System and program

Abstract

SOLUTION: A system includes: a classification model acquisition unit which acquires a classification model for classifying texture data of a building into classes; a classification unit which classifies the texture data of the building into classes, using the classification model acquired by the classification model acquisition unit; an estimation model acquisition unit which acquires a material estimation model configured to estimate a material corresponding to the class of the texture data of the building, from the texture data of the building classified into classes; an estimation unit which estimates a material corresponding to the class of the texture data of the building, from the texture data of the building classified into classes by the classification unit, using the material estimation model acquired by the estimation model acquisition unit; and a map data generation unit which generates three-dimensional map data including the 3D model of the building by associating the 3D model of the building with material data indicating the material corresponding to the class of the texture data of the building, on the basis of the result estimated by the estimation unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a system and a program. [Background technology]

[0002] Patent Document 1 describes a physical property prediction method, a physical property prediction system, and a program capable of predicting the physical property values ​​of a material using machine learning. Patent Document 2 describes an image processing device and a program capable of easily creating high-quality 3DCG data using previously created low-quality 3DCG data. Patent Document 3 describes an annotation device and method that enables simple and high-precision annotation. Patent Document 4 describes a more effective and robust technique for generating UV mapping seams for 3D models. Patent Document 5 describes a technique for improving the estimation accuracy of radio wave propagation characteristics by machine learning using spatial image data. [Prior art document] [Patent documents] [Patent Document 1] Patent No. 7078944 [Patent Document 2] JP 2023-039470 A [Patent Document 3] JP 2023-080509 A [Patent Document 4] Patent No. 7129529 [Patent Document 5] JP 2023-060988 A Summary of the Invention [Means for solving the problem]

[0003] According to an embodiment of the present invention, a system is provided. The system may include a classification model acquisition unit that acquires a classification model for classifying texture data of a building into classes. The system may include a classification unit that classifies the texture data of a building into classes using the classification model acquired by the classification model acquisition unit. The system may include an estimation model acquisition unit that acquires a material estimation model that estimates a material corresponding to a class of texture data of a building from the texture data of the building classified into a class. The system may include an estimation unit that estimates a material corresponding to a class of texture data of the building from the texture data of the building classified into a class by the classification unit, using the material estimation model acquired by the estimation model acquisition unit. The system may include a map data generation unit that generates three-dimensional map data including a 3D model of the building by associating material data indicating a material corresponding to a class of the texture data of the building with the 3D model of the building based on a result of estimation by the estimation unit.

[0004] The system may further include a first learning data storage unit that stores first learning data including texture data of buildings classified into classes and material data indicating a material corresponding to the class of the texture data of the buildings, and an estimation model generation unit that generates the material estimation model by machine learning using a plurality of the first learning data stored in the first learning data storage unit as first teacher data, wherein the estimation model acquisition unit may acquire the material estimation model generated by the estimation model generation unit, and the estimation unit may use the material estimation model generated by the estimation model generation unit to estimate a material corresponding to the class of the texture data of the buildings, from the texture data of the buildings classified into classes by the classification unit, using the material estimation model generated by the estimation model generation unit.

[0005] Any of the systems may further include a second learning data storage unit that stores second learning data including texture data of a building and a classification result of classifying the texture data of the building into classes, and a classification model generation unit that generates the classification model by machine learning using a plurality of the second learning data stored in the second learning data storage unit as second teacher data, wherein the classification model acquisition unit may acquire the classification model generated by the classification model generation unit, and the classification unit may classify the texture data of the building into classes using the classification model generated by the classification model generation unit.

[0006] Any of the systems includes a second learning data storage unit that stores second learning data including texture data of a building and a classification result of classifying the texture data of the building into classes; a classification model generation unit that uses a plurality of the second learning data stored in the second learning data storage unit as second teacher data to generate the classification model by machine learning; a first learning data storage unit that stores first learning data including the texture data of the building classified into classes and material data indicating a material corresponding to the class of the texture data of the building; and a classification model generation unit that generates the material estimation model by machine learning using a plurality of the first learning data stored in the first learning data storage unit as first teacher data. the classification model acquisition unit may acquire the classification model generated by the classification model generation unit, the classification unit may classify the building texture data into classes using the classification model generated by the classification model generation unit, the estimation model acquisition unit may acquire the material estimation model generated by the estimation model generation unit, and the estimation unit may estimate a material corresponding to the class of the building texture data from the building texture data classified into classes by the classification unit, using the material estimation model generated by the estimation model generation unit.

[0007] In any of the systems described above, the estimation unit may estimate, from the texture data of the building classified into classes, whether a material corresponding to a class of the texture data of the building is concrete or glass.

[0008] Any of the systems may further include a simulation unit that uses the three-dimensional map data generated by the map data generation unit to perform a simulation of radio wave propagation in an area corresponding to the three-dimensional map data.

[0009] Any of the systems may further include a simulation unit that uses the three-dimensional map data generated by the map data generation unit to perform a simulation of a damage situation of the building in the event of a disaster occurring to the building located within the area corresponding to the three-dimensional map data.

[0010] Any of the systems may further include a simulation unit that uses the three-dimensional map data generated by the map data generation unit to perform a simulation of the amount of solar radiation irradiated inside the building located within the area corresponding to the three-dimensional map data.

