Renovation support system, renovation support method, and program

Abstract

It automates everything from structural analysis of floor plans to design, estimates, and recommendations of construction candidates, streamlining renovation work all at once. [Solution] The renovation support system 1 includes a renovation support device 100 that proposes renovation plans based on floor plan images of existing dwelling units. The renovation support device 100 analyzes floor plan images acquired from a user terminal 200 via a network N using an image analysis unit 102 to generate structural information, creates an optimal renovation floor plan proposal using a generation AI 400 using a plan generation unit 103, and further calculates the types and quantities of building materials using an estimation unit 104 to output a construction estimate, thereby streamlining the entire process from design, estimation, and company selection all at once.

Description

[Technical Field]

[0001] The present invention relates to a renovation support system, a renovation support method, and a program. [Background technology]

[0002] In recent years, with the expansion of the market for reuse and renovation of used homes, there has been an increasing demand for systems that propose new renovation plans based on floor plans of existing homes (for example, Patent Documents 1 and 2).

[0003] Patent Document 1 discloses a technology that uses AI to analyze floor plans, calculate the necessary building materials and equipment based on the type and dimensions of the space, and automatically create an estimate for renovation work.

[0004] Furthermore, Patent Document 2 provides an interface that allows users to select floor plans and furniture arrangements on the web and change rental conditions according to the selected contents. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2023-119559 [Patent Document 2] Patent No. 7560933 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the conventional technologies mentioned above only provided partial support and did not provide comprehensive support for the entire renovation process, including design, estimates, and construction proposals. In other words, conventional renovation support technologies separate multiple processes, such as generating design proposals, evaluating constraints, creating estimates, and submitting proposals to construction companies, and this lacks efficiency and consistency in allowing users and businesses to make consistent decisions and proposals.

[0007] The present invention was made in consideration of these circumstances, and aims to provide a renovation support system, renovation support method, and program that can automatically process everything from structural analysis of floor plans to design, estimates, and recommendation of construction candidates, thereby streamlining renovation work all at once. [Means for solving the problem]

[0008] In order to solve the above problems, the present invention is a renovation support system that proposes a renovation plan based on a floor plan image of an existing dwelling unit, and is characterized by comprising: an image analysis means that recognizes from the floor plan image a plurality of components that make up the interior of the dwelling unit to be renovated and generates structural information including room dimensions and wall thickness; a plan generation means that uses a generation AI to generate a plurality of renovation floor plan proposals including furniture placement information based on the structural information, calculates an evaluation value for each floor plan proposal in light of predetermined design constraints and traffic flow evaluation criteria, and selects the renovation floor plan proposal with the highest suitability based on the evaluation value; and an estimation means that calculates the type and quantity of necessary building materials based on the components included in the selected renovation floor plan proposal, calculates a construction estimate based on unit price information for each material and labor cost information for workers, and generates a construction estimate including the construction estimate.

[0009] In addition, in order to solve the above-mentioned problems, the present invention is a computer-based renovation support method that proposes a renovation plan using an input floor plan image of an existing dwelling unit, and is characterized by having the following steps: an image analysis step that recognizes, from the floor plan image, multiple components that make up the interior of the dwelling unit to be renovated and generates structural information including room dimensions and wall thickness; a plan generation step that uses a generation AI to generate multiple renovation floor plan proposals including furniture placement information based on the structural information, calculates an evaluation value for each floor plan proposal in light of specified design constraints and traffic flow evaluation criteria, and selects the renovation floor plan proposal with the highest suitability based on the evaluation value; and an estimation step that calculates the type and quantity of necessary building materials based on the components included in the selected renovation floor plan proposal, calculates a construction estimate based on unit price information for each material and labor cost information for workers, and generates a construction estimate including the construction estimate.

[0010] Furthermore, in order to solve the above-mentioned problems, the present invention provides a computer-readable program to be executed by a renovation support system that proposes renovation plans based on floor plan images of existing dwelling units, and is characterized by executing the following steps: an image analysis step that recognizes, from the floor plan image, multiple components that make up the interior of the dwelling unit to be renovated and generates structural information including room dimensions and wall thickness; a plan generation step that uses a generation AI to generate multiple renovation floor plan proposals including furniture placement information based on the structural information, calculates an evaluation value for each floor plan proposal in light of specified design constraints and traffic flow evaluation criteria, and selects the renovation floor plan proposal with the highest suitability based on the evaluation value; and an estimation step that calculates the type and quantity of necessary building materials based on the components included in the selected renovation floor plan proposal, calculates a construction estimate based on unit price information for each material and labor cost information for workers, and generates a construction estimate including the construction estimate. [Effects of the Invention]

