2d CAD-based automatic integrated data generation device using artificial intelligence and control method thereof

Abstract

The present disclosure relates to a 2D CAD-based automatic integrated data generation device using artificial intelligence and a control method thereof. By generating integrated data at a lower cost than existing 3D model-based auto integration methods, the integration technology can be applied to small and medium-sized buildings or projects with limited budgets. Since it is not necessary to use the 3D model-based integration method, of which the high cost has limited its use, the integration technology can be applied to a greater variety of buildings and projects.

Description

2D CAD-based automatic integration data generation device using artificial intelligence and its control method

[0001] The present disclosure relates to a 2D CAD-based automatic cost data generation device and a control method thereof utilizing artificial intelligence that analyzes various drawings with an artificial intelligence model to recognize the space of the drawings and generate cost data for each construction item.

[0002] Unless otherwise indicated herein, the materials described in this section are not prior art to the claims of this application, and their inclusion in this section is not intended to be admitted as prior art.

[0003] In construction projects, cost estimation is the process of calculating the cost of a construction project. Cost estimation systematically calculates all costs—materials, labor, equipment, time, and more—necessary from the project planning stage to completion. This allows for accurate cost estimates, budgeting, and cost management. Cost estimation encompasses a variety of items, including material, labor, and equipment costs. Cost estimation is essential for the successful execution of a construction project, allowing for systematic cost management and preventing budget overruns.

[0004] The process of accurately calculating the quantity of raw materials required for construction has been performed manually for decades. Figure 1 illustrates the traditional estimation process. Many construction companies perform estimation in-house, but this process typically requires visual inspection of dozens of drawings and inputting thousands of lines into Excel calculation reports. This process typically takes more than four weeks for a single project. Meanwhile, estimation offices, which perform estimation on behalf of construction companies, are expensive and time-consuming. Therefore, many specialized construction companies forgo estimation and instead proceed with construction based on the calculation reports provided by general construction companies. However, proceeding with construction based on the calculation reports provided by general construction companies often results in losses.

[0005] Furthermore, to address the significant time, manpower, and cost involved in estimation, a technology has been developed that converts drawings to 3D and performs automatic estimation. However, conventional automatic estimation methods, when performed using only a 3D model of the drawing, require converting all 2D drawings to 3D, resulting in significant costs and limited applicability to low-rise buildings of seven or eight stories.

[0006] With the recent rapid rise in raw material costs, it's difficult to generate construction profits if construction progresses without knowing the exact cost. Therefore, the importance of cost estimation has become increasingly important. The construction industry requires rapid cost estimation within two weeks of a building owner's announcement. However, conventional technology struggles to generate accurate cost estimation data quickly. Furthermore, existing manual cost estimation methods allow for a certain percentage of the quantity (e.g., ±3%) of error to account for human error.

[0007] The purpose of the embodiment disclosed in the present disclosure is to enable generation of accumulated data at low cost by automatically generating accumulated data using an artificial intelligence model.

[0008] In addition, the embodiment disclosed in the present disclosure aims to provide a 2D CAD-based automatic accumulation data generation device utilizing artificial intelligence that automatically generates accumulation data using a 2D drawing, and a control method thereof.

[0009] However, the problems to be solved according to one embodiment are not limited to those mentioned above.

[0010] In order to achieve the above-described technical task, a 2D CAD-based automatic integration data generation device utilizing artificial intelligence according to the present disclosure comprises: a memory storing at least one command for generating 2D CAD-based automatic integration data utilizing artificial intelligence; And a processor for performing an operation according to the command, wherein the processor preprocesses a drawing collected through an artificial intelligence model, extracts a quantity of materials required for construction from the preprocessed drawing, calculates a quantity of materials required according to an analysis result of the extracted quantity, generates an estimated data according to the calculated quantity of materials required, and generates a calculation statement applying the generated estimated data, and when calculating the quantity of materials required, the processor identifies a required area of ​​materials including materials in the drawing through the artificial intelligence model, measures the length of the required area of ​​materials, and calculates an area of ​​the required area of ​​materials based on the measured length, and calculates the quantity of materials required according to the unit size and loading method of the materials, and when generating the estimated data, the processor scans a drawing input from a user terminal and converts it into a CAD image, recognizes a quantity of materials including construction items, materials, and ingredients required for the construction in the preprocessed drawing through an artificial intelligence-based detection model, derives a feature of the recognized quantity, and generates raw data at drawing coordinates corresponding to the coordinates of the feature based on the derived feature. Loading, and generating the accumulated data for the quantity required with the loaded raw data, and when generating the calculation statement, the processor recognizes a residential space including a bedroom, a living room, and a bathroom in the CAD image through the artificial intelligence-based space recognition model, and collects the accumulated data that calculates the quantity based on the loaded raw data, and creates the calculation statement based on the name of the recognized residential space and the accumulated data,In order to generate basis data for the accumulated elements included in the calculation statement through the artificial intelligence-based line drawing model, a part corresponding to the accumulated elements included in the calculation data is indicated as a line in a drawing image, and exception processing details are extracted based on special conditions, additional requirements and exceptions of the construction project specified in the special specifications in the drawing through the detection model, and the calculation statement is corrected according to the extracted exception processing details, and when the processor corrects the calculation statement, if a special material other than the originally planned material must be used among the exceptions, the cost and quantity of the special material can be added, and the calculation statement can be corrected based on the cost and quantity of the added special material.

[0011] At this time, the height of the raw data is calculated through a cross-sectional view, the height of the beam is calculated through a structural view, and when the processor detects an area where the beam is located through the structural view, the height of the raw data can be corrected by deducting the height by the height of the beam.

[0012] In addition, the processor may analyze the window drawing in the drawing to calculate the wall area excluding the area occupied by openings including windows and doors on the exterior wall of the building, calculate a deduction value due to the opening from the calculated wall area, and reflect the calculated deduction value in the generation of the accumulated data.

[0013] In addition, the processor may map a calculation basis to each of the accumulated data uploaded to the calculation statement, and the calculation basis may include an image of a drawing, and raw data including the length and height of the quantity may be output in the image. At this time, the processor may provide a user interface for verifying the calculation basis, and the user interface may output a drawing for verifying the calculation basis, and when a value requiring verification is selected, a CAD image matching the value may be provided. In addition, the processor may, through the user interface, change the color of or emphasize an area corresponding to the calculation basis including a wall and a window, output it by overlapping it with the CAD image, and display attribute values ​​of the drawing including the width, height, and area of ​​the wall and the window included in the area corresponding to the calculation basis.

[0014] Additionally, the processor can select a generation for which work has been completed based on the accumulated data and generate a calculation statement for ready-made billing.

