JP2019200721A - Design support device and design support model learning device - Google Patents

Abstract

To automatically generate conditions of designing a building from image information or language information.SOLUTION: An information acquisition unit 22 of a design support device 100 acquires input information showing image information on a building and language information on a building. A design information output unit 24 of the design support device 100 inputs the input information acquired by the information acquisition unit 22 into a learned model having learned in advance according to the learning data and outputs the condition of designing a design target building by an operation using the learned model.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a design support apparatus and a design support model learning apparatus.

  Conventionally, an aggregation system using a virtual three-dimensional model of an assembly of parts constituting a building is known (for example, Patent Document 1). Moreover, a processing system using a virtual three-dimensional model of an assembly of parts constituting a building is known (for example, Patent Document 2). In addition, a design support system that links the design of a refrigerator with human image and sensitivity is known (for example, Patent Document 3).

Japanese Unexamined Patent Publication No. 2016-115040 JP 2017-224014 JP 2004-295619 A

  A request regarding a design image of a building may be presented to a construction company by image or language. However, the techniques described in Patent Document 1 to Patent Document 3 cannot automatically generate building design conditions from images or languages.

  In view of the above facts, an object of the present invention is to automatically generate a building design condition from image information or language information.

  In order to achieve the above object, the design support apparatus of the present invention is acquired by an information acquisition unit that acquires input information that represents at least one of image information that represents a building and language information that represents a building, and the information acquisition unit. Based on learning data representing a combination of at least one of learning image information representing a building for learning and learning language information representing a building for learning, and design conditions of the building for learning, the input information A design information output unit that inputs to a learned model that has been learned and outputs a design condition of a building that is a design target by calculation using the learned model. Thereby, the design condition of a building can be automatically generated from image information or language information.

  The design information output unit of the design support apparatus of the present invention may further output a BIM (Building Information Modeling) model corresponding to the design condition. Thereby, the BIM model according to the design condition of a building can be automatically generated from image information or language information.

  In the design support apparatus of the present invention, the input information is image information representing the building, and the rendering process is performed on each of the BIM models according to the design condition, thereby corresponding to each of the BIM models. 3D image information generating unit for generating 3D image information to be performed and image information obtained by viewing the 3D image information from a predetermined position for each of the 3D image information generated by the 3D image information generating unit And an evaluation unit that calculates evaluation information representing a degree of coincidence between the input information and the image information representing the building. Thereby, the degree of coincidence between the input information and the three-dimensional image information corresponding to the BIM model according to the design condition can be automatically evaluated.

  The design support apparatus of the present invention outputs, based on the evaluation information calculated by the evaluation unit, output at least one of the three-dimensional image information and the BIM model in which the evaluation information is greater than or equal to a preset threshold value. A part may be further included. Thereby, appropriate three-dimensional image information according to the input information can be obtained.

  The design condition may be a design condition related to at least one of the exterior and interior of the building. Thereby, the design condition of at least one of the exterior and interior of a building can be output.

  The design support model learning device of the present invention is a learning that represents a combination of at least one of learning image information representing a building for learning and learning language information representing a building for learning, and the design conditions of the building for learning. A learning unit that learns a model for outputting design conditions of the building from at least one of image information representing the building and linguistic information representing the building based on the business data. Thereby, a learned model for automatically generating a building design condition from image information and language information can be obtained.

  According to the present invention, it is possible to automatically generate a building design condition from image information or language information.

It is a block diagram which shows an example of a structure of the design assistance apparatus which concerns on this embodiment. It is explanatory drawing for demonstrating the image information of this embodiment. It is a figure which shows an example of the learned model of this embodiment. It is a figure which shows an example of the part of the rectangular figure among BIM models. It is a block diagram which shows an example of a structure of the design support model learning apparatus which concerns on this embodiment. It is explanatory drawing for demonstrating the data for learning of this embodiment. It is a figure which shows an example of the learning process routine which concerns on this embodiment. It is a figure which shows an example of the design assistance process routine which concerns on this embodiment. It is a figure which shows an example of a sensitivity evaluation sheet.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

<Configuration of Design Support Device According to Embodiment of the Present Invention>

  FIG. 1 shows an example of the configuration of a design support apparatus 100 according to an embodiment of the present invention. Functionally, the design support apparatus 100 can be represented by a configuration including a reception unit 10, a computer 20, and a display unit 40, as shown in FIG. The design support apparatus 100 according to the present embodiment supports architectural design performed by an architect.

  In the field of architectural design, the use of BIM (Building Information Modeling) is progressing. When an architectural designer performs architectural design, he / she proceeds with design work by creating a BIM model representing a three-dimensional model of a building to be designed.

By utilizing BIM, an object representing the shape of a building is formed in a space on the computer, and attribute information is given to the object. For example, as attribute information, information on physical characteristics (specific gravity, material, strength, rigidity, etc.) or information on costs (for example, unit price per 1 m ² and construction cost, etc.) can be linked to an object. It becomes. Therefore, for example, by linking the BIM model and various analysis tools, structural analysis, thermal load analysis, and the like can be automatically performed according to the BIM model.

  When designing a building, the architect considers the building owner's request regarding the exterior or interior of the building, the cost of the building, and various performances of the building (for example, wind resistant load and thermal insulation performance), and Design conditions need to be determined. This work is performed based on the sense and experience of the architect.

  However, since the image regarding the exterior or interior of the building depends on the individual's sense, it is difficult to appropriately reflect the request of the architect regarding the exterior or interior of the building in the design conditions. In addition, it is difficult to propose a plurality of design conditions related to the exterior or interior of a building to the architect in a short time.

  Therefore, in this embodiment, regarding the design of the exterior or interior of a building, image information and language information are input information and design conditions for the exterior or interior of the building are output. Since the image information and the language information reflect the potential demands of the building owner, according to the present embodiment, design conditions according to the potential demands of the building owner can be output. Further, according to the present embodiment, the efficiency of building design can be improved by automatically generating a BIM model according to the design conditions.

  As design conditions for the exterior of a building, for example, an opening ratio for creating an exterior image, a proportion of a window, a dimension of a waist wall or a hanging wall, a dimension of an opening depth, a material of the exterior, and the like are output as design conditions. In addition, as design conditions for the interior of a building, for example, an aperture ratio that fosters the image of the interior, a material of a wall, and the like are output as design conditions.

  Hereinafter, this embodiment will be specifically described.

  The reception unit 10 receives input information input from a user. The input information of this embodiment is image information representing a building and language information representing a building. The accepting unit 10 is realized by, for example, a keyboard, a mouse, or an input / output device that accepts input from an external device. In this embodiment, a case where the exterior of a building is a design object will be described as an example.

  For example, a building owner selects a picture of the exterior of a building when a new construction project is started. For example, as shown in FIG. 2, the building owner selects a photo close to the image the building owner has from photos P 1, P 2 and P 3 of a plurality of buildings already built. Then, the user of the design support apparatus 100 inputs the photograph selected by the building owner to the receiving unit 10 as image information. In addition, the photograph input as image information may be single or plural.

  The building owner selects a word representing an image of the exterior of the building that is the building target. For example, words representing building images such as “warm”, “open”, “modern”, “healthy”, “nature”, and “harmony” are selected. And the user of the design support apparatus 100 inputs the word selected by the building owner into the reception unit 10 as language information. In addition, the word input as language information may be single or plural. Further, the input language information may be a sentence.

  The computer 20 includes a CPU (Central Processing Unit), a ROM (Read Only Memory) that stores a program for realizing each processing routine, a RAM (Random Access Memory) that temporarily stores data, and a memory as a storage means. And a network interface and the like. As shown in FIG. 1, the computer 20 functionally includes an information acquisition unit 22, a learned model storage unit 23, a design information output unit 24, a 3D image information generation unit 26, and an evaluation unit 28. And an output unit 30.

  The information acquisition unit 22 acquires input information received by the reception unit 10.

  The learned model storage unit 23 stores a learned model learned in advance based on learning data. The learning data of the present embodiment is data representing a combination of learning image information representing a learning building, learning language information representing a learning building, and design conditions of the learning building. The probabilities of building design condition candidates are output from the learned model. Details of the learned model will be described later.

  The design information output unit 24 inputs the input information acquired by the information acquisition unit 22 to the learned model stored in the learned model storage unit 23, and calculates the building that is the design target by calculation using the learned model. Output design conditions. The design conditions output in the present embodiment are design conditions related to the exterior of the building.

  In the present embodiment, a learned model as shown in FIG. 3 is used. For example, as shown in FIG. 3, a neural network can be used as an example of a model, and deep learning can be used as an example of a learning algorithm. The learned model shown in FIG. 3 outputs, for example, probabilities related to design condition candidates. The input information shown in FIG. 3 is expressed in a format that can be read by the neural network. For example, as language information, a frequency vector representing the frequency of appearance of each word is input to the neural network. When a sentence is input as language information, a frequency vector representing the frequency of appearance of each word included in the sentence is input to the neural network. As image information, pixel values of a photograph of a building are input to the neural network.

  Design conditions include specifications related to the exterior of the building (for example, information on the type of outer wall (RC or ALC, etc.), information on the opening ratio, information on the arrangement of windows, information on the dimensions of the waist wall or hanging wall, and opening depth dimensions. Information etc.) is used. For example, as the probabilities of design condition 1 output from the learned model, there are probabilities of outer wall type: ALC, opening ratio: A1, window arrangement: B1, waist wall dimension: C1, and opening depth dimension: D1. Is output.

  Next, the design information output unit 24 generates a BIM model according to the design conditions. Specifically, the design information output unit 24 includes information represented by each of the design conditions (information regarding the type of outer wall (RC or ALC, etc.), information regarding the aperture ratio, information regarding the arrangement of windows, and the dimensions of the waist wall or the hanging wall. The BIM model is generated in accordance with the information on the opening depth dimension and the like. Regarding the generation of the BIM model, an object representing the shape of the outer wall of the building is generated in accordance with information on dimensions and the like. For example, as shown in FIG. 4, each part (window Wi, top light TL, outer wall Wa, etc.) corresponding to the rectangular view of the BIM model is generated according to the information represented by each design condition. . Also, attribute information representing design conditions is assigned to the BIM model object.

  The three-dimensional image information generation unit 26 generates three-dimensional image information corresponding to each of the BIM models by performing rendering processing on each of the BIM models according to the design conditions output by the design information output unit 24. To do. As the three-dimensional image information, for example, a perspective representing an exterior image of a building is generated.

  For each of the three-dimensional image information generated by the three-dimensional image information generation unit 26, the evaluation unit 28 includes image information obtained by viewing the three-dimensional image information from a predetermined position, and image information representing a building in the input information, Evaluation information representing the degree of matching is calculated. For example, the predetermined position is set by the user so that image information in the three-dimensional image information corresponds to image information representing a building in the input information.

  For example, the degree of coincidence between the color information of the image information obtained by viewing the 3D image information from a predetermined position and the color information of the image information of the input information, or the window of the image information of the 3D image information The degree of coincidence between the shape and the shape of the window of the image information in the input information is calculated as evaluation information.

  Based on the evaluation information calculated by the evaluation unit 28, the output unit 30 outputs three-dimensional image information whose evaluation information is greater than or equal to a preset threshold value. Note that the output unit 30 may output the BIM model generated by the design information output unit 24 together with the 3D image information to be output. The output unit 30 may output the BIM model generated by the design information output unit 24 instead of the three-dimensional image information.

  The display unit 40 displays the 3D image information output by the output unit 30.

  The user confirms the 3D image information displayed by the display unit 40. The user checks whether the three-dimensional image information displayed by the display unit 40 matches the image information of the input information and the image of the language information. In addition, the user presents the 3D image information displayed on the display unit 40 to the builder and confirms whether or not the builder meets the request of the builder.

<Configuration of Design Support Model Learning Device According to Embodiment of the Present Invention>

  FIG. 5 shows an example of the configuration of the design support model learning apparatus 200 according to the embodiment of the present invention. Functionally, the design support model learning device 200 can be represented by a configuration including a data receiving unit 210 and a computer 220 as shown in FIG.

  The data receiving unit 210 receives learning data for generating a learned model.

  The computer 220 includes a CPU, a ROM that stores programs for realizing each processing routine, a RAM that temporarily stores data, a memory as a storage unit, a network interface, and the like. As shown in FIG. 5, the computer 220 functionally includes a learning data acquisition unit 221, a learning data storage unit 223, a learning unit 225, and a learned model storage unit 226.

  The learning data acquisition unit 221 acquires the learning data received by the data reception unit 210. Then, the learning data acquisition unit 221 stores the learning data in the learning data storage unit 223. The learning data of the present embodiment is data representing a combination of learning image information representing a learning building, learning language information representing a learning building, and design conditions of the learning building.

  The learning data storage unit 223 stores the learning data acquired by the learning data acquisition unit 221. FIG. 6 is an explanatory diagram for explaining the learning data. As shown in FIG. 6, the learning data of this embodiment corresponds to learning image information representing a learning building, learning language information representing a learning building, and design conditions of the learning building. Attached and stored. In the learning data with the data ID “00001” shown in FIG. 6, when the learning image information is “XXX1” and the learning language information is “YYY1”, the design condition of the building for learning is It is shown that it was “ZZZ1”.

  As the learning language information, for example, language information representing an image of a building such as “warmth”, “modern”, “healthy”, “nature”, and “harmony” is set in advance. When learning a learned model, for example, a frequency vector representing the frequency of appearance of words is used as language information. As the learning image information, for example, image information of a photograph of the appearance of a learning building is used.

  Based on a plurality of learning data stored in the learning data storage unit 223, the learning unit 225 learns a model for outputting the design conditions of the building from the image information representing the building and the language information representing the building. To obtain a trained model. Thereby, a learned model as shown in FIG. 3 is generated.

  The learned model storage unit 226 stores the learned model learned by the learning unit 225.

<Operation of Design Support Model Learning Device 200>

  Next, the operation of the design support model learning device 200 will be described. When the data receiving unit 210 of the design support model learning device 200 receives the input of learning data, the data is stored in the learning data storage unit 223. When the computer 220 of the design support model learning device 200 receives the instruction signal for executing the learning process, the computer 220 executes a learning process routine shown in FIG.

  In step S <b> 100, the learning unit 225 acquires a plurality of learning data stored in the learning data storage unit 223.

  In step S102, the learning unit 225 learns a model for outputting the design condition of the building from the image information representing the building and the language information representing the building based on the plurality of learning data acquired in step S100. To obtain a trained model.

  In step S104, the learning unit 225 stores the learned model generated in step S102 in the learned model storage unit 226, and ends the learning process routine.

<Design support processing routine>

  When the learned model generated by the design support model learning apparatus 200 is input to the design support apparatus 100, it is stored in the learned model storage unit 23 of the design support apparatus 100. When the computer 20 of the design support apparatus 100 receives the design support process execution instruction signal, the computer 20 executes a design support process routine shown in FIG.

  In step S200, the information acquisition unit 22 acquires the input information received by the reception unit 10.

  In step S202, the design information output unit 24 inputs the input information acquired by the information acquisition unit 22 to the learned model stored in the learned model storage unit 23, and performs the design object by calculation using the learned model. The candidate of the design condition of the building is output.

  In step S204, the design information output unit 24 generates each BIM model corresponding to each of the design condition candidates output in step S202.

  In step S206, the three-dimensional image information generation unit 26 performs rendering processing on each of the BIM models generated in step S204, thereby generating three-dimensional image information corresponding to each of the BIM models.

  In step S208, the evaluation unit 28 includes, for each of the three-dimensional image information generated in step S206, image information obtained by viewing the three-dimensional image information from a predetermined position and the input information acquired in step S100. Evaluation information representing the degree of coincidence with the image information is calculated.

  In step S210, the output unit 30 selects three-dimensional image information whose evaluation information is greater than or equal to a preset threshold value based on each of the evaluation information calculated in step S208.

  In step S212, the output unit 30 outputs the 3D image information selected in step S210 as a result, and ends the design support processing routine.

  The user of the design support apparatus 100 confirms the 3D image information displayed by the display unit 40. Then, the user confirms whether the three-dimensional image information displayed by the display unit 40 matches the image information of the input information and the image of the language information. In addition, the user presents the 3D image information displayed on the display unit 40 to the builder and confirms whether or not the builder meets the request of the builder.

  As described above in detail, the design support apparatus 100 according to the present embodiment inputs image information representing a building and input information representing language information representing a building to a learned model that has been learned in advance based on learning data. Then, the design condition of the building to be designed is output by calculation using the learned model. Thereby, building design conditions can be automatically generated from image information and language information.

  Further, according to the design support model learning device 200 of the present embodiment, a model for outputting design conditions of the building is learned from image information representing the building and language information representing the building based on the learning data. Thus, a learned model for automatically generating a building design condition from image information and language information can be obtained.

  In addition, according to the present embodiment, in the design of the exterior, which is determined through extensive meetings with the building owner under various required performances and constraints such as cost, aesthetics, heat insulation performance, and load, It is possible to present a plurality of design candidates that can match the main required performance, particularly image information and linguistic information that are directly related to the sense, in a short time.

  The present invention is not limited to the above-described embodiment, and various modifications and applications can be made without departing from the gist of the present invention.

  For example, in the above embodiment, a case where a neural network model as an example of a model is learned by deep learning has been described as an example, but the present invention is not limited to this. For example, another model different from the neural network model may be learned by another learning method different from deep learning. For example, a model used in statistical analysis may be learned.

  Moreover, although this embodiment demonstrated to the example the case where the design of the exterior of a building was made into object, it is not limited to this. For example, the design of the interior of a building may be targeted.

  In this embodiment, the case where image information and language information are used has been described as an example. However, the present invention is not limited to this, and either image information or language information may be used.

  Further, by linking a predetermined analysis program and a database storing performance information related to the exterior, the performance (for example, heat insulation performance, load bearing performance, cost, etc.) of the generated BIM model is further calculated. You may do it. By linking the BIM model with an analysis program or the like, it is possible to automatically perform heat insulation performance calculation, load bearing performance calculation, cost calculation, etc. corresponding to the exterior of the design condition candidate. Thereby, the candidate of the design conditions which satisfy | filled the constraint conditions regarding heat insulation performance, load-bearing performance, cost, etc. can be shown efficiently to the builder.

  Further, when outputting the design condition candidates, a reference rectangular figure may be selected from a predetermined database and displayed. Further, when outputting a BIM model corresponding to the exterior according to the design condition, information on past quality problems (for example, cracks, water leakage, tile peeling, etc.) may be further output. In this case, a warning and a countermeasure method may be further recommended.

  Moreover, you may make it input the keyword obtained by mining the conversation at the time of hearing with an architect as language information.

  Further, learning data may be collected using a sensitivity evaluation sheet. For example, image information and language information in the learning data can be collected using a sensitivity evaluation sheet as shown in FIG. In the sensitivity evaluation sheet shown in FIG. 9, for example, a plurality of sentences Se and a photograph Im representing the exterior of a building are associated with each viewpoint of S1 to S5. In this case, for example, image information and language information as learning data can be collected by conducting a questionnaire as to which sentence corresponding to the picture Im corresponds to Se. In addition, the design condition of the building shown in the photograph Im can be used as the design condition as the learning data.

  Further, when operating the design support apparatus 100, a reward is given to the design condition selected by the building owner among a plurality of design condition candidates output from the learned model, and the learned model is learned again. You may do it. This makes it possible to preferentially output candidates that can be easily selected by the owner.

  In addition, regarding the exterior of the building, further considering the orientation of each surface of the building and the tangential condition (frontality), etc., the exterior surface of the building is distinguished, and design condition candidates are output for each distinguished exterior surface. You may do it.

  In the above description, the program is stored (installed) in a storage unit (not shown) in advance. However, the program is recorded on any recording medium such as a CD-ROM, a DVD-ROM, and a micro SD card. It is also possible to provide it in the form in which it is provided.

DESCRIPTION OF SYMBOLS 10 Reception part 20,220 Computer 22 Information acquisition part 23,226 Learned model memory | storage part 24 Design information output part 26 3D image information generation part 28 Evaluation part 30 Output part 40 Display part 100 Design support apparatus 200 Design support model learning apparatus 210 Data reception unit 221 Learning data acquisition unit 223 Learning data storage unit 225 Learning unit

Claims (6)

An information acquisition unit for acquiring input information representing at least one of image information representing a building and language information representing a building;
The input information acquired by the information acquisition unit is a combination of at least one of learning image information representing a learning building and learning language information representing a learning building, and a design condition of the learning building A design information output unit that inputs to a learned model that has been learned in advance based on learning data that represents, and outputs a design condition of a building that is a design target by calculation using the learned model;
Design support device including The design information output unit further outputs a BIM (Building Information Modeling) model according to the design conditions.
The design support apparatus according to claim 1. The input information is image information representing the building,
A three-dimensional image information generation unit that generates three-dimensional image information corresponding to each of the BIM models by performing rendering processing on each of the BIM models according to the design conditions;
For each of the three-dimensional image information generated by the three-dimensional image information generation unit, the degree of coincidence between the image information obtained by viewing the three-dimensional image information from a predetermined position and the image information representing the building in the input information An evaluation unit that calculates evaluation information representing
The design support apparatus according to claim 2. An output unit that outputs at least one of the three-dimensional image information and the BIM model in which the evaluation information is greater than or equal to a preset threshold based on the evaluation information calculated by the evaluation unit;
The design support apparatus according to claim 3. The design condition is a design condition related to at least one of the exterior and interior of the building.
The design support apparatus of any one of Claims 1-3. An image representing a building based on learning data representing a combination of at least one of learning image information representing a building for learning and learning language information representing a building for learning, and design conditions of the building for learning A design support model learning device including a learning unit that learns a model for outputting a design condition of a building from at least one of information and language information representing the building.