[0011] According to the present invention, the entire renovation process can be automated, from structural analysis of floor plans to design, estimates, and recommendation of construction candidates, making it possible to streamline renovation work all at once. More specifically, this invention provides a system that uses a generation AI to generate, evaluate, and select multiple new floor plan proposals that take into consideration furniture placement and daily activity patterns based on floor plan images of existing dwellings, and that can comprehensively and automatically perform a series of renovation support processes based on the proposed floor plans, including creating estimates, outputting renderings of the completed project (CG perspective drawings), and recommending construction companies. This allows users to perform everything from structural analysis of floor plans to design, estimates, and selection of construction candidates in one go with minimal operations, resulting in the entire renovation work workflow, which was previously separated, being integrated and processed quickly with high consistency. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a block diagram showing the configuration of a renovation support system according to one embodiment of the present invention. [Figure 2] 1 is a block diagram showing a renovation support device according to an embodiment of the present invention; [Figure 3] 3 is a flowchart showing the overall processing procedure of the renovation support device according to one embodiment of the present invention. [Figure 4] 10 is a flowchart for explaining details of the analysis process (step S200) by the image analysis unit of the renovation support device according to one embodiment of the present invention. [Figure 5] 10 is a flowchart for explaining details of a renovation plan generation process (step S300) by a plan generation unit of a renovation support device according to an embodiment of the present invention. [Figure 6] 10 is a flowchart for explaining details of a construction estimate generation process (step S400) performed by an estimating unit of the renovation support device according to one embodiment of the present invention. [Figure 7]10 is a flowchart for explaining details of a matching process (step S500) performed by a matching unit of the renovation support apparatus according to one embodiment of the present invention. [Figure 8] 10 is a flowchart for explaining details of a CG perspective generation process (step S600) performed by a CG generation unit of a renovation support apparatus according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, like components are designated by like reference numerals, and the description thereof will be omitted as appropriate. In the following description, each component of each device or system represents a functional block, not a hardware configuration, unless otherwise specified. Each component of each device or system is realized by any combination of hardware and software, centered around the CPU of any computer, memory, a program loaded into the memory, a storage medium such as a hard disk that stores the program, and a network connection interface. There are many variations in the realization methods and devices.

[0014] [1. Configuration of Renovation Support System 1] Figure 1 is a block diagram showing the configuration of a renovation support system 1 according to one embodiment of the present invention. The renovation support system 1 shown in Figure 1 is an information processing system that analyzes floor plan information of an existing dwelling unit to be renovated based on input operations by a user from a user terminal 200, and automatically and comprehensively supports a series of renovation tasks, such as design proposals, creating estimates, recommending construction companies, and outputting renderings of the completed unit. The "dwelling unit to be renovated" here refers to the dwelling space that is treated as the target of design and construction in the renovation support system 1, and is a concept that includes the interior space of a single dwelling unit in an apartment building or a detached house.

[0015] As shown in Figure 1, in the renovation support system 1 of this embodiment, a renovation support device 100 analyzes floor plan information of existing dwelling units and provides a series of renovation support services such as design proposals, creating estimates, recommending construction companies, and outputting renderings of completed units, a user terminal 200, a group of databases (DBs) 300, and a generation AI 400 are configured to be mutually connectable via a network N such as the Internet.

[0016] [1.1. Configuration and Function of Renovation Support Device 100] The renovation support device 100 is a device that provides services such as providing renovation plans to users of the renovation support system 1, for example in SaaS format, and is a computer or server that generates renovation plans, etc. using a generation AI 400 based on a prompt, which is text entered by the user on the user terminal 200.

[0017] The renovation support device 100 receives floor plan images as input from the user terminal 200 and incorporates multiple processing means for consistently executing processes such as extracting structural information, generating renovation proposals, creating estimates, matching potential construction companies, and outputting CG perspective drawings. These processes are shared among various modules (various means) implemented within the system, and are configured to be processed automatically while coordinating with each other. Here, the renovation support device 100 mainly includes the following functional blocks.

[0018] [1.1.1. Configuration and Function of Input / Output Unit 101] The input / output unit 101 has an interface function for transmitting and receiving information between the user terminal 200 and the renovation support apparatus 100 . Specifically, the input / output unit 101 receives floor plan image data uploaded from the user terminal 200 and renovation conditions specified by the user (desired floor plan, family composition, preferred equipment, etc.), and supplies the data to internal processing modules such as the image analysis unit 102 and the plan generation unit 103.

[0019] In addition, the input / output unit 101 transmits structural information, floor plan proposals, construction estimates, matching candidate lists, CG perspective drawings of completed projects, etc., generated as a result of internal processing to the user terminal 200, and provides them in a format that can be displayed or downloaded through a specified user interface screen. Here, the input / output unit 101 may be configured to present the status of ongoing processing and completion notifications to the user in real time depending on the overall system usage and processing status, and may include control logic that enables interactive dialogue with the system. In this way, the input / output unit 101 is a component that serves as the starting point and the ending point of the entire processing flow as a contact point with the user, and also functions as a hub for data exchange in cooperation with other internal processing means.

[0020] [1.1.2. Configuration and Function of Image Analysis Unit 102] The image analysis unit 102 is a processing module that acquires as input from the user terminal 200, for example, a floor plan image corresponding to a single unit in an apartment building, and generates "structural information" for quantitatively grasping the spatial configuration of the unit.

[0021] The image analysis unit 102 first automatically detects elements such as walls, rooms, doors, windows, and storage areas contained in the floor plan image using image processing and image recognition algorithms, and models the spatial configuration by analyzing the geometric shapes, positional relationships, and connection relationships of these elements. Next, the image analysis unit 102 calculates dimensional attributes such as the area of each room, the length of walls, the placement of openings, and the width of corridors based on the detected components, and converts these into numerical data as structural information. The structural information obtained here is used by the plan generation unit 103 to generate floor plan proposals and evaluate furniture placement.