[0015] In addition, a method for controlling an automatic 2D CAD-based data generation device utilizing artificial intelligence, performed by a processor according to the present disclosure for achieving the above-described technical task, comprises the steps of: preprocessing a drawing collected through an artificial intelligence model; extracting a quantity of materials required for construction from the preprocessed drawing; calculating a quantity of materials required according to an analysis result of the extracted quantity; generating estimation data according to the calculated quantity of materials required; and generating a calculation statement applying the generated estimation data, wherein the quantity of materials required calculation step comprises: identifying a required area of ​​materials including materials in the drawing through the artificial intelligence model; measuring a length of the required area of ​​the quantity of materials; calculating an area of ​​the required area of ​​the quantity of materials based on the measured length; and calculating the quantity of materials required according to a unit size and loading method of the quantity of materials, wherein the estimation data generation step comprises: scanning a drawing input from a user terminal and converting it into a CAD image; recognizing a quantity of materials including construction items, materials, and ingredients required for the construction in the preprocessed drawing through an artificial intelligence-based detection model; A step of deriving a feature of the recognized quantity; a step of loading raw data in drawing coordinates corresponding to the coordinates of the feature based on the derived feature; and a step of generating the accumulated data for the quantity required using the loaded raw data, wherein the step of generating the calculation statement includes a step of recognizing a residential space including a bedroom, a living room, and a bathroom in the CAD image through the artificial intelligence-based space recognition model; a step of collecting accumulated data that calculates the quantity based on the loaded raw data; a step of generating the calculation statement based on the name of the recognized residential space and the accumulated data;In order to generate basis data for the accumulated elements included in the calculation statement, the step of indicating a portion corresponding to the accumulated elements included in the calculation data as a line in a drawing image through the artificial intelligence-based line drawing model; the step of extracting exception processing details based on special conditions, additional requirements, and exceptions of a building project specified in a special specification in the drawing through the detection model; and the step of correcting the calculation statement according to the extracted exception processing details, wherein the step of correcting the calculation statement may include the step of adding the cost and quantity of the special material when a special material other than the originally planned material must be used among the exceptions; and the step of correcting the calculation statement based on the cost and quantity of the added special material.

[0016] At this time, the height of the raw data is calculated through a cross-sectional view, the height of the beam is calculated through a structural view, and the control method of the present integrated data generation device may further include a step of correcting the height of the raw data by deducting the height by the height of the beam when an area with the beam is detected through the structural view.

[0017] In addition, the step of generating the accumulated data may further include a step of analyzing a window drawing in the drawing to calculate a wall area excluding the area occupied by openings including windows and doors on the exterior wall of the building; a step of calculating a deduction value due to the opening from the calculated wall area; and a step of reflecting the calculated deduction value in the generation of the accumulated data.

[0018] In addition, the control method of the present accumulated data generation device further includes a step of mapping a calculation basis to each of the accumulated data uploaded to the calculation statement, wherein the calculation basis includes an image of a drawing, and raw data including the length and height of the quantity can be output to the image.

[0019] In addition, the control method of the present accumulated data generation device further includes a step of providing a user interface for verifying the calculation basis, wherein the user interface outputs a drawing for verifying the calculation basis, and when a value requiring verification is selected, a CAD image matching the value can be provided.

[0020] In addition, the control method of the present integrated data generation device may further include a step of changing the color of an area corresponding to a calculation basis including a wall and a window through the user interface or emphasizing the color thereof and outputting it by overlapping it with the CAD image; and a step of displaying attribute values ​​of the drawing including the width, height, and area of ​​the wall and the window included in the area corresponding to the calculation basis.

[0021] In addition, the step of generating the above-mentioned calculation statement may further include a step of selecting a generation for which work has been completed based on the above-mentioned accumulated data and generating a calculation statement for a ready-made claim.

[0022] According to the aforementioned problem solving means of the present disclosure, a 2D CAD-based automatic estimation data generation device and a control method thereof utilizing artificial intelligence generate estimation data at a lower cost than the existing 3D model-based automatic estimation method, thereby providing the effect of enabling the application of estimation technology to small and medium-sized buildings or projects with limited budgets.

[0023] In addition, the aforementioned problem-solving means of the present disclosure eliminates the need to use a 3D model-based estimation method that was limited due to high cost, thereby providing the effect of enabling the application of estimation technology to more types of buildings and projects.

[0024] In addition, the aforementioned problem solving means of the present disclosure provides the effect of reducing human error and increasing the speed and accuracy of estimation work by automatically generating estimation data using artificial intelligence, thereby significantly reducing construction preparation time and costs.

[0025] In addition, since the aforementioned problem solving means of the present disclosure generates the accumulated data based on the 2D drawing, it provides a user-friendly effect by allowing the accumulated data to be used as is and thus allowing the accumulated data to be performed without additional modeling work.

[0026] Furthermore, the aforementioned problem-solving method of this disclosure allows the application of estimation data to projects of various shapes and sizes through AI-based automation technology, thereby increasing the versatility of the technology. This can aid in efficient cost management and budgeting across the construction industry.

[0027] In addition, the aforementioned problem-solving means of the present disclosure significantly improves cost management and efficiency of construction projects, and enables the use of estimation technology in more projects, thereby providing the effect of increasing productivity in the overall construction industry.

[0028] The effects of the present invention are not limited to the effects described above, and should be understood to include all effects that can be inferred from the detailed description of the present invention or the composition of the invention described in the claims.

[0029] Figure 1 is a drawing showing a conventional accumulation operation process.

[0030] Figure 2 is a drawing showing a 2D CAD-based automatic accumulation data generation system utilizing artificial intelligence according to an embodiment.

[0031] Fig. 3 is a block diagram showing a cumulative generation device according to an embodiment.

[0032] Figure 4 is a drawing for explaining the process of generating accumulated data according to an embodiment.

[0033] Figure 5 is a drawing for explaining the types and functions of artificial intelligence models used in the embodiment.

[0034] Figure 6 is a drawing for explaining a quantity calculation process according to an embodiment.

[0035] Figure 7 is a drawing for explaining the process of reflecting the deduction value according to an embodiment.

[0036] Figures 8 to 12 are drawings showing the user interface of a 2D CAD-based automatic accumulation data generation system using artificial intelligence according to an embodiment.

[0037] Figure 13 is a diagram showing the process of generating accumulated data according to an embodiment.

[0038] Hereinafter, the embodiments disclosed in this specification will be described in detail with reference to the attached drawings. Regardless of the drawing numbers, identical or similar components will be given the same reference numbers, and redundant descriptions thereof will be omitted. The suffixes "module" and "part" used for components in the following description are assigned or used interchangeably only for the convenience of writing the specification, and do not in themselves have distinct meanings or roles. In addition, when describing the embodiments disclosed in this specification, if it is determined that a specific description of a related known technology may obscure the gist of the embodiments disclosed in this specification, a detailed description thereof will be omitted. In addition, the attached drawings are only intended to make it easier to understand the embodiments disclosed in this specification, and the technical ideas disclosed in this specification are not limited by the attached drawings, and it should be understood that they include all modifications, equivalents, and substitutes included in the spirit and technical scope of the present invention.