[0011] According to an embodiment of the present invention, a system is provided. The system may include a classification model acquisition unit that acquires a classification model for classifying texture data of a building into classes. The system may include a classification unit that classifies the texture data of a building into classes using the classification model acquired by the classification model acquisition unit. The system may include an estimation model acquisition unit that acquires a reflection coefficient estimation model that estimates a reflection coefficient of a radio wave corresponding to a class of the texture data of a building from the texture data of the building classified into a class. The system may include an estimation unit that estimates a reflection coefficient of a radio wave corresponding to a class of the texture data of the building from the texture data of the building classified into a class by the classification unit, using the reflection coefficient estimation model acquired by the estimation model acquisition unit. The system may include a map data generation unit that generates three-dimensional map data including a 3D model of the building by associating a reflection coefficient of a radio wave corresponding to a class of the texture data of the building with a 3D model of the building based on a result of estimation by the estimation unit.

[0012] The system includes a second learning data storage unit that stores second learning data including texture data of buildings and a classification result of classifying the texture data of the buildings into classes; a classification model generation unit that generates the classification model by machine learning using a plurality of the second learning data stored in the second learning data storage unit as second teacher data; a third learning data storage unit that stores third learning data including texture data of buildings classified into classes and a reflection coefficient of a radio wave corresponding to the class of the texture data of the buildings; and a classification model generation unit that generates the reflection coefficient estimation model by machine learning using a plurality of the third learning data stored in the third learning data storage unit as third teacher data. the classification model acquisition unit may acquire the classification model generated by the classification model generation unit, the classification unit may classify the building texture data into classes using the classification model generated by the classification model generation unit, the estimation model acquisition unit may acquire the reflection coefficient estimation model generated by the estimation model generation unit, and the estimation unit may estimate a reflection coefficient of radio waves corresponding to the class of the building texture data from the building texture data classified into classes by the classification unit, using the reflection coefficient estimation model generated by the estimation model generation unit.

[0013] According to one embodiment of the present invention, there is provided a program for causing a computer to function as any one of the above systems.

[0014] The above summary of the invention does not list all of the necessary features of the present invention. Also, subcombinations of these features may also be inventions. [Brief description of the drawings]

[0015] [Figure 1] An example of a system 10 is shown diagrammatically. [Diagram 2] FIG. 2 is an explanatory diagram for explaining an example of generating a 3D model of a building 50. [Diagram 3]11 is an explanatory diagram for explaining an example of associating material data with a 3D model of a building 50. FIG. [Figure 4] 2 illustrates an example of a functional configuration of an information processing device 100. [Diagram 5] 2 is an explanatory diagram illustrating an example of a processing flow of the information processing device 100. FIG. [Figure 6] 2 illustrates an example of a functional configuration of a classification model generating device 200. [Figure 7] 2 illustrates an example of a functional configuration of a classification device 300. [Figure 8] 2 shows an example of a functional configuration of an estimation model generating device 400. [Figure 9] 5 illustrates an example of a functional configuration of an estimation device 500. [Figure 10] 6 shows an example of a functional configuration of a map data generating device 600. [Figure 11] 7 shows an example of a functional configuration of a simulation device 700. [Figure 12] FIG. 2 is an explanatory diagram for explaining an example of a processing flow of the system 10. [Figure 13] An example of a hardware configuration of a computer 1200 functioning as an information processing device 100, a classification model generating device 200, a classification device 300, an estimation model generating device 400, an estimation device 500, a map data generating device 600, or a simulation device 700 is shown roughly. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] In existing radio wave propagation simulations, area simulations on the scale of city blocks or cities are performed using 3D models with LOD (Level of Details) 1 that represent buildings in three dimensions. However, 3D models with LOD 1 represent buildings with uniform materials. Therefore, one of the reasons for the occurrence of areas in which there is a large difference between the simulation results of the area simulation and the measurement results of actually measuring radio waves in the target area of ​​the area simulation is that the material of the 3D model of the building is different from the material of the reflection points of the building that actually reflect radio waves. On the other hand, in the current market, there are radio wave propagation simulation models that can perform high-resolution area simulations using 3D models with LOD 4 that can represent buildings in detail, including not only the external structure and materials but also the internal structure and materials. However, creating a 3D model with LOD 4 on the scale of city blocks or cities is not realistic in terms of cost. Therefore, it is desirable to realize area simulations on the scale of city blocks or cities using high-resolution radio wave propagation simulations. The system according to the present embodiment employs a mechanism capable of generating a 3D model on a block or city scale, in which the exterior of a building is represented at a level of detail equivalent to the exterior of the building represented by a 3D model of LOD4, based on a 3D model of LOD2 that can represent the shape of the building and includes texture data of the building. Specifically, the system according to the present embodiment first estimates the material of the exterior of the building by performing machine learning-based image analysis on the texture data of the 3D model of LOD2. Then, the system according to the present embodiment generates a 3D model in which the exterior of the building is represented at a level of detail equivalent to the exterior of the building represented by a 3D model of LOD4, based on material data indicating the estimated material of the exterior of the building.

[0017] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the invention according to the claims. In addition, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0018] 1 illustrates an example of a system 10. The system 10 may include a classification model generating device 200. The system 10 may include a classification device 300. The system 10 may include an estimation model generating device 400. The system 10 may include an estimation device 500. The system 10 may include a map data generating device 600. The system 10 may include a simulation device 700.

[0019] The classification model generating device 200 generates a classification model for classifying texture data into classes. The texture data is data that expresses the texture and color of an object. The texture data is, for example, image data.

[0020] The object is, for example, a building. The building may be any object that is fixed to land and has a roof and surrounding walls. The building is, for example, a building. The building is, for example, a house. The building is, for example, a store. The building is, for example, a school. The building may be a warehouse. The object is, for example, a structure such as a road, a railway, a road sign, a bridge, etc. The object may be a natural object such as a branch, a stone, etc.

[0021] For example, if the object is a building, the classes of the building's texture data include classes such as a window class, an entrance class, an exterior wall class, a signboard class, a front door class, a roof class, etc. For example, if the object is a structure, the classes of the structure's texture data include a roadway class, a sidewalk class, a road marking class, a road sign class, a rail class, a sleeper class, a ballast class, etc. For example, if the object is a natural object, the classes of the natural object's texture data include a foliage class, a stone class, etc.

[0022] The classification model generating device 200 generates a classification model by machine learning, for example. The classification model generating device 200 generates a classification model by machine learning, for example, by using a plurality of learning data including texture data of an object and a classification result of classifying the texture data of the object into classes as teacher data.

[0023] The classification model is, for example, an image segmentation model. The classification model is, for example, a semantic segmentation model. The classification model may be any other model capable of classifying texture data into classes.

[0024] The classification model generation device 200 transmits, for example, the generated classification model. The classification model generation device 200 transmits, for example, the generated classification model via the network 20.

[0025] The network 20 includes, for example, the Internet. The network 20 may include a core network provided by a telecommunications carrier. The core network may be compliant with any mobile communication system, such as a 3G (3rd Generation) communication system, an LTE (Long Term Evolution) communication system, a 5G (5th Generation) communication system, and a 6G (6th Generation) communication system or later. The network 20 may include a wired communication network.

[0026] For example, the classification model generation device 200 transmits the generated classification model to the classification device 300. The classification model generation device 200 may transmit the generated classification model to any other external device.

[0027] The classification device 300 classifies the texture data of objects into classes. For example, the classification device 300 classifies the texture data of buildings into classes.

[0028] The classification device 300 classifies the texture data of an object into classes using, for example, a classification model. The classification device 300 receives the classification model from the classification model generation device 200 via the network 20, and classifies the texture data of the object into classes using the received classification model.

[0029] The classification device 300, for example, acquires texture data of the object to be classified. The classification device 300 acquires three-dimensional map data, and acquires texture data of the object to be classified included in the acquired three-dimensional map data, thereby acquiring the texture data of the object to be classified.

[0030] For example, if a 3D model includes texture data of an object that is a model subject, the LOD of the 3D model is LOD 2. Three-dimensional map data including a 3D model of LOD 2 may be open data.

[0031] The area corresponding to the three-dimensional map data is, for example, an urban area. The area corresponding to the three-dimensional map data is, for example, a city block area. The area corresponding to the three-dimensional map data is, for example, an urban area. The area corresponding to the three-dimensional map data is, for example, an office district area. The area corresponding to the three-dimensional map data is, for example, a residential area. The area corresponding to the three-dimensional map data is, for example, a suburban area. The area corresponding to the three-dimensional map data may be any other area. Note that the area corresponding to the three-dimensional map data may be an area covered by the three-dimensional map data.

[0032] The classification device 300 transmits the classification result obtained by classifying the texture data of the object via the network 20, for example.

[0033] The classification device 300, for example, transmits the classification result to the estimation device 500. The classification device 300 may transmit the classification result to any other external device.

[0034] The classification result includes, for example, texture data of the object classified into the class, and position data indicating the position of the 3D model of the object in the three-dimensional coordinate system of the three-dimensional map data.

[0035] The estimation model generating device 400 generates an estimation model that estimates information about texture data of an object classified into a class. For example, the estimation model generating device 400 generates a material estimation model that estimates a material corresponding to the class of the texture data of an object classified into a class from the texture data of the object.

[0036] The estimation model generating device 400 generates a material estimation model by machine learning, for example. The estimation model generating device 400 generates a material estimation model by machine learning, for example, by using as learning data including texture data of objects classified into classes and material data indicating materials corresponding to the classes of the texture data of the objects.

[0037] The material estimated by the material estimation model is, for example, concrete. The concrete includes, for example, dry concrete. The concrete includes, for example, wet concrete. The material estimated by the material estimation model is, for example, glass. The material estimated by the material estimation model is, for example, brick. The material estimated by the material estimation model is, for example, plasterboard. The material estimated by the material estimation model is, for example, wood. The material estimated by the material estimation model is, for example, metal. The material estimated by the material estimation model is, for example, branches and leaves. The material estimated by the material estimation model may be any one or more other materials.

[0038] The estimation model generating device 400 transmits the generated estimation model. The estimation model generating device 400 transmits the generated estimation model via the network 20, for example.

[0039] The estimation model generating device 400, for example, transmits the generated estimation model to the estimation device 500. The estimation model generating device 400 may transmit the generated estimation model to any other external device.

[0040] The estimation device 500 estimates information about the texture data of an object classified into a class. The estimation device 500 receives a classification result from the classification device 300 via the network 20, for example, and estimates information about the texture data of an object classified into a class indicated by the received classification result.

[0041] The estimation device 500 estimates information about the texture data of an object classified into a class, for example, by using an estimation model. The estimation device 500 receives the estimation model from the estimation model generation device 400 via the network 20, for example, and estimates information about the texture data of an object classified into a class, by using the received estimation model.

[0042] The estimation device 500 estimates a material corresponding to the class of texture data of an object classified into classes from the texture data of the object using, for example, a material estimation model.The estimation device 500 estimates a material corresponding to the class of texture data of a building classified into classes from the texture data of the building using, for example, a material estimation model.

[0043] The estimation device 500 transmits an estimation result obtained by estimating information about texture data of an object classified into a class. The estimation device 500 transmits the estimation result via the network 20, for example.

[0044] The estimation result includes, for example, texture data of an object classified into a class. The estimation result includes, for example, estimated material information indicating an estimated material corresponding to the class of the texture data of the object. The estimation result includes, for example, position data of a 3D model of the object.

[0045] The estimation device 500, for example, transmits the estimation result to the map data generating device 600. The estimation device 500 may transmit the estimation result to any other external device.

[0046] The map data generating device 600 generates three-dimensional map data including a 3D model. For example, the map data generating device 600 generates three-dimensional map data including a 3D model that takes into account the material of the outer surface of the model object. For example, the map data generating device 600 generates three-dimensional map data including a 3D model in which the outer surface is expressed at a level of detail equivalent to the level of detail of the outer surface expressed in a 3D model of LOD4. Note that the outer surface may include the front, back, side, and top surfaces.

[0047] The map data generation device 600 generates three-dimensional map data including a 3D model of an object based on an estimation result obtained by estimating information about texture data of the object classified into a class. The map data generation device 600 receives the estimation result from the estimation device 500 via the network 20, for example.

[0048] For example, based on the estimation result, the map data generation device 600 generates three-dimensional map data including a 3D model of an object that takes into account the material of the exterior surface of the object by associating material data indicating a material corresponding to a class of the texture data of the object with the 3D model of the object. For example, based on the estimation result, the map data generation device 600 generates three-dimensional map data including a 3D model of a building that takes into account the material of the exterior surface of the building by associating material data corresponding to a class of the texture data of the building with the 3D model of the building.

[0049] The map data generating device 600 transmits the generated three-dimensional map data. The map data generating device 600 transmits the generated three-dimensional map data via the network 20, for example.

[0050] For example, the map data generation device 600 transmits the generated three-dimensional map data to the simulation device 700. The map data generation device 600 may transmit the generated three-dimensional map data to any other external device.

[0051] The simulation device 700 executes various simulations using the three-dimensional map data. The simulation device 700 executes various simulations using the three-dimensional map data received from the map data generating device 600 via the network 20, for example.

[0052] The simulation device 700, for example, executes a simulation of radio wave propagation in an area corresponding to the three-dimensional map data. The simulation device 700, for example, executes a simulation of a damage situation in an area corresponding to the three-dimensional map data when a disaster occurs in the area. The simulation device 700 may also execute a simulation of the amount of solar radiation irradiated into a building located in the area corresponding to the three-dimensional map data.

[0053] At least two of the classification model generation device 200, the classification device 300, the estimation model generation device 400, the estimation device 500, the map data generation device 600, and the simulation device 700 may be configured in a single device. For example, the classification model generation device 200 and the classification device 300 are configured in a single device. For example, the estimation model generation device 400 and the estimation device 500 are configured in a single device. The classification model generation device 200 and the estimation model generation device 400 may be configured in a single device.

[0054] Currently, in order to generate a 3D model of a building with LOD4, it is usually necessary to generate it manually from a blueprint of the building. Therefore, when generating three-dimensional map data using a 3D model with LOD4, the burden of costs such as time cost, monetary cost, and human cost becomes enormous as the range of the area corresponding to the three-dimensional map data becomes wider. In particular, when the area corresponding to the three-dimensional map data is an urban area or a block area, buildings are densely packed and there are many buildings with complex structures, so the cost of generating three-dimensional map data per unit area is extremely high compared to other areas. On the other hand, in the current market, there is a high-resolution radio wave propagation simulation model using a 3D model with LOD4. In the case of outdoor radio wave propagation simulation, the radio wave propagation simulation model can perform radio wave propagation simulation by using three-dimensional map data including a 3D model that takes into account the material of the outer surface of the model object. This is because the reflection coefficient of the radio wave of the outer surface of the model object can be derived from the material of the outer surface of the model object. Therefore, if three-dimensional map data that can cover a wide area can be generated at low cost using a 3D model that takes into account the exterior material of the model object, a high-resolution radio wave propagation simulation can be realized for a wide outdoor area. In particular, in urban areas and city block areas, there are many communication terminals that use radio waves and the radio wave usage conditions are complex, so there is a high demand for high-resolution radio wave propagation simulation. For the above reasons, it is desirable to be able to generate three-dimensional map data that includes a 3D model that takes into account the exterior material of the model object and can cover a wide area at low cost.

[0055] In contrast, according to the system 10 of the present embodiment, the classification model is used to classify the texture data of the building into classes, the material estimation model is used to estimate the material corresponding to the class of the texture data of the building, and the material data indicating the material corresponding to the class of the texture data of the building is associated with the 3D model of the building based on the estimation result of the material, thereby generating three-dimensional map data including the 3D model of the building. As a result, three-dimensional map data including a 3D model that takes into account the material of the exterior of the model object can be automatically generated, so that the system 10 of the present embodiment can generate three-dimensional map data that includes a 3D model that takes into account the material of the exterior of the model object and can cover a wide area at low cost. Furthermore, by performing a simulation of radio wave propagation outdoors using three-dimensional map data including a 3D model that takes into account the material of the exterior of the model object, the system 10 of the present embodiment can realize a high-resolution simulation of radio wave propagation targeting a wide area outdoors, particularly an urban area or a block area.

[0056] 2 is an explanatory diagram for explaining an example of generating a 3D model of a building 50. Here, an example of generating a 3D model of a building 50 taking into account the material of the exterior surface of the building 50 from the 3D model of the building 50 of LOD2 will be mainly described.

[0057] 2 shows a schematic diagram of an example of a 3D model of a building 50 at LOD 2. The building 50 includes a plurality of windows including a window 52, ​​a doorway 54, and an exterior wall.

[0058] The classification device 300 classifies the texture data of the building 50 included in the 3D model of the building 50 at LOD2 into classes using, for example, the classification model generated by the classification model generation device 200. Here, the explanation will be continued assuming that the classification device 300 has classified the texture data of the building 50 into a "window" class, an "entrance" class, and an "exterior wall" class.

[0059] The diagram shown in the middle part of Fig. 2 roughly illustrates an example of a 3D model of a building 50 of LOD2 in which texture data has been classified into classes. The 3D model of the building 50 of LOD2 in which texture data has been classified into classes shown in the middle part of Fig. 2 is drawn with dotted lines in order to emphasize the difference from the 3D model of the building 50 of LOD2 shown in the upper part of Fig. 2.

[0060] The estimation device 500 estimates a material corresponding to a class of the texture data of the building 50 from the texture data of the building 50 classified into classes by the classification device 300, using, for example, a material estimation model generated by the estimation model generation device 400. Here, the explanation will be continued assuming that the estimation device 500 estimates that the material corresponding to the class of "window" and the class of "entrance" is "glass", and that the material corresponding to the class of "exterior wall" is "concrete".

[0061] For example, the map data generation device 600 generates a 3D model of the building 50 taking into account the material of the exterior surface of the building 50 by associating material data indicating a material corresponding to the class of the texture data of the building 50 with the 3D model of the building 50 based on the estimation result by the estimation device 500. Here, the explanation will be continued assuming that the map data generation device 600 generates a 3D model of the building 50 taking into account the material of the exterior surface of the building 50 by processing the 3D model of the building 50 so that areas corresponding to the "window" class and the "entrance" class are represented using material data indicating "glass" and areas corresponding to the "exterior wall" class are represented using material data indicating "concrete".

[0062] The diagram shown in the lower part of Fig. 2 shows an example of a 3D model of building 50 taking into account the material of the exterior of building 50. In the example of the 3D model of building 50 taking into account the material of the exterior of building 50 shown in the lower part of Fig. 2, an area corresponding to a plurality of "window" classes corresponding to a plurality of windows including window 52 and an area corresponding to an "entrance" class corresponding to entrance 54 are represented using material data indicating "glass", and an area corresponding to an "exterior wall" class corresponding to the exterior wall of building 50 is represented using material data indicating "concrete".

[0063] 3 is an explanatory diagram for explaining an example of associating material data with a 3D model of a building 50. Here, an example of associating material data with a 3D model of a building 50 of LOD2 will be mainly described.

[0064] The diagram shown in the upper part of Fig. 3 shows an example of a 3D model of a building 50 at LOD2, in which texture data is classified into classes. The diagram shown in the upper part of Fig. 3 is the same as the diagram shown in the middle part of Fig. 2, except for the difference between dotted lines and solid lines.

[0065] The estimation device 500 estimates a material corresponding to a class of the texture data of the building 50 from the texture data of the building 50 classified into classes by the classification device 300, using, for example, a material estimation model generated by the estimation model generation device 400. Here, the explanation will be continued assuming that the estimation device 500 estimates that the material corresponding to the class of "window" and the class of "entrance" is "glass", and that the material corresponding to the class of "exterior wall" is "concrete".

[0066] For example, the map data generation device 600 divides the 3D model of the building 50 into areas corresponding to classes estimated to be the same material, based on the estimation result by the estimation device 500. Here, the explanation will be continued assuming that the map data generation device 600 divides the 3D model of the building 50 into areas corresponding to the "window" class and the "entrance" class estimated to be "glass", and an area corresponding to the "exterior wall" class estimated to be "concrete".

[0067] The diagram shown in the middle part of Fig. 3 shows an example of a 3D model of a building 50 divided into areas corresponding to classes estimated to be the same material. For example, the map data generation device 600 assigns material data corresponding to each divided area to the divided area. Here, the explanation will be continued assuming that the map data generation device 600 assigns material data indicating "glass" to areas corresponding to the "window" class and the "entrance" class, and assigns material data indicating "concrete" to an area corresponding to the "exterior wall" class.

[0068] The diagram shown in the lower part of Fig. 3 shows an example of a 3D model of the building 50 in which material data is assigned to the divided regions. Thereafter, the map data generation device 600 may generate a 3D model of the building 50 that takes into account the material of the exterior surface of the building 50 shown in the lower part of Fig. 2 by integrating the regions corresponding to the "window" class and the "entrance" class to which material data indicating "glass" is assigned and the region corresponding to the "exterior wall" class to which material data indicating "concrete" is assigned.

[0069] 4 illustrates an example of a functional configuration of the information processing device 100. The information processing device 100 processes various types of information. The system 10 may further include the information processing device 100.

[0070] The information processing device 100 includes a learning data acquisition unit 102, a learning data storage unit 104, a preprocessing unit 105, a classification model generation unit 106, an estimation model generation unit 108, a texture data acquisition unit 110, a texture data storage unit 111, a classification model acquisition unit 112, a classification unit 114, an estimation model acquisition unit 116, an estimation unit 118, a map data generation unit 120, a reflection coefficient data storage unit 122, and a simulation unit 124. It is not essential that the information processing device 100 includes all of these components.

[0071] The learning data acquiring unit 102 acquires various types of learning data. For example, the learning data acquiring unit 102 acquires various types of learning data by receiving various types of learning data from an external device via the network 20. The learning data acquiring unit 102 may acquire various types of learning data by accepting input from a user of the information processing device 100 via an input unit included in the information processing device 100. The learning data acquiring unit 102 may store the acquired various types of learning data in the learning data storage unit 104.

[0072] The learning data acquiring unit 102 acquires, for example, various types of learning data that have been pre-processed before a model is generated by machine learning. The learning data acquiring unit 102 may acquire various types of learning data that have not been pre-processed.

[0073] The learning data acquisition unit 102 acquires, for example, learning data for generating a classification model. The learning data for generating a classification model includes, for example, texture data of an object and a classification result in which the texture data of the object is classified into classes. The learning data for generating a classification model includes, for example, texture data of a building and a classification result in which the texture data of the building is classified into classes. The learning data for generating a classification model includes, for example, texture data of a structure and a classification result in which the texture data of the structure is classified into classes. The learning data for generating a classification model may include texture data of a natural object and a classification result in which the texture data of the natural object is classified into classes. The learning data for generating a classification model may be referred to as second learning data.

[0074] The learning data acquisition unit 102 acquires, for example, learning data for generating a material estimation model. The learning data for generating a material estimation model includes, for example, texture data of an object classified into a class, and material data indicating a material corresponding to the class of the texture data of the object. The learning data for generating a material estimation model includes, for example, texture data of a building classified into a class, and material data indicating a material corresponding to the class of the texture data of the building. The learning data for generating a material estimation model includes, for example, texture data of a structure classified into a class, and material data indicating a material corresponding to the class of the texture data of the structure. The learning data for generating a material estimation model includes, for example, texture data of a natural object classified into a class, and material data indicating a material corresponding to the class of the texture data of the natural object. The learning data for generating a material estimation model may be referred to as first learning data.

[0075] The learning data acquisition unit 102 may acquire a reflection coefficient estimation model that estimates a reflection coefficient of radio waves corresponding to the class of the texture data of an object classified into a class from the texture data of the object. The learning data for generating the reflection coefficient estimation model includes, for example, texture data of an object classified into a class and a reflection coefficient of radio waves corresponding to the class of the texture data of the object. The learning data for generating the reflection coefficient estimation model includes, for example, texture data of a building classified into a class and a reflection coefficient of radio waves corresponding to the class of the texture data of the building. The learning data for generating the reflection coefficient estimation model includes, for example, texture data of a structure classified into a class and a reflection coefficient of radio waves corresponding to the class of the texture data of the structure. The learning data for generating the reflection coefficient estimation model may include texture data of a natural object classified into a class and a reflection coefficient of radio waves corresponding to the class of the texture data of the natural object. The learning data for generating the reflection coefficient estimation model may be referred to as third learning data.

[0076] The learning data storage unit 104 stores various types of learning data, for example, various types of learning data acquired by the learning data acquisition unit 102.

[0077] The learning data storage unit 104 stores, for example, first learning data. The learning data storage unit 104 may be an example of a first learning data storage unit.

[0078] The learning data storage unit 104 stores, for example, second learning data. The learning data storage unit 104 may be an example of a second learning data storage unit.

[0079] The learning data storage unit 104 may store third learning data. The learning data storage unit 104 may be an example of a third learning data storage unit.

[0080] The preprocessing unit 105 performs preprocessing on various learning data stored in the learning data storage unit 104. The preprocessing unit 105 performs preprocessing on, for example, first learning data. The preprocessing unit 105 performs preprocessing on, for example, second learning data. The preprocessing unit 105 may perform preprocessing on third learning data.

[0081] The preprocessing unit 105 performs preprocessing on various types of learning data stored in the learning data storage unit 104, for example, by performing data augmentation processing on the various types of learning data stored in the learning data storage unit 104. The data augmentation processing is processing for increasing the amount of data in the learning data by performing inversion processing, rotation processing, translation processing, brightness change processing, enlargement processing, reduction processing, synthesis processing, shearing processing, etc. on image data included in the learning data. The preprocessing unit 105 may store the various types of learning data after the preprocessing in the learning data storage unit 104.

[0082] The classification model generation unit 106 generates a classification model that classifies texture data into classes. The classification model generation unit 106 generates the classification model by machine learning, for example, using a plurality of second training data stored in the training data storage unit 104 as training data. Note that the training data for generating the classification model may be referred to as second training data.

[0083] The classification model generation unit 106 generates, for example, a classification model that classifies texture data of an object into classes by machine learning. The classification model generation unit 106 generates, for example, a classification model that classifies texture data of a building into classes by machine learning. The classification model generation unit 106 generates, for example, a classification model that classifies texture data of a structure into classes by machine learning. The classification model generation unit 106 may generate, for example, a classification model that classifies texture data of a natural object into classes by machine learning.

[0084] The estimation model generation unit 108 generates an estimation model. For example, the estimation model generation unit 108 uses a plurality of pieces of training data stored in the training data storage unit 104 as teacher data to generate the estimation model through machine learning.

[0085] The estimation model generation unit 108 generates a material estimation model by machine learning using, for example, a plurality of first training data stored in the training data storage unit 104 as training data. Note that the training data for generating the material estimation model may be referred to as first training data.

[0086] For example, the estimation model generation unit 108 generates, by machine learning, a material estimation model that estimates a material corresponding to the class of texture data of an object classified into a class from the texture data of the object. For example, the estimation model generation unit 108 generates, by machine learning, a material estimation model that estimates a material corresponding to the class of texture data of a building classified into a class from the texture data of the building. For example, the estimation model generation unit 108 generates, by machine learning, a material estimation model that estimates a material corresponding to the class of texture data of a structure classified into a class from the texture data of the structure. The estimation model generation unit 108 may generate, by machine learning, a material estimation model that estimates a material corresponding to the class of texture data of a natural object from the texture data of the natural object classified into a class.

[0087] The estimation model generation unit 108 may generate the reflection coefficient estimation model by machine learning, using the multiple pieces of third training data stored in the training data storage unit 104 as training data. Note that the training data for generating the reflection coefficient estimation model may be referred to as third training data.

[0088] The estimation model generation unit 108 generates, for example, a reflection coefficient estimation model by machine learning, which estimates a reflection coefficient of radio waves corresponding to a class of texture data of an object from texture data of the object classified into classes. The estimation model generation unit 108 generates, for example, a reflection coefficient estimation model by machine learning, which estimates a reflection coefficient of radio waves corresponding to a class of texture data of a building from texture data of the building classified into classes. The estimation model generation unit 108 generates, for example, a reflection coefficient estimation model by machine learning, which estimates a reflection coefficient of radio waves corresponding to a class of texture data of a structure from texture data of the structure classified into classes. The estimation model generation unit 108 may generate, for example, a reflection coefficient estimation model by machine learning, which estimates a reflection coefficient of radio waves corresponding to a class of texture data of a natural object from texture data of the natural object classified into classes.

[0089] The texture data acquisition unit 110 acquires various texture data. For example, the texture data acquisition unit 110 acquires various texture data by receiving various texture data from an external device via the network 20. The texture data acquisition unit 110 may acquire the various texture data by accepting input from a user of the information processing device 100 via an input unit included in the information processing device 100. The texture data acquisition unit 110 may store the acquired various texture data in the texture data storage unit 111.

[0090] The texture data acquisition unit 110 acquires, for example, texture data of an object. The texture data acquisition unit 110 acquires, for example, texture data of a building. The texture data acquisition unit 110 acquires, for example, texture data of a structure. The texture data acquisition unit 110 may acquire texture data of a natural object.

[0091] The texture data acquisition unit 110 acquires, for example, a 3D model of LOD2. The texture data acquisition unit 110 may acquire three-dimensional map data including the 3D model of LOD2.

[0092] The classification model acquisition unit 112 acquires a classification model. For example, the classification model acquisition unit 112 acquires a classification model generated by the classification model generation unit 106. The classification model acquisition unit 112 may acquire a classification model generated by the classification model generation device 200. In this case, the classification model acquisition unit 112 may receive the classification model from the classification model generation device 200 via the network 20.

[0093] The classification unit 114 classifies the texture data into classes. The classification unit 114 classifies the texture data into classes using, for example, the classification model acquired by the classification model acquisition unit 112.

[0094] For example, the classification unit 114 classifies texture data of objects into classes. For example, the classification unit 114 classifies texture data of buildings into classes. For example, the classification unit 114 classifies texture data of structures into classes. The classification unit 114 may classify texture data of natural objects into classes.

[0095] The estimation model acquisition unit 116 acquires an estimation model. For example, the estimation model acquisition unit 116 acquires an estimation model generated by the estimation model generation unit 108. The estimation model acquisition unit 116 may acquire an estimation model generated by the estimation model generation device 400. In this case, the estimation model acquisition unit 116 may receive the estimation model from the estimation model generation device 400 via the network 20.

[0096] The estimation model acquisition unit 116 acquires, for example, a material estimation model. The estimation model acquisition unit 116 may acquire a reflection coefficient estimation model.

[0097] The estimation unit 118 estimates information about the texture data of the object classified into the class. The estimation unit 118 estimates information about the texture data of the object classified into the class, for example, by using the estimation model acquired by the estimation model acquisition unit 116.

[0098] For example, the estimation unit 118 estimates information about texture data of an object classified into a class indicated by the classification result by the classification unit 114. The estimation unit 118 may estimate information about texture data of an object classified into a class indicated by the classification result by the classification device 300. In this case, the estimation unit 118 may receive the classification result from the classification device 300 via the network 20.

[0099] The estimation unit 118 estimates a material corresponding to the class of texture data of an object classified into a class from the texture data of the object, for example, using a material estimation model. The estimation unit 118 estimates a material corresponding to the class of texture data of a building, for example, using a material estimation model, from the texture data of the building, for example, classified into a class. The estimation unit 118 estimates a material corresponding to the class of texture data of a structure, for example, using a material estimation model, from the texture data of the structure, for example, classified into a class. The estimation unit 118 may estimate a material corresponding to the class of texture data of a natural object, from the texture data of the natural object, for example, using a material estimation model.

[0100] The estimation unit 118 estimates whether a material corresponding to the class of the texture data of an object classified into a class is concrete or glass, for example, using a material estimation model. The estimation unit 118 estimates whether a material corresponding to the class of the texture data of an object classified into a class is one of dry concrete, wet concrete, and glass, for example, using a material estimation model, from the texture data of the object classified into a class. The estimation unit 118 estimates whether a material corresponding to the class of the texture data of an object classified into a class is one of dry concrete, wet concrete, glass, brick, gypsum board, wood, metal, and foliage, for example, using a material estimation model, from the texture data of the object classified into a class. Note that the material to be the estimation candidate may be appropriately selected from dry concrete, wet concrete, glass, brick, gypsum board, wood, metal, foliage, and any other one or more materials.

[0101] The estimation unit 118 estimates a reflection coefficient of radio waves corresponding to the class of texture data of an object classified into a class from the texture data of the object, for example, using a reflection coefficient estimation model. The estimation unit 118 estimates a reflection coefficient of radio waves corresponding to the class of texture data of a building classified into a class from the texture data of the building, for example, using a reflection coefficient estimation model. The estimation unit 118 estimates a reflection coefficient of radio waves corresponding to the class of texture data of a structure, for example, using a reflection coefficient estimation model, from the texture data of the structure classified into a class. The estimation unit 118 may estimate a reflection coefficient of radio waves corresponding to the class of texture data of a natural object, from the texture data of the natural object classified into a class, using the reflection coefficient estimation model. Note that when estimating a reflection coefficient of radio waves corresponding to the class of texture data of an object from the texture data of the object classified into a class using the reflection coefficient estimation model, the estimation result may include estimated reflection coefficient information indicating an estimated reflection coefficient of radio waves corresponding to the class of texture data of the object.

[0102] The map data generating unit 120 generates three-dimensional map data including a 3D model. The map data generating unit 120 generates three-dimensional map data including a 3D model that takes into account, for example, the material of the outer surface of the model object. The map data generating unit 120 generates three-dimensional map data including a 3D model that takes into account, for example, the radio wave reflection coefficient of the outer surface of the model object. The map data generating unit 120 generates three-dimensional map data including a 3D model in which the outer surface is expressed with a level of detail equivalent to the level of detail of the outer surface expressed in a 3D model of LOD4.

[0103] The map data generating unit 120 generates three-dimensional map data including, for example, a 3D model of an object. The map data generating unit 120 generates three-dimensional map data including, for example, a 3D model of a building. The map data generating unit 120 generates three-dimensional map data including, for example, a 3D model of a structure. The map data generating unit 120 generates three-dimensional map data including, for example, a 3D model of a natural object.

[0104] For example, the map data generation unit 120 generates three-dimensional map data including a 3D model of the object based on the estimation result by the estimation unit 118. The map data generation unit 120 may generate three-dimensional map data including a 3D model of the object based on the estimation result by the estimation device 500. In this case, the map data generation unit 120 may receive the estimation result from the estimation device 500 via the network 20.

[0105] The map data generating unit 120 generates three-dimensional map data including a 3D model of an object by associating material data indicating a material corresponding to a class of the texture data of the object with the 3D model of the object based on the estimation result, for example. Note that the material corresponding to the class of the texture data of the object may be an estimated material indicated by estimated material information included in the estimation result.

[0106] The map data generating unit 120 associates the material data with the 3D model of the object, for example, by processing the 3D model of the object so that an area of ​​the 3D model of the object corresponding to a class of the texture data of the object is expressed using material data indicating a material corresponding to the class. For example, the map data generating unit 120 divides the 3D model of the object into areas corresponding to classes estimated to be the same material. Next, for each divided area, the map data generating unit 120 assigns the material data corresponding to the divided area to the divided area. In this way, the map data generating unit 120 associates the material data with the 3D model of the object. After that, the map data generating unit 120 may integrate the divided areas.

[0107] The map data generating unit 120 may generate three-dimensional map data including a 3D model of the object by associating the reflection coefficient of the radio wave corresponding to the class of the texture data of the object with the 3D model of the object based on the estimation result. Note that the reflection coefficient of the radio wave corresponding to the class of the texture data of the object may be an estimated reflection coefficient indicated by the estimated reflection coefficient information included in the estimation result.

[0108] The map data generating unit 120 associates the radio wave reflection coefficient with the 3D model of the object, for example, by processing the 3D model of the object so that an area of ​​the 3D model of the object corresponding to a class of the texture data of the object is expressed using a radio wave reflection coefficient corresponding to the class. The map data generating unit 120 associates the radio wave reflection coefficient with the 3D model of the object, for example, by assigning a radio wave reflection coefficient corresponding to the class to an area of ​​the 3D model of the object corresponding to a class of the texture data of the object.

[0109] According to the information processing device 100 of this embodiment, the information processing device 100 generates three-dimensional map data including a 3D model that takes into account the reflection coefficient of radio waves from the outer surface of the model object. As a result, it is possible to automatically generate three-dimensional map data including a 3D model that takes into account the reflection coefficient of radio waves from the outer surface of the model object, so the information processing device 100 of this embodiment can generate three-dimensional map data at low cost that includes a 3D model that takes into account the reflection coefficient of radio waves from the outer surface of the model object, can cover a wide area, and is specialized for use in simulating radio wave propagation.

[0110] The map data generation unit 120 may determine the position of the 3D model of the object in the three-dimensional coordinate system of the three-dimensional map data, based on the three-dimensional map data stored in the texture data storage unit 111. The map data generation unit 120 may determine the position of the 3D model of the object in the three-dimensional coordinate system of the three-dimensional map data, based on position data of the 3D model of the object included in the estimation result.

[0111] The reflection coefficient data storage unit 122 stores reflection coefