing terminal (500) that may include a PC, a laptop, a smart pad, and a smartphone; and a design recommendation web page (800) provided through an online server (600) connected to an online network.

7. In paragraph 6, The design recommendation application (700) and the design recommendation webpage (800) are formed by dividing the image / drawing input / output window (2) and the command input window (3) within a single design recommendation work screen window (1). A building wall composition object design recommendation system characterized in that input of image / design drawing information, input of command text information / object specific information / new construction-remodeling design classification information / design overall characteristic keyword information / object design keyword information / weight keyword information, output of a building image with object blurring, output of multiple design drafts, input of user selection information for multiple design drafts, and output of a building image with design draft applied are performed within a single design recommendation work screen window (1).

Images