[0039] Terms that include ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by these terms. These terms are used solely to distinguish one component from another.

[0040] When a component is referred to as being "connected" or "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but that there may be other components intervening. Conversely, when a component is referred to as being "directly connected" or "connected" to another component, it should be understood that there are no other components intervening.

[0041] In this application, terms such as “include” or “have” are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but should be understood not to exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

[0042] In this specification, the term "unit" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Furthermore, a single unit may be realized using two or more pieces of hardware, and two or more units may be realized using a single piece of hardware.

[0043] Some of the operations or functions described herein as being performed by a terminal, apparatus, or device may instead be performed by a server connected to the terminal, apparatus, or device. Similarly, some of the operations or functions described herein as being performed by a server may also be performed by a terminal, apparatus, or device connected to the server.

[0044] Hereinafter, the present invention will be described in detail with reference to the attached drawings.

[0045] Figure 2 is a drawing showing a 2D CAD-based automatic accumulation data generation system utilizing artificial intelligence according to an embodiment.

[0046] Referring to FIG. 2, a 2D CAD-based automatic cost data generation system utilizing artificial intelligence according to an embodiment may be configured to include a cost data generation device (100) and a user terminal (200). The cost data generation device (100) analyzes drawings including floor plans, window drawings, and cross-section drawings collected from the user terminal (200) using various artificial intelligence models to determine the quantity of materials required for construction and generates cost data based on the determined quantity. In the embodiment, the drawings include, but are not limited to, floor plans, cross-section drawings, elevation drawings, window details, beam details, finishing material details, and special specifications.

[0047] In addition, in the embodiment, the accumulated data generation device (100) generates accumulated data according to the drawing analysis result, and maps the generated accumulated data and the drawing from which the accumulated data was generated as supporting data to generate a calculation statement. In the embodiment, the calculation statement can provide detailed information including the general construction company, region, site name, design document upload date, calculation completion date, work type, calculation cost, and progress status. In addition, the calculation statement outputs the drawing from which the corresponding calculation item was calculated and the items in the drawing by marking the information item of the accumulated cost, thereby allowing the user to check the supporting data for the calculation result together.

[0048] In the embodiment, the user terminal (200) is the terminal of the client requesting the creation of the integrated data and calculation statement. Once the integrated data creation is complete, the user terminal (200) can check it together with the calculation statement. According to the embodiment, the 2D CAD-based automatic integrated data creation system utilizing artificial intelligence automatically creates integrated data and automatically creates an integrated calculation statement based on this data, thereby enabling the rapid and accurate creation of integrated data required for construction at a low cost.

[0049] Fig. 3 is a block diagram showing a cumulative generation device according to an embodiment.

[0050] In an embodiment, the cumulative generation device (100) may be configured as a server. A server is a computing system that provides services to other computers or devices in a computer network or stores and manages data. The server accepts requests from other computers or devices, called clients, and provides responses or data in response to those requests. The configuration of the cumulative generation device (100) illustrated in FIG. 3 is merely a simplified example.

[0051] The communication module (110) can be configured regardless of the communication mode, such as wired or wireless, and can be configured with various communication networks, such as a personal area network (PAN) and a wide area network (WAN). In addition, the communication module (110) can operate based on the well-known World Wide Web (WWW), and can also utilize a wireless transmission technology used for short-distance communication, such as infrared (IrDA: Infrared Data Association) or Bluetooth. For example, the communication module (110) can be responsible for transmitting and receiving data required to perform a technique according to an embodiment of the present disclosure.

[0052] The memory (120) may refer to any type of storage medium. For example, the memory (120) may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), a RAM (Random Access Memory), a SRAM (Static Random Access Memory), a ROM (Read-Only Memory), an EEPROM (Electrically Erasable Programmable Read-Only Memory), a PROM (Programmable Read-Only Memory), a magnetic memory, a magnetic disk, and an optical disk. Such a memory (120) may also constitute a database as illustrated in FIG. 1.

[0053] The memory (120) can store at least one instruction that can be executed by the processor (130). In addition, the memory (120) can store any type of information generated or determined by the processor (130) and any type of information received by the server (200). For example, the memory (120) stores RM data and RM protocols according to the user, as will be described later. In addition, the memory (120) stores various types of modules, instruction sets, and models.

[0054] The processor (130) may perform technical features according to embodiments of the present disclosure, which will be described later, by executing at least one instruction stored in the memory (120). In one embodiment, the processor (130) may be configured with at least one core and may include a processor for data analysis and / or processing, such as a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), or a tensor processing unit (TPU) of a computer device.

[0055] This processor (130) can train a neural network or model designed using machine learning or deep learning methods. To this end, the processor (130) can perform calculations for neural network training, such as processing input data for training, extracting features from the input data, calculating errors, and updating the weights of the neural network using backpropagation. In addition, the processor (130) can also perform inference for a predetermined purpose using a model implemented using an artificial neural network method.

[0056] Figure 4 is a drawing for explaining the process of generating accumulated data according to an embodiment.

[0057] Referring to FIG. 4, a processor (130) according to an embodiment collects drawings and preprocesses the drawings using an artificial intelligence model. In an embodiment, the processor (130) may use an artificial intelligence model that performs space recognition, window recognition, wall detection, and beam recognition during the preprocessing of the drawings. In addition, the processor (130) stores each object, such as a space, window, wall, and beam, recognized through artificial intelligence in a database. Thereafter, the processor (130) extracts quantities, which are elements of quantity calculation, from the preprocessed drawings, and generates quantity data of the quantities included in the drawings based on the results of the analysis of the extracted quantities. Thereafter, a calculation statement applying the generated quantity data is generated or prepared. In an embodiment, quantity includes, but is not limited to, construction items, materials, and supplies required for construction, such as bricks, sashes, and windows.

[0058] To this end, the processor (130) collects drawings such as floor plans, window drawings, and cross-sections from the user terminal and classifies the collected drawings according to type. In the embodiment, the processor (130) performs preprocessing to convert CAD into images in the case of floor plans and cross-sections. For example, in the case of floor plans and cross-sections, the processor (130) converts the collected CAD (Computer-Aided Design) drawings into image formats. This may include converting CAD files into general image file formats such as JPG, PNG, and BMP. In addition, in the case of window drawings, the detailed elements of the window drawings are analyzed by a window drawing analyzer and necessary information is extracted. In addition, the processor (130) performs processes such as noise removal, outlier removal, and missing value processing so that the artificial intelligence model can accurately analyze the input image.

[0059] In an embodiment, the processor (130) analyzes images of floor plans and window plans using an artificial intelligence model and recognizes objects included in the drawings, such as spaces, walls, and windows, based on the analysis results. Thereafter, the area of ​​the masonry wall is calculated to derive the quantity of materials required for construction, and the derived quantity can be used to generate accumulated data and calculation details. In an embodiment, the processor (130) can perform preprocessing of the drawings, height recognition, and area calculation using a rule-based algorithm. In addition, the artificial intelligence model used in the embodiment includes a vision artificial intelligence engine.

[0060] Referring to FIG. 4, in the embodiment, the processor (130) scans a drawing input from a user terminal, converts it into a CAD image, and then detects the quantity using a detection model, which is an artificial intelligence model. In the embodiment, the quantity is an object recognized as a material, substance, or construction item required for construction in the drawing. For example, objects recognized as the quantity required for construction include, but are not limited to, walls, windows, bricks, windows, doors, sashes, etc. Thereafter, the processor (130) derives a feature from the quantity. In the embodiment, the feature is a main characteristic of each object, and may include the size, location, shape, etc. of the quantity. Thereafter, the processor (130) loads raw data from the coordinates of the drawing corresponding to the coordinates of the feature based on the derived feature. Thereafter, the accumulated data is generated based on the loaded raw data. In the embodiment, the raw data is data indicating the size and number of logistics used for calculating the quantity. For example, raw data includes, but is not limited to, the width and height of materials required for construction, such as walls and windows.

[0061] In an embodiment, the processor (130) maps the coordinates of the derived features to the drawing coordinates to generate the accumulated data based on the raw data. This is a process of determining where each feature is exactly located in the drawing. Thereafter, the processor (130) loads the raw data at the corresponding coordinates of the drawing based on the mapped feature coordinates. In an embodiment, the raw data includes detailed information of the drawing. For example, the detailed information of the drawing includes the type, quantity, dimensions, width, height, etc. of materials. Thereafter, the processor (130) analyzes the loaded raw data to finally generate the accumulated data. In an embodiment, the accumulated data includes, but is not limited to, the type, quantity, size, and material list of the exact quantity required for construction.

[0062] In addition, the processor (130) recognizes a living space including a bedroom, a living room, and a bathroom in a CAD image through an artificial intelligence-based space recognition model, collects accumulated data that calculates the amount of raw materials based on raw data, and creates an accounting statement based on the name of the recognized space and accumulated data.

[0063] To this end, the processor (130) scans the drawing input from the user terminal, converts it into a digital image file, and then converts it into a CAD image. Then, the processor (130) recognizes residential spaces including a bedroom, a living room, and a bathroom from the CAD image using an artificial intelligence-based space recognition model. In an embodiment, the artificial intelligence model that recognizes and divides the space analyzes the characteristics and layout of each space and assigns a name to the corresponding space. Thereafter, the processor (130) loads raw data of each recognized residential space from the drawing based on the coordinates of the corresponding space. At this time, the loaded raw data may include information on area, structure, and material. Thereafter, the processor (130) analyzes the loaded raw data to calculate the amount of raw materials required for each space. In this process, the processor (130) may perform calculations by considering the size and shape of each space, the type of material to be used, and the like. Thereafter, the processor (130) collects the calculated raw material quantity data and generates an estimate data. In an embodiment, the estimate data generated includes the type and quantity of materials required for each space. Thereafter, the processor (130) creates a calculation report based on the names of the recognized spaces (e.g., bedroom, living room, bathroom, etc.) and the calculated accumulated values. The calculation report may include the space name, the name of each residential space (e.g., bedroom 1, living room, bathroom 2, etc.), the quantity of raw materials, and the type and quantity of raw materials required for each space.

[0064] Figure 5 is a drawing for explaining the types and functions of artificial intelligence models used in the embodiment.

[0065] Referring to FIG. 5, in an embodiment, the processor (130) may use a detection model that performs object detection, a space recognition model that performs area segmentation, and an artificial intelligence model that performs line drawing. In an embodiment, the detection model is an artificial neural network model that recognizes objects in drawings, including quantities of windows and walls, text, and numbers, and analyzes them. The space segmentation model is an artificial neural network model that recognizes a region of space in a drawing and segments and names the recognized space. In addition, in an embodiment, the line drawing model is an artificial neural network model that displays a part corresponding to an accumulation element included in an accumulation data as a line on a drawing image in order to generate an accumulation element included in the calculation details and basis data therefor.

[0066] In an embodiment, the detection models may include YOLO (You Only Look Once), Faster R-CNN (Region-based Convolutional Neural Networks), SSD (Single Shot MultiBox Detector), EAST (Efficient and Accurate Scene Text Detector) for text detection, CRAFT (Character Region Awareness for Text detection), Mask R-CNN, RetinaNet, etc.

[0067] Additionally, models for recognizing and segmenting space may include Fully Convolutional Networks (FCN), U-Net, Seg Net, Mask R-CNN, Deep Lab, Pyramid Scene Parsing Network (PSPNet), etc. Additionally, line drawing models may include Holistically-Nested Edge Detection (HED), LINE-Net, Deep Hough Transform, etc. The algorithms described above in the embodiments are merely examples and each model is not limited to the algorithms described above.

[0068] In the embodiment, the processor (130) analyzes the drawing using a detection model to identify an area requiring a quantity of material, measures the length of the area requiring the quantity, and calculates the area of ​​the area requiring the quantity. Thereafter, the required quantity, i.e. the quantity of material required, is calculated based on the unit size and loading method of the quantity.

[0069] Figure 6 is a drawing for explaining the process of calculating the quantity required according to an embodiment.

[0070] In an embodiment, the processor (130) identifies areas in the drawing that require quantities of materials for quantity calculation. In an embodiment, the detection model analyzes the drawing to identify areas that require quantities.

[0071] For example, the processor (130) identifies a specific wall section in the drawing and measures the length of the identified wall. As illustrated in FIG. 6, the processor (130) may measure the length of the wall measured in the drawing as 2300 mm (2.3 m). Thereafter, the processor (130) calculates the area of ​​the section requiring materials. In an embodiment, the area may be calculated by assuming the height. As illustrated in FIG. 6, assuming the height to be 2 m, the area is calculated as 2.3 m x 2 m = 4.6 m2. Thereafter, the processor (130) calculates the required quantity of materials by considering the unit size and stacking method of bricks, etc. For example, if the brick size is 190 mm x 90 mm x 57 mm and the stacking method is 0.5B stacking with a width of 90 mm, the processor (130) may calculate that 75 bricks are required per 1 m2 when stacking 0.5 B. Accordingly, the processor (130) can calculate the required quantity of bricks as a total quantity of 4.6㎡ x 75 sheets = 345 sheets. In the embodiment, the processor (130) can systematically calculate the required quantity of bricks for a section of the drawing that requires materials by identifying the area requiring materials and measuring raw data including length and area.

[0072] In an embodiment, the processor (130) may calculate the height of raw data through a cross-sectional drawing, and the height of a beam may be calculated through a structural drawing. In addition, if an area with a beam is detected through the structural drawing, the processor (130) corrects the height of the raw data by deducting the height by the height of the beam. In an embodiment, the height of raw data may be calculated through a cross-sectional drawing. A cross-sectional drawing is a drawing showing a vertical cut surface of a building or structure, and the actual height of each point can be determined. In addition, the processor (130) may calculate the height of a beam through the structural drawing. A structural drawing is a drawing showing the location and size of each part of a building, and the location and height of a beam can be confirmed from this. To this end, the processor (130) analyzes the structural drawing to detect an area with a beam. Through this, it is determined in which part the beam is located. In an embodiment, if an area with a beam is detected, the height of the beam is deducted from the height of the raw data of the corresponding area. For example, if the raw data height at a specific point is 10 meters and there is a 2-meter high beam at that point, the corrected height will be 8 meters, which is 10 meters minus 2 meters.

[0073] In addition, the processor (130) analyzes the window drawing in the drawing through the detection model and calculates the wall area by excluding the area occupied by openings including windows and doors on the exterior wall of the building. Thereafter, the processor (130) calculates a deduction value from the calculated wall area and reflects the calculated deduction value in the calculation of material quantity requirements and accumulated data. To this end, the processor (130) analyzes the window drawing in the drawing through the detection model. The window drawing is a drawing showing the location and size of windows and doors on the exterior wall of the building. In the embodiment, the processor (130) analyzes the window drawing and calculates the area occupied by openings such as windows and doors located on the exterior wall of the building. Through this, the total opening area is calculated by adding up the areas of each opening. Thereafter, the wall area is calculated by excluding the opening area from the total area of ​​the exterior wall. For example, if the total area of ​​the exterior wall is 100 square meters and the total area of ​​the opening is 20 square meters, the wall area is 80 square meters. In addition, the processor (130) calculates the required deduction value from the calculated wall area. In an embodiment, the deduction value refers to a value that is excluded or adjusted by considering a specific factor in the process of calculating the area of ​​a building or calculating materials. The deduction value can be used for various reasons during the architectural design and construction process, and may be caused by the area of ​​openings (windows, doors, etc.), structural elements, design standards and regulations, etc. The area of ​​openings (windows, doors, etc.) is used when excluding the area occupied by openings such as windows and doors from the total area of ​​the exterior wall. For example, when calculating the area of ​​a wall, the areas of windows and doors are excluded from the actual wall area.

[0074] In the example, structural elements are used to exclude the area occupied by structural elements such as columns, beams, and fixtures. These elements often need to be excluded from the net area of ​​walls or floors. Design standards and regulations refer to cases where specific areas need to be excluded or adjusted based on building design standards or legal regulations. For example, fire safety or energy efficiency standards may require separate calculations of the area of ​​a specific section.

[0075] In addition, in the embodiment, when calculating the required quantity of building materials, which is the actual amount of building materials required, the processor (130) may apply a deduction value considering the efficiency or waste of materials. For example, a certain percentage may be deducted considering material loss that may occur during construction. In addition, in the embodiment, the processor (130) reflects the finally calculated deduction value in the calculation of the accumulated data. The accumulated data is used to calculate the amount of building materials, construction costs, etc. For example, if the wall area is 80 square meters and the deduction value is 10 square meters, 70 square meters is ultimately reflected in the accumulated data. Through this process, the processor (130) can accurately exclude the opening area of ​​the exterior wall based on the window drawing and calculate the wall area, thereby generating more accurate accumulated data.

[0076] Figure 7 is a drawing for explaining the process of reflecting the deduction value according to an embodiment.

[0077] Referring to FIG. 7, in the embodiment, the processor recognizes a space in the drawing through an artificial intelligence model for space segmentation, and recognizes a "bathroom" space in the image. This is a process of determining the length and height of the space using a space segmentation technique. Thereafter, the basic area is calculated by multiplying the length and the height. For example, if the length of the bathroom is 2.3 meters and the height is 2.45 meters, the basic area is 2.3 × 2.45 = 5.6352.3, and 5.6352.3 × 2.45 = 5.635 square meters. Thereafter, the door deduction value is calculated through a detection model. In the embodiment, an object detection technique is used to recognize a door in the bathroom space, and its area is calculated. For example, if the size of the door is 0.75 meters x 2.1 meters, the area of ​​the door is 0.75 × 2.1 = 1.5750 square meters. Thereafter, the processor (130) deducts the area of ​​the door from the basic area. For example, it is calculated as 5.635-1.575=4.065.635-1.575=4.065.635-1.575=4.06 square meters. Thereafter, the processor (130) finally calculates the area reflecting the deduction value. In this example, the final calculated value is 4.06 square meters. Referring to Fig. 7, in the embodiment, the processor (130) calculates the basic area of ​​the space through an artificial intelligence model, and then calculates the deduction value such as the area of ​​the door, and deducts it from the total area to obtain the final calculated value. This enables more accurate area calculation and generation of accumulated data.

[0078] The processor (130) maps a basis for calculation to each of the accumulated data uploaded to the calculation statement. In an embodiment, the basis for calculation includes an image of a drawing, and raw data including area and height may be output from the image of the drawing. To this end, in an embodiment, the processor (130) uploads the accumulated data to the calculation statement. In an embodiment, the accumulated data may include various information such as area, volume, and material quantity for each element of a building. Thereafter, the processor (130) analyzes the image on the drawing. The image on the drawing refers to an architectural design drawing and includes information such as the area and height of each space. Thereafter, the processor (130) extracts raw data of each space through the image. The raw data represents the actual dimensions of the quantity of materials required for construction of each space on the drawing. Thereafter, the processor (130) maps the extracted raw data to the accumulated data. In an embodiment, the basis for calculation includes an image extracted from the drawing during accumulation and dimensional data such as area and height included in the image. For example, area and height data for calculating the area of ​​a bathroom are extracted from the drawing and then mapped to the bathroom item in the estimation data. Then, the basis for calculating each estimation data item is displayed. This clarifies where and how the estimation data was derived.

[0079] In an embodiment, the calculation basis may include images on a drawing, dimensional data such as width and height, etc. The processor (130) then maps the calculation basis for each accumulated data item and generates a calculation statement. The calculation statement generated in the embodiment includes the calculation basis for each item, allowing verification of the source and accuracy of the data.

[0080] Additionally, the processor (130) can extract exception handling details through analysis of the special specifications and correct the calculation details based on the extracted exception handling details. The special specifications are documents that describe special conditions, additional requirements, exceptions, etc. of a construction project. To correct the calculation details, the processor (130) uploads the special specifications to the processor (130). Thereafter, the processor (130) analyzes the special specifications and extracts exception handling details.

[0081] In this embodiment, exceptions specified in the special specifications are identified to extract exception handling details. Exceptions may include the use of specific materials, changes in construction methods, and additional work. The processor (130) then organizes the exception handling details and stores them in a database. The processor (130) then maps each item of the estimation data with the exception handling details and identifies items to which exceptions should be applied. For example, if a special material must be used for a specific wall, the cost and quantity of that material are reflected in the estimation data. Furthermore, the processor (130) corrects the calculation statement based on the exception handling details. In this embodiment, the estimation data is modified or added for items that reflect exceptions. For example, if a special material other than the originally planned material must be used, the cost and quantity of that material are added or existing items are corrected. The processor (130) then verifies the corrected calculation statement to ensure accuracy. It verifies that all exceptions have been accurately reflected and that the quantity and cost have been calculated correctly. In this embodiment, a final calculation statement is created after verification. This report contains accurate cumulative data with exceptions reflected.

[0082] In addition, the processor (130) can select a generation for which work has been completed based on the accumulated data and generate a calculation statement for the completion billing. To this end, the processor (130) collects the accumulated data for each generation of the construction project. The collected accumulated data may include material usage, work hours, details of completed work, etc. Thereafter, the accumulated data is analyzed to identify the generation for which work has been completed. In this process, the processor (130) checks the work progress status of each generation and selects the generation for which work has been completed. Thereafter, the processor (130) sets the completion billing items. In the embodiment, the completion billing items can be set for each generation. These items may include material costs, labor costs, equipment usage fees, and other expenses. Thereafter, the processor (130) generates a calculation statement for each generation based on the accumulated data. The calculation statement includes:

[0083] This may include generation number, work completed, materials used and costs, labor costs, equipment usage fees, other costs, and total cost totals.

[0084] Figure 8 is a diagram illustrating a drawing upload interface of a 2D CAD-based automatic estimation data generation system utilizing artificial intelligence according to an embodiment. Referring to Figure 8, in the embodiment, a user terminal uploads a drawing when requesting the estimation data generation device to generate estimation data and calculation details.

[0085] FIG. 9 is a diagram illustrating a drawing viewer interface provided by an integrated data generation device according to an embodiment. Referring to FIG. 9, the integrated data generation device according to an embodiment can output analyzed drawings and integrated data calculated from the drawings together. As illustrated in FIG. 9, in the embodiment, the building, floor, real name, logistics item, area, and integrated data calculation formula corresponding to the drawing can be output together with the drawing.

[0086] Figure 10 is a drawing showing an output interface of accumulated details generated in an embodiment.

[0087] As illustrated in Fig. 10, the device for generating the accumulated data according to the embodiment can output the calculated accumulated data and the information items of each accumulated data. The information of the accumulated data output according to the embodiment can include the general contractor, region, site name, upload date, person completing the accumulated data, type of work, accumulated cost, manager in charge, progress status, etc. In addition, in the embodiment, the accumulated data can be output by applying filters according to the type of work, region, type, progress status, and general contractor. Fig. 11 is a drawing showing an interface that displays the basis for calculating the accumulated data according to the embodiment. Referring to Fig. 11, the device for generating the accumulated data according to the embodiment can confirm the object for which the accumulated data was calculated and the length and width of each object in the drawing together with the calculation formula for calculating the accumulated data.

[0088] In addition, in the past, a person had to directly find the location of the drawing and compare it to find the basis data for the calculation statement, but the calculation program according to the embodiment allows the user to immediately check the drawing through a user interface (UI, User Interface, hereinafter referred to as a verification-only UI) for verifying the basis for the calculation.

[0089] Referring to FIG. 11, the processor (130) of the accumulated data generation device (100) outputs and provides the verification-only UI to the display (not shown) of the accumulated data generation device (100) or an external display connected wired or wirelessly to the accumulated data generation device (100), and when a user selects (or inputs) a corresponding value in the calculation statement to confirm or verify the basis for calculation through the verification-only UI, a CAD image matching the value is displayed and provided.

[0090] Additionally, the processor (130) can extract raw data (e.g., attribute values) corresponding to features derived from an artificial intelligence detection (AI Detection) model and display them on the verification-only UI.

[0091] In addition, as illustrated in FIG. 11 in the embodiment, the processor (130) can display an area such as a wall or window that is the basis for calculation by overlapping it with the CAD image displayed on the verification-only UI using a method such as color or emphasis.

[0092] In addition, the processor (130) can display the attribute values ​​(e.g., width, height, area, etc.) of the actual drawing along with the walls, windows, etc. on the verification-only UI. In addition, the processor (130) can provide the verification-only UI screen by configuring it differently according to various types of construction work, including brickwork or masonry work for building a building structure and plastering or finishing work for leveling and finishing the surface of a wall or ceiling.

[0093] Fig. 12 is a diagram illustrating a user interface for providing a ready-made billing service provided in an embodiment. In the embodiment, ready-made billing is a process of requesting a corresponding cost when a certain task is completed. Referring to Fig. 12, in the embodiment, when a household for which work has been completed (e.g., February workload: 1st to 10th floors of Building 101) is selected through the user interface, ready-made billing for the selected household can be performed. At this time, the accumulated data generation device according to the embodiment generates visual data such as a graph showing logistics information such as bricks and blocks used in each area of ​​the selected household. In addition, as illustrated in Fig. 12, the accumulated data generation device can generate visual data showing the number of bricks and blocks used in each household for which construction has been completed, and the number of block meshes, reinforcing blocks, etc. used in spaces such as the master bedroom, in the form of a graph.

[0094] Hereinafter, let us examine Fig. 13. The automatic accumulation data generation method illustrated in Fig. 13 can be performed by an automatic accumulation data generation device (100) including a processor (130).

[0095] Meanwhile, Fig. 13 is merely exemplary, and the spirit of the present invention is not limited to what is illustrated in Fig. 13. For example, each step may be configured in a different order than that illustrated in Fig. 13, at least one of the steps illustrated in Fig. 13 may not be performed, or one or more steps not illustrated in Fig. 13 may be additionally performed.

[0096] Below, a method for automatically generating 2D CAD-based integrated data using artificial intelligence is sequentially described. Since the operation (function) of the method for automatically generating 2D CAD-based integrated data using artificial intelligence according to the embodiment is essentially the same as the function of the system, any description overlapping with that in FIGS. 2 to 12 will be omitted.

[0097] Figure 13 is a diagram showing the process of generating accumulated data according to an embodiment.

[0098] Referring to FIG. 13, in step S100, drawings including floor plans, cross-sections, elevations, structural drawings, window details, beam details, and finishing material details are collected, and in step S200, the collected images are preprocessed. In an embodiment, the preprocessing process may include a process of converting CAD into images and a process of databaseizing extracted objects. Thereafter, in step S300, material quantity data is extracted from the preprocessed images. In an embodiment, the material quantity includes materials, supplies, and construction items required for drawing construction. Thereafter, in step S400, the extracted material quantity data is analyzed to calculate the material quantity required for construction. In step S500, a calculation statement is created (written) based on the calculated material quantity required.

[0099] A device, method and system for generating automatic estimation data based on 2D CAD using artificial intelligence according to an embodiment generates estimation data at a lower cost than the existing 3D model-based automatic estimation method, thereby providing the effect of enabling the application of estimation technology to small and medium-sized buildings or projects with limited budgets.

[0100] Additionally, the 2D CAD-based automatic estimation data generation system utilizing artificial intelligence eliminates the need for 3D model-based estimation methods, which were limited due to high costs, thereby enabling the application of estimation technology to a wider range of buildings and projects.

[0101] In addition, the 2D CAD-based automatic estimation data generation system utilizing artificial intelligence reduces human error and increases the speed and accuracy of estimation work through the automatic estimation data generation using artificial intelligence, thereby significantly reducing construction preparation time and costs.

[0102] Furthermore, the AI-powered 2D CAD-based automatic cost data generation system generates cost data based on 2D drawings, enabling users to utilize existing 2D CAD drawings without additional modeling, thereby enhancing user experience. Furthermore, AI-powered automation technology allows cost data to be applied to projects of various shapes and sizes, increasing the technology's versatility. This, in turn, can aid in efficient cost management and budget planning across the construction industry.

[0103] Furthermore, the 2D CAD-based automatic cost data generation system utilizing artificial intelligence according to the embodiment significantly improves the cost management and efficiency of construction projects, and enables the use of cost-saving technology in more projects, thereby increasing the productivity of the overall construction industry.

[0104] A model in this specification may refer to any form of computer program that operates based on a network function, an artificial neural network, and / or a neural network. Throughout this specification, the terms model, neural network, network function, and neural network may be used interchangeably. A neural network is a network in which one or more nodes are interconnected through one or more links to form input node and output node relationships within the neural network. The characteristics of a neural network can be determined based on the number of nodes and links within the neural network, the correlation between the nodes and links, and the weight value assigned to each link. A neural network may be composed of a set of one or more nodes. A subset of the nodes constituting the neural network may constitute a layer.

[0105] A deep neural network (DNN) may refer to a neural network that includes multiple hidden layers in addition to an input layer and an output layer. A deep neural network may include a convolutional neural network (CNN), a recurrent neural network (RNN), an autoencoder, a generative adversarial network (GAN), a restricted boltzmann machine (RBM), a deep belief network (DBN), a Q network, a U network, a Siamese network, a generative adversarial network (GAN), a transformer, and the like. The description of the above-described deep neural network is merely an example, and the present disclosure is not limited thereto.

[0106] Neural networks can learn through at least one of the following methods: supervised learning, unsupervised learning, semi-supervised learning, self-supervised learning, or reinforcement learning. Neural network learning can be the process of applying knowledge to the neural network to perform a specific action.

[0107] Neural networks can be trained to minimize output errors. This process involves repeatedly inputting training data into the neural network, calculating the neural network output and target error for the training data, and backpropagating the neural network error from the output layer to the input layer to update the weights of each node in the neural network to reduce the error. In supervised learning, labeled data is used for each training data, while unsupervised learning uses unlabeled data. The amount of change in the connection weights of each updated node can be determined by the learning rate. The neural network's calculation of input data and backpropagation of errors can constitute a learning cycle (epoch). The learning rate can vary depending on the number of iterations in the neural network's training cycle. Additionally, to prevent overfitting, methods such as increasing the training data, regularization, dropout that disables some nodes, and batch normalization layers can be applied.

[0108] In one embodiment, the model may borrow at least a portion of a transformer. The transformer may be composed of an encoder that encodes embedded data and a decoder that decodes the encoded data. The transformer may have a structure that receives a series of data and outputs a series of data of different types through encoding and decoding steps. In one embodiment, the series of data may be processed into a form operable by the transformer. The process of processing the series of data into a form operable by the transformer may include an embedding process. Expressions such as data tokens, embedding vectors, and embedding tokens may refer to data embedded in a form operable by the transformer.

[0109] To encode and decode a series of data, a transformer can utilize an attention algorithm to process the encoders and decoders within the transformer. An attention algorithm can refer to an algorithm that, for a given query, calculates the similarity for one or more keys, reflects this similarity in the values ​​corresponding to each key, and then weights and sums the values ​​to which the similarity is reflected to calculate an attention value.

[0110] Depending on how the query, key, and value are configured, various types of attention algorithms can be categorized. For example, if attention is obtained by setting the query, key, and value all to the same value, this could be a self-attention algorithm. If attention is obtained by reducing the dimensionality of the embedding vector to process a series of input data in parallel and then generating individual attention heads for each segmented embedding vector, this could be a multi-head attention algorithm.

[0111] In one embodiment, the transformer may be composed of modules that perform multiple multi-head self-attention algorithms or multi-head encoder-decoder algorithms. In one embodiment, the transformer may also include additional components other than attention algorithms, such as embedding, normalization, and softmax. Methods for constructing a transformer using attention algorithms may include methods disclosed in Vaswani et al., Attention Is All You Need, 2017 NIPS, which is incorporated herein by reference.

[0112] A transformer can be applied to various data domains, such as embedded natural language, segmented image data, and audio waveforms, to transform a series of input data into a series of output data. To transform data with various data domains into a series of data that can be input to a transformer, the transformer can embed the data. The transformer can process additional data that expresses the relative positional relationship or phase relationship between the series of input data. Alternatively, vectors expressing the relative positional relationship or phase relationship between the input data can be additionally reflected in the series of input data to embed the series of input data. In one example, the relative positional relationship between the series of input data may include, but is not limited to, word order within a natural language sentence, the relative positional relationship between each segmented image, and the time order of segmented audio waveforms. The process of adding information expressing the relative positional relationship or phase relationship between the series of input data may be referred to as positional encoding.

[0113] In one embodiment, the model may include, but is not limited to, at least one of a Recurrent Neural Network (RNN), a Long Short Term Memory (LSTM) network, a Deep Neural Network (DNN), a Convolutional Neural Network (CNN), and a Bidirectional Recurrent Deep Neural Network (BRDNN).

[0114] In one embodiment, the model may be a model trained using transfer learning. Transfer learning, in this context, refers to a learning method that pre-trains a large amount of unlabeled training data using semi-supervised or self-learning methods to obtain a pre-trained model for a first task, then fine-tunes the pre-trained model to suit a second task, and trains it on labeled training data using supervised learning to implement a target model.

[0115] The disclosed content is merely an example, and various modifications and implementations can be made by a person skilled in the art without departing from the gist of the claims claimed in the patent, so the scope of protection of the disclosed content is not limited to the specific embodiments described above.

Claims

A memory storing at least one command for generating automatic 2D CAD-based integrated data using artificial intelligence; and A processor comprising: a processor that performs an operation according to the above command; The above processor preprocesses drawings collected through an artificial intelligence model, extracts the quantity of materials required for construction from the preprocessed drawings, calculates the quantity required based on the analysis results of the extracted quantity, generates estimated data based on the calculated quantity of materials required, and generates a calculation statement applying the generated estimated data. The processor, when calculating the required quantity, identifies the required area of ​​the quantity including the material in the drawing through the artificial intelligence model, measures the length of the required area of ​​the quantity, calculates the area of ​​the required area of ​​the quantity based on the measured length, and calculates the required quantity of the quantity based on the unit size and loading method of the quantity. The processor, when generating the accumulated data, scans a drawing input from a user terminal and converts it into a CAD image, recognizes the quantity including construction items, materials and ingredients required for the construction in the preprocessed drawing through an artificial intelligence-based detection model, derives a feature of the recognized quantity, loads raw data at drawing coordinates corresponding to the coordinates of the feature based on the derived feature, and generates the accumulated data for the quantity required using the loaded raw data. The processor, when generating the calculation statement, recognizes a residential space including a bedroom, a living room, and a bathroom in the CAD image through the artificial intelligence-based space recognition model, collects the accumulated data that calculates the quantity based on the loaded raw data, creates the calculation statement based on the name of the recognized residential space and the accumulated data, and displays a part corresponding to the accumulated element included in the calculated statement as a line in the drawing image to generate basis data for the accumulated element included in the calculation statement through the artificial intelligence-based line drawing model, and extracts exception processing details based on the special conditions, additional requirements, and exceptions of the building project specified in the special specifications in the drawing through the detection model, and corrects the calculation statement according to the extracted exception processing details. The above processor is a data generation device that, when correcting the above production statement, adds the cost and quantity of the special material if a special material other than the originally planned material must be used among the above exceptions, and corrects the production statement based on the cost and quantity of the added special material. In the first paragraph, Among the above raw data, the height is calculated through a cross-section, and the height of the beam is calculated through a structural drawing. The processor is a data generation device that, when an area with the beam is detected through the structural diagram, corrects the height of the raw data by deducting the height by the height of the beam. In the first paragraph, The above processor, By analyzing the window drawing in the above drawing, the wall area excluding the area occupied by openings including windows and doors on the exterior wall of the building is calculated, Calculate the deduction value due to the opening from the calculated wall area above, An accumulation data generation device that reflects the calculated deduction value in the generation of the accumulation data. In the first paragraph, The above processor, Mapping the basis for calculation to each of the accumulated data uploaded to the above calculation statement, The above calculation basis includes an image of the drawing, The above image is a cumulative data generation device that outputs raw data including the length and height of the quantity. In paragraph 4, The above processor provides a user interface for verifying the above calculation basis, The above user interface, Print out the drawing to confirm the above calculation basis, A data generation device that provides a CAD image matching the above-mentioned value when the value requiring verification is selected. In paragraph 5, The above processor, Through the above user interface, the color of the area corresponding to the output basis including the wall and window is changed or emphasized, and output by overlapping it with the CAD image, A data generation device that displays the attribute values ​​of the drawing including the width, height and area of ​​the wall and the window included in the area corresponding to the above calculation basis. In the first paragraph, The above processor, A device for generating accumulated data that selects a generation for which work has been completed based on the above accumulated data and generates a calculation statement for a ready-made claim. In a control method of an automatic 2D CAD-based data generation device utilizing artificial intelligence, performed by a processor, A step of preprocessing the drawings collected through an artificial intelligence model; A step of extracting the quantity required for construction from the above preprocessed drawing; A step of calculating the required quantity based on the analysis results of the extracted quantity; A step of generating accumulated data according to the above calculated quantity requirement; and It includes a step of creating a calculation statement by applying the above-generated accumulated data, The above steps for calculating the required quantity are: A step of identifying the required area of ​​the quantity including the material in the drawing through the artificial intelligence model; A step of measuring the length of the required area of ​​the above quantity; A step of calculating the area of ​​the required area of ​​the quantity based on the measured length; and It includes a step of calculating the required quantity of the quantity according to the unit size and loading method of the quantity of the quantity, The above accumulated data generation step is: A step of scanning a drawing input from a user terminal and converting it into a CAD image; A step of recognizing the quantity of construction items, materials and supplies required for the construction within the preprocessed drawing using an artificial intelligence-based detection model; A step of deriving features of the above recognized quantity; A step of loading raw data in drawing coordinates corresponding to the coordinates of the feature based on the above-described feature; and Including a step of generating the accumulated data for the required quantity using the loaded raw data, The above output statement generation step is: A step of recognizing a residential space including a bedroom, living room, and bathroom in the CAD image using the artificial intelligence-based spatial recognition model; A step of collecting accumulated data that calculates the quantity based on the loaded raw data; A step of generating the calculation statement based on the name of the recognized residential space and the accumulated data; A step of displaying a portion corresponding to the accumulated element included in the accumulated data as a line in a drawing image to generate basis data for the accumulated element included in the calculated statement through the artificial intelligence-based line drawing model; A step of extracting exception handling details based on special conditions, additional requirements and exceptions of the building project specified in the special specifications in the drawing through the above detection model; and Including a step of correcting the above-mentioned output statement according to the above-mentioned extracted exception handling details, The above calculation statement correction step is: In the case where special materials other than the originally planned materials must be used among the above exceptions, a step of adding the cost and quantity of the special materials; and A control method of a data generation device, comprising a step of correcting the calculation statement based on the cost and quantity of the special material added above. In paragraph 8, Among the above raw data, the height is calculated through a cross-section, and the height of the beam is calculated through a structural drawing. A control method for a data generation device, further comprising a step of correcting the height of the raw data by deducting the height by the height of the beam when an area with the beam is detected through the above structural diagram. In paragraph 8, The above accumulated data generation step is: A step of analyzing the window drawing in the above drawing and calculating the wall area excluding the area occupied by openings including windows and doors on the exterior wall of the building; A step of calculating a deduction value due to the opening from the calculated wall area; and A control method of an integrated data generation device, further comprising a step of reflecting the calculated deduction value in the generation of the integrated data. In paragraph 8, Further comprising a step of mapping the basis for calculation to each of the accumulated data uploaded to the above calculation statement, The above calculation basis includes an image of the drawing, A control method for a cumulative data generation device, in which raw data including the length and height of the quantity is output in the image above. In Article 11, Further comprising a step of providing a user interface for verifying the above calculation basis, The above user interface, Print out the drawing to confirm the above calculation basis, A control method of a data generation device that provides a CAD image matching the above-mentioned value when the value requiring verification is selected. In paragraph 12, A step of changing the color of an area corresponding to the output basis including walls and windows through the user interface or emphasizing the color and overlapping it with the CAD image and outputting it; and A control method of a data generation device, further comprising a step of displaying attribute values ​​of the drawing including the width, height and area of ​​the wall and the window included in the area corresponding to the calculation basis. In paragraph 8, The above output statement generation step is: A control method of a cumulative data generation device, further comprising a step of selecting a generation for which work has been completed based on the cumulative data and generating a calculation statement for a ready-made claim.

Images