[0022] Furthermore, the image analysis unit 102 can extract at least one of secondary information contained in the floor plan image, namely, annotation information (e.g., room name, room number, direction symbol, etc.), scale description (scale line, numerical scale, etc.), and room name (living room, Japanese-style room, bedroom, etc.), and use this to perform the following correction or complementation processing: Drawing scaling based on scale description (pixel-based dimensions converted to actual size) Room classification based on room name (labeling of space use based on functional classification) - Consistency check using information obtained from annotations (preventing false positives and complementing room attributes) In other words, these processes function as "pre-processing complementation functions" and constitute technical means for scaling structural information, complementing room attributes, and aligning dimensional values using the extracted secondary information. Note that structural information includes the positional relationships, sizes, and layout coordinates of each component, as well as room dimensions (length, width, area, etc.), which are used as important input information in subsequent design and cost estimation processes.

[0023] In this way, the image analysis unit 102 accurately reads the spatial configuration of the dwelling unit from the pixel information of the drawing image and outputs it as structural information expressed quantitatively and structurally. As a result, it plays an important role as the starting point for the renovation plan proposal process in the renovation support system 1 according to this embodiment, supporting the quality and accuracy of subsequent processes.

[0024] [1.1.3. Configuration and Function of Plan Generator 103] The plan generation unit 103 is a core processing module that receives the structural information generated by the image analysis unit 102 as input and designs, evaluates, and selects renovation proposals based on the structural information and user-specified conditions.

[0025] The plan generation unit 103 first creates a prompt based on the input structural information and conditions specified by the user (e.g., desired floor plan type, family composition, furniture type, preferred flow line, etc.), and inputs the prompt to the generation AI 400. Based on this prompt, the generation AI 400 outputs multiple renovation floor plan proposals including furniture type, position, dimensions, arrangement pattern, etc. The plan generation unit 103 evaluates each output floor plan proposal in light of predetermined design constraints and flow line evaluation criteria. For example, the design constraints include preventing furniture from interfering with each other, ensuring a minimum amount of movement space, and adhering to functional zoning, while the flow line evaluation criteria include the length of the route between rooms, the number of intersections, and smooth circulation.

[0026] The plan generating unit 103 calculates an evaluation value for each floor plan based on the constraints and evaluation criteria described above. This evaluation value may be a quantitative score, or may be a weighted and integrated evaluation value of multiple evaluation indexes. The plan generation unit 103 then selects the most suitable floor plan based on the calculated evaluation values. This optimal plan is passed on to subsequent processes such as estimation, matching, and CG generation, and is used as the basis for proposing a final renovation plan to the user.

[0027] In this way, the plan generation unit 103 flexibly creates floor plan proposals by utilizing the generation AI 400 in response to user requests, and also has the function of objectively judging the merits or demerits of the proposals through a predetermined evaluation logic, thereby forming the core of the intelligent space proposals of the present invention.

[0028] [1.1.4. Configuration and Function of Integration Unit 104] The estimating unit 104 has a function of calculating the construction costs required to realize the renovation floor plan selected by the plan generating unit 103, and outputting the calculation results as a construction estimate.

[0029] First, the estimation unit 104 identifies the type of construction material according to the components included in the proposed floor plan and calculates the required quantity for each. For example, the specifications, area, and quantity of materials required for construction are automatically calculated for each element, such as wall area, floor area, ceiling finish, fixtures, storage equipment, and housing equipment (kitchen, bathroom, toilet, etc.) to be newly constructed or removed. Next, the estimating unit 104 references unit price information for each material from the material database 320 and acquires labor cost information for each type of work. Based on this information, the estimating unit 104 calculates the cost (= unit price × quantity + labor cost) for each material and work item, and adds these up to calculate the construction cost (construction estimate) for the entire renovation. Furthermore, the estimating unit 104 organizes the cost information calculated above by item and automatically generates a construction estimate including the quantity, unit price, subtotal, total amount, etc. of each component. This construction estimate is output to the user terminal 200, and is configured so that the user can immediately grasp the budget corresponding to the renovation content.

[0030] The estimation process in the estimation unit 104 can also be used to automatically estimate rough costs at the design stage, and is also effective as a decision-making support before preparing detailed drawings or requesting estimates. In this way, the estimation unit 104 can quickly and precisely calculate material and labor costs based on structural information and design proposals, and performs an important function of enhancing the practicality of the entire renovation support system 1 and the completeness of automated processing.

[0031] [1.1.5. Configuration and Function of Matching Unit 105] The matching unit 105 is a component that is responsible for extracting construction companies or real estate companies as recommended candidates based on the renovation floor plan selected by the plan generation unit 103 and the conditions specified by the user, and presenting them to the user.

[0032] First, the matching unit 105 receives as input the components included in the proposed floor plan (e.g., new wall construction, kitchen replacement, toilet relocation, etc.) and user-specified conditions (e.g., location, budget, desired construction period, contractor type, etc.), and based on this information, creates a prompt for the generation AI 400. The prompt is a semantic integration of contractor search conditions, and is generated in natural language or structured data format. Here, the generation AI 400 uses the prompt as input to expand or complement the internal conditions (search targets) for extracting suitable construction companies or real estate companies, and outputs more accurate recommended candidate conditions. This utilizes information recorded in the company database 330, namely, feature quantities such as each company's past construction performance, available areas, areas of expertise, customer evaluation scores, price range trends, and response speed.

[0033] The matching unit 105 uses this information to score the candidate companies and generate a list with a recommendation ranking based on the evaluation value. The final matching results are presented to the user via the user terminal 200, allowing the user to directly request a quote or make an inquiry if desired. In this way, the matching unit 105 constitutes an important feature of the present invention in that it can automatically and semantically optimize the selection of construction companies in conjunction with design information, something that previously required manual searching and inquiries.

[0034] [1.1.6. Configuration and Function of CG Generation Unit 106] The CG generation unit 106 is responsible for the process of visualizing in three dimensions the completed image of the renovated residential space based on the renovation floor plan selected by the plan generation unit 103, and generating image data (a rendering of the completed space, so-called CG perspective) to present to the user. Specifically, the CG generation unit 106 acquires spatial structure information about the components included in the selected renovation floor plan, namely, walls, floors, ceilings, doors, windows, storage, and equipment, as well as furniture layout information, such as the type, placement position, dimensions, and orientation of the furniture, and then uses this information to virtually construct the interior of the apartment in three-dimensional space.

[0035] The CG generation unit 106 executes the following processes in stages on the constructed three-dimensional model: (1) Material application process Material information such as texture, color, and pattern is assigned to each component (flooring, wallpaper, fixtures, furniture, etc.) to create an appearance that faithfully reproduces the actual finishing materials. (2) Lighting setting processing Lighting conditions are set based on natural light and the placement of lighting fixtures, and the brightness, shadows, and degree of reflection of the entire scene are adjusted, making it possible to create a sense of realism and atmosphere as an indoor space. (3) Rendering process Based on the above 3D model, viewpoint information, materials, and lighting settings, a CG perspective is output as 2D image data through computer graphics processing. It may be generated from multiple viewpoints depending on the user's viewpoint and shooting angle.

[0036] The generated CG perspective is provided to users to intuitively grasp the completed image after renovation, and is used for plan consideration, decision-making, customer proposals, etc. It can also be linked to other systems, such as by converting it into AR / VR format as needed. In this way, the CG generation unit 106 generates a realistic rendering of the completed design from drawing-based design information, thereby matching the design intent with the impression of the space, and dramatically improving the user's satisfaction and visibility.

[0037] [1.1.7. Configuration and Function of Storage Unit 107] The memory unit 107 is an information storage device for temporarily or permanently storing data handled by each of the above-mentioned processing units within the renovation support system 1, and is a fundamental component that supports the stability and reproducibility of the overall information processing of the present invention.

[0038] The storage unit 107 stores, for example, the following information: Input information such as floor plan images, user-specified conditions, desired specifications, desired construction area, etc. acquired from the user terminal 400 via the input / output unit 101. Structural information generated by the image analysis unit 102. A plurality of renovation floor plan proposals generated by the plan generating unit 103 and their evaluation values. The material quantity, unit price, labor cost, estimated amount, and construction estimate calculated by the estimation unit 104. Construction company information and matching results extracted by the matching unit 105. A three-dimensional model and a CG perspective drawing of the completed model generated by the CG generation unit 106.

[0039] The storage unit 107 can organize, classify and record this data by user or by case, and is also used for database reference for suspending and resuming processing, tracking history and reusing. The storage unit 107 is configured by a hard disk device, semiconductor memory, cloud storage, or a combination of these, and provides a flexible and scalable data processing environment for the entire renovation support system 1.

[0040] [1.1.8. Example of hardware configuration of renovation support device 100] 2 is a diagram showing an example of the hardware configuration of a renovation support apparatus 100 according to an embodiment of the present invention. The renovation support apparatus 100 includes a bus 1010, a processor 1020, a memory 1030, a storage device 1040, an input / output interface 1050, and a network interface 1060.

[0041] The bus 1010 is a data transmission path for transmitting and receiving data among the processor 1020, memory 1030, storage device 1040, input / output interface 1050, and network interface 1060. However, the method of connecting the processor 1020 and the like to each other is not limited to bus connection.

[0042] The processor 1020 is implemented by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.

[0043] The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.

[0044] The storage device 1040 is an auxiliary storage device realized by removable media such as a hard disk drive (HDD), a solid state drive (SSD), or a memory card, or a read-only memory (ROM), and has a recording medium. The recording medium of the storage device 1040 stores program modules that realize each function of the renovation support device 100. The processor 1020 loads each of these program modules into the memory 1030 and executes them, thereby realizing each function corresponding to the program module.

[0045] The input / output interface 1050 is an interface for connecting the renovation support apparatus 100 to various input / output devices (not shown).

[0046] The network interface 1060 is an interface for connecting the renovation support apparatus 100 to the network N. The method for connecting the network interface 1060 to the network N may be a wireless connection or a wired connection. The renovation support apparatus 100 may communicate with a user terminal 200 or the like via the network interface 1060.

[0047] [1.2. User terminal configuration and functions] The user terminal 200 is an external device connected to the renovation support device 100 via the network N, and is an information processing terminal operated by a user such as a real estate company representative or a sales or design staff member of a construction company.

[0048] The user terminal 200 provides the following functions to support the renovation work for one of the apartment units in the housing complex that the user is in charge of: Select and upload floor plan images - Input of user-specified conditions such as area information, desired specifications, budget, construction period, etc. Receiving and displaying various processing results provided by the renovation support device 100

[0049] Specifically, the user terminal 200 can be a smartphone, tablet, laptop, desktop, or web browser-compatible business terminal. These terminals are equipped with Internet communication functions, a GUI-based user interface, image display functions, and data transmission and reception functions.

[0050] Through the user terminal 200, the user selects and transmits a floor plan image corresponding to the target dwelling unit in the apartment complex, and inputs the area conditions, desired floor plan, interior design preferences, priorities, etc. This information is imported into the renovation support device 100 via the input / output unit 101 and is used in each stage of the image analysis process, plan generation process, estimate creation process, contractor matching process, and CG generation process.

[0051] In addition, the renovation proposals, construction estimates, list of recommended construction companies, CG perspective drawings of the completed project, etc. generated as a result of the processing are displayed on the user terminal 200, and the user can refer to them to carry out tasks such as examining proposals, presenting them to clients, sharing them internally, and applying for approval. In this way, the user terminal 200 constitutes the starting point and the ending point of input and output in this system, and functions as an interface terminal that mediates between user operations and information presentation.

[0052] [1.3. Configuration and Function of Database Group 300] The database group 300 is a group of storage devices for systematically storing and providing information necessary for various processes in the renovation support system 1, and is composed of multiple databases divided by purpose, such as a plan database 310, a materials database 320, and a company database 330. These databases 300 are connected to the renovation support device 100 via a network N, and are configured so that various processing units can refer to these data in real time as necessary.

[0053] The plan database 310 stores a plurality of renovation floor plan proposals generated by the plan generating unit 103, their evaluation values, selection results, user input conditions, etc., in chronological order or on a case-by-case basis. The materials database 320 stores information about the construction materials used by the estimation unit 104, such as material type, dimensions, unit price, finishing specifications, and construction classification. The company database 330 stores feature quantities such as company name, location, service area, specialty, price range, past construction performance, and customer evaluations for the matching unit 105 to select a construction company.

[0054] These databases may be stored in the memory unit 107 provided inside the renovation support device 100, or may be configured to be installed on cloud storage or an external server, and may be configured to be distributed, expandable, and synchronized as needed. In this way, the database group 300 functions as an information infrastructure that supports the accuracy and consistency of various design, estimation, and recommendation processes, and is an essential component for ensuring the reliability and scalability of the automated series of processes of the present invention.

[0055] [1.4. Configuration and Function of Generation AI 400] The generation AI 400 is an intelligent processing module in the renovation support system 1 that assists in understanding the floor plan structure, analyzing the spatial composition, and automatically generating new renovation plans based on a floor plan image of a unit in an apartment building.

[0056] The generation AI 400 receives as input in the plan generation unit 103 a prompt constructed based on the structural information output from the image analysis unit 102 and user-specified conditions, and outputs multiple renovation floor plan proposals including furniture types, placement patterns, dimensions, functional zoning, etc. The generation AI 400 also evaluates the consistency of each output plan in light of predetermined design constraints and flow line evaluation criteria, and assists the plan generation unit 103 in selecting the optimal plan.

[0057] The generation AI 400 according to this embodiment can use a large-scale language model (LLM) such as GPT-4, GPT-4 Turbo, or GPT-4o, which are associated with OpenAI's "GPT (registered trademark)," or Claude 3, Claude 3 Haiku, Claude 3 Sonnet, or Claude 3 Opus, which are associated with Anthropic's "CLAUDE (registered trademark)." Furthermore, it is also possible to apply a multi-modal large-scale language model (Multi-Modal LLM: MMLLM), which is capable of integrated processing of different modalities such as text, drawings, images, and voice, to the generation AI 400. This enables applications such as converting drawing-based input instructions and configuration information read from images into natural language or structured data, and directly generating floor plan proposals and layout drawings that maintain spatial and configuration consistency. In this way, the generation AI 400 is the technical core of the present invention in terms of both understanding the floor plan of the dwelling unit and generating new designs, and is the main means for realizing intelligent automation of spatial proposals.

[0058] In this way, in the renovation support system 1 of this embodiment, the renovation support device 100 works in cooperation with the user terminal 200, database group 300, and generation AI 400 connected via the network N to comprehensively and automatically realize the process of creating a floor plan based on user input and a renovation plan based on various specified conditions. This allows for automated processing of everything from structural analysis of floor plans to design, estimates, and recommendations of construction candidates, streamlining renovation work all at once.

[0059] More specifically, the renovation support system 1 of this embodiment uses a generation AI to generate, evaluate, and select multiple new floor plan proposals that take into consideration furniture placement and daily activity patterns based on floor plan images of an existing dwelling unit, and is capable of automatically and comprehensively executing a series of renovation support processes based on the proposed floor plans, including creating estimates, outputting renderings of the completed building (CG perspective drawings), and recommending construction companies. This allows users to perform everything from structural analysis of floor plans to design, estimates, and selection of construction candidates in one go with minimal operations, resulting in the entire renovation work workflow, which was previously separated, being integrated and processed quickly with high consistency.

[0060] [2. Example of operation in Renovation Support System 1] 3 is a flowchart showing the overall processing procedure of the renovation support device 100 according to one embodiment of the present invention. The input / output unit 101 of the renovation support device 100 acquires (receives / inputs) user-specified conditions such as a floor plan image and area information corresponding to a property to be renovated, which are transmitted from the user terminal 200 (step S100).

[0061] Here, regional information refers to information about the location of the dwelling unit designated by the user as the property to be renovated, including, for example, the postal code, city, ward, town, village name, nearest station, prefecture, and the building's floor and building number. More detailed conditions can also be included, including information about the desired construction area, local building regulations, and the surrounding environment (commercial facilities, public transportation, sunlight conditions, etc.). This regional information is treated as an important decision parameter because it influences the recommendation of construction companies, price range adjustments, and design constraints (height restrictions, securing evacuation routes, etc.) in the subsequent matching process.

[0062] When the input / output unit 101 acquires these data, the image analysis unit 102 performs image analysis processing to detect walls, rooms, doors, etc. from the floor plan image and generate floor plan structure data (step S200). Figure 4 is a flowchart for explaining the details of the analysis processing (step S200) by the image analysis unit 102 of the renovation support device 100 according to one embodiment of the present invention.

[0063] As shown in FIG. 4, the image analysis unit 102 performs preprocessing such as tilt correction, noise removal, and resolution conversion on the input floor plan image (step S201).

[0064] Next, the image analysis unit 102 extracts elements such as walls, rooms, doors, windows, storage areas, etc. from the pre-processed image using an image processing algorithm or the like (step S202).

[0065] Next, the image analysis unit 102 estimates room attributes such as the name and use of the room (for example, living room, bedroom, kitchen, etc.) from notes, layout, etc. (step S203).

[0066] Next, the image analysis unit 102 digitizes information such as the area of the extracted components, the length of the walls, the positions of the openings, and the orientation of the doors, and converts it into quantitative data of the spatial configuration (step S204).

[0067] Furthermore, the image analysis unit 102 generates floor plan structure data including room IDs, components, dimensional information, etc., based on the digitized structural attribute information (step S205).

[0068] Then, the image analysis unit 102 outputs the generated floor plan structure data to the plan generation unit 103, which is in charge of subsequent processing (step S206). This floor plan structure data becomes the basic information for the renovation plan created by the plan generation unit 103 in the subsequent steps.

[0069] Returning to the flowchart of Fig. 3, after the analysis process (step S200) by the image analysis unit 102, the plan generation unit 103 generates a renovation plan based on the floor plan structure data and user-specified conditions (step S300). Fig. 5 is a flowchart for explaining the details of the renovation plan generation process (step S300) by the plan generation unit 103 of the renovation support device 100 according to one embodiment of the present invention.

[0070] As shown in FIG. 5, the plan generation unit 103 acquires floor plan structure data from the image analysis unit 102 and user-specified conditions from the input / output unit 101 (step S301), and then configures this information as a prompt and inputs it to the generation AI 400.

[0071] The plan generation unit 103 automatically generates a plurality of renovation plans that take into consideration room configuration, traffic flow, functional zoning, furniture arrangement, etc., through processing in cooperation with the generation AI 400 (step S302).

[0072] In the next step S303, each generated plan is verified based on predetermined design constraints (furniture interference prevention, spatial balance, storage rate, etc.) and traffic flow evaluation criteria (movement efficiency, circulation of traffic flow, safety, etc.).

[0073] As a result, for plans that are determined to be problem-free in step S303, the plan generation unit 103 maps furniture placement information (furniture type, size, orientation, coordinates, etc.) to floor plan structure data, and ultimately constructs one or more renovation plan data (step S304). On the other hand, if a plan is determined to have a problem in step S303, it is discarded, or if necessary, the process returns to step S302 to generate a new renovation plan.

[0074] After one or more renovation plan data are constructed in step S304, the renovation plan with the highest evaluation value is selected from among them, and that plan is output to the user terminal 200 via the input / output unit 101 and stored (saved) in the plan database 310 (step S305). In this way, the plan generation unit 103 utilizes the generation AI 400 to create, evaluate, and select flexible and logical floor plan proposals, and can provide the user with a consistent renovation plan.

[0075] Returning to the flowchart of Fig. 3, in parallel with the renovation plan generation process (step S300) by the plan generation unit 103, or after the renovation plan has been generated, the estimating unit 104 generates a construction estimate for the generated renovation plan (step S400). Fig. 6 is a flowchart for explaining the details of the construction estimate generation process (step S400) by the estimating unit 104 of the renovation support device 100 according to one embodiment of the present invention.

[0076] The estimating unit 104 acquires information on the components of the renovation plan, that is, the wall area, floor area, number of doors and windows, and arrangement of household equipment, from the plan generating unit 103 (step S401).

[0077] Next, the estimating unit 104 refers to the material database 320 to acquire unit price information corresponding to the building materials (finishing materials, structural materials, equipment, etc.) required for each component (step S402). At this time, the labor unit price of the workers required for the renovation work is also acquired at the same time from any of the databases in the database group 300 (including databases not shown), or from user-specified conditions, etc.

[0078] Furthermore, the estimating unit 104 calculates the cost for each component based on the acquired quantity of building materials and their respective unit price information, adds labor costs as necessary, and then estimates the cost for each construction item (step S403).

[0079] Furthermore, the estimating unit 104 automatically generates a construction estimate in a format including the cost breakdown by material, unit price, quantity, subtotal, and total amount based on the estimation result in step S403 (step S404).

[0080] Then, the estimating unit 104 outputs the created construction estimate to the user terminal 200 via the input / output unit 101 (step S405). This allows the user to immediately refer to the estimate information. In this way, the estimating unit 104 can quickly and accurately calculate related material information and labor costs based on the contents of the renovation plan, and automatically generate a construction estimate. This allows the user to conduct more specific considerations while grasping a realistic budget that corresponds to the design intent of the renovation proposal.

[0081] Returning to the flowchart of Figure 3, in parallel with the renovation plan generation process by the plan generation unit 103 (step S300) and the generation of a construction estimate by the estimating unit 104 (step S400), or after the renovation plan and construction estimate have been generated, the matching unit 105 executes matching with a construction company that will carry out the renovation according to the generated renovation plan (step S500). Figure 7 is a flowchart for explaining the details of the matching process (step S500) by the matching unit 105 of the renovation support device 100 according to one embodiment of the present invention.

[0082] As shown in FIG. 7, the matching unit 105 first acquires a proposed renovation plan (a data structure including room configuration, construction items, estimated budget, etc.) acquired from the plan generating unit 103 or the storage unit 107 (step S501).

[0083] Next, the matching unit 105 acquires the area information and user-specified conditions (for example, desired price range, construction period, service area, etc.) acquired from the user terminal 200 or the storage unit 107 (step S502).

[0084] Furthermore, based on the acquired information, the matching unit 105 performs formatting processes such as tagging the plan components (e.g., replacing plumbing, changing partitions, etc.), clarifying the specified conditions, and converting the generated AI 400 into a prompt format (step S503).

[0085] Furthermore, the matching unit 105, in cooperation with the generation AI 400, extracts conditions of construction companies that are suitable in terms of regional response, construction track record, areas of expertise, customer evaluations, price range, etc., and generates candidate matching conditions (step S504).

[0086] Furthermore, the matching unit 105 queries the company database 330 to acquire company information that matches the output of the generation AI 400 (step S505). Here, detailed information such as the company name, location, service area, specialty construction type, price trend, etc. are also acquired.

[0087] Furthermore, the matching unit 105 performs a scoring process on each candidate company based on a plurality of evaluation indices (trust score, past evaluation, regional suitability, etc.) (step S506).

[0088] Then, the matching unit 105 formats and outputs the scoring-processed matching results (each company candidate) as a list sorted according to recommendation order (step S507). In this way, the matching unit 105 can dynamically and automatically present construction company candidates that are most suitable for the user's wishes and design content through advanced semantic processing using the generation AI 400.

[0089] Returning to the flowchart of Fig. 3, after the plan generation unit 103 generates a renovation plan (step S300), or after the estimating unit 104 generates a construction estimate (step S400) and the matching unit 105 performs matching (step S500), the CG generation unit 106 generates a three-dimensional model based on the renovation plan, etc., and performs rendering to generate a CG perspective (a conceptual drawing of the completed building) (step S600). Fig. 8 is a flowchart for explaining the details of the CG perspective generation process (step S600) by the CG generation unit 106 of the renovation support device 100 according to one embodiment of the present invention.

[0090] The CG generation unit 106 acquires the renovation plan proposal and furniture arrangement information (types, arrangement positions, dimensions, orientations, etc. of furniture) from the plan generation unit 103 or the storage unit 107 (step S601).

[0091] Next, the CG generating unit 106 converts the acquired information in the two-dimensional coordinate system into a three-dimensional space format, and prepares it for placement in the virtual space (step S602).

[0092] Furthermore, the CG generation unit 106 specifically arranges furniture models in the three-dimensional space in addition to architectural structural elements such as walls, floors, ceilings, and fittings, to form a structure that is close to the real space (step S603).

[0093] Furthermore, the CG generation unit 106 performs a material application process for each of the formed components (step S604). This material application process includes applying material information such as the texture, color, pattern, and gloss of flooring, wallpaper, and furniture surface finishes. This allows the CG object to have an appearance that matches the real material.

[0094] Furthermore, the CG generation unit 106 executes lighting setting processing such as the arrangement of lighting (natural light, ceiling lighting, etc.), color temperature of the light source, brightness, and shadow expression, to create a sense of realism throughout the space (step S605).

[0095] Furthermore, the CG generation unit 106 sets a camera angle and perspective (bird's-eye view, eye level, dynamic viewpoint, etc.) that matches the user's viewpoint, and adjusts it so that the space is captured from a viewpoint that looks good (step S606).

[0096] Finally, the CG generation unit 106 performs rendering processing on the entire virtual space reflecting the above configuration and settings, and generates a CG perspective (a conceptual drawing of the completed work) as a two-dimensional image (step S607). This image is output from multiple viewpoints as needed, and is stored in the storage unit 107, and is also transmitted via the input / output unit 101 so that it can be displayed or saved on the user terminal 200. In this way, the CG generation unit 106 automatically generates highly accurate 3D visuals from blueprint-based plan information, enabling the user to intuitively and visually visualize the completed space. This ensures consistency between the design intent and the impression of the space, and realizes renovation support with a high level of proposal power and persuasiveness.

[0097] As described above, the renovation support system 1 of this embodiment can automatically execute each process from structural analysis of floor plans, plan generation, estimate creation, construction candidate recommendation, and CG generation, starting from input from the user, in accordance with the various processing flows shown in Figures 3 to 8, with each module coordinating with the others.

[0098] 3 to 8 are merely examples of the present invention. Within the scope of the present invention, processes other than those described in Figures 3 to 8 may be included, some of the processes described in Figures 3 to 8 may be omitted, or the order of the processes may be changed.

[0099] Although the embodiments of the present invention have been described above, the embodiments disclosed herein are illustrative in all respects and should not be considered limiting. The scope of the present invention is defined by the claims rather than the above description, and it is intended to include meanings equivalent to the claims and all modifications within the scope of the claims. [Explanation of symbols]

[0100] 1. Renovation support system 100 Renovation Support Device 101 Input / output section 102 Image analysis unit 103 Plan Generation Unit 104 Integration Unit 105 Matching Department 106 CG generation department 107 Storage section 200 user terminals 300 databases (DB) 310 Plan Database 320 Materials Database 330 company database 400 Generation AI 1010 Bus 1020 processor 1030 memory 1040 Storage Device 1050 Input / Output Interface 1060 Network Interface N Internet

Claims

1. A renovation support system that proposes renovation plans based on floor plan images of existing dwelling units, an image analysis means for recognizing a plurality of components constituting the interior of the dwelling unit to be renovated from the floor plan image and generating structural information including room dimensions and wall thickness; A plan generation means for generating a plurality of renovation floor plan proposals including furniture placement information using a generation AI based on the structural information, calculating an evaluation value for each floor plan proposal in light of predetermined design constraint conditions and traffic flow evaluation criteria, and selecting the most suitable renovation floor plan proposal based on the evaluation value; an estimation means for calculating the type and quantity of necessary building materials based on the components included in the selected renovation floor plan, calculating a construction estimate based on unit price information for each material and labor cost information for workers, and generating a construction estimate including the construction estimate; A renovation support system comprising:

2. The renovation support system according to claim 1, The renovation support system further comprises a matching means for using a generation AI to input the characteristics of the selected renovation floor plan and the specified conditions specified by the user as prompts into the generation AI, searching for a group of candidate construction companies or real estate companies based on the prompts, and generating matching results with a recommendation ranking.

3. In the renovation support system according to claim 1 or 2, The renovation support system further comprises a CG generation means for generating an interior model of the dwelling unit and a furniture model in three-dimensional space based on the selected renovation floor plan, applying materials and setting lighting, then performing rendering processing, and outputting a CG perspective as a rendering of the completed project.

4. In the renovation support system according to claim 1 or 2, The renovation support system is characterized in that the plan generation means inputs prompts generated based on the structural information and user-specified conditions into the generation AI, which generates multiple floor plan proposals including furniture arrangement patterns, and the generation AI calculates the evaluation value for each proposal based on predetermined evaluation criteria and selects the optimal floor plan proposal based on the evaluation value.

5. The renovation support system according to claim 2, The matching means constructs a prompt for company search to be input into the generation AI based on the components included in the proposed renovation floor plan and the specified conditions, extracts candidate companies based on features including at least one of past construction performance, areas of expertise, price range, customer reviews, or local responsiveness, and generates a list of candidate companies in a recommendation order based on the matching score as the matching result, a renovation support system characterized by this.

6. The renovation support system according to claim 1, The renovation support system is characterized in that the image analysis means extracts secondary information including at least one of annotation information, scale description, and room name contained in the floor plan image, and has a pre-processing completion function that uses the secondary information to scale the structural information, complement room attributes, and align dimensional values.

7. A computer-based renovation support method that proposes a renovation plan using an input floor plan image of an existing dwelling unit, comprising: an image analysis step of recognizing a plurality of components constituting the interior of the dwelling unit to be renovated from the floor plan image and generating structural information including room dimensions and wall thickness; a plan generation step of generating a plurality of renovation floor plan proposals including furniture placement information using a generation AI based on the structural information, calculating an evaluation value for each floor plan proposal in light of predetermined design constraints and traffic flow evaluation criteria, and selecting the most suitable renovation floor plan proposal based on the evaluation value; an estimation step of calculating the type and quantity of necessary building materials based on the components included in the selected renovation floor plan, calculating a construction estimate based on unit price information for each material and labor cost information for workers, and generating a construction estimate including the construction estimate; A renovation support method comprising:

8. A computer-readable program to be executed by a renovation support system that proposes renovation plans based on floor plan images of existing dwelling units, an image analysis step of recognizing a plurality of components constituting the interior of the dwelling unit to be renovated from the floor plan image and generating structural information including room dimensions and wall thickness; a plan generation step of generating a plurality of renovation floor plan proposals including furniture placement information using a generation AI based on the structural information, calculating an evaluation value for each floor plan proposal in light of predetermined design constraints and traffic flow evaluation criteria, and selecting the most suitable renovation floor plan proposal based on the evaluation value; an estimation step of calculating the type and quantity of necessary building materials based on the components included in the selected renovation floor plan, calculating a construction estimate based on unit price information for each material and labor cost information for workers, and generating a construction estimate including the construction estimate; A computer-readable program for executing the program.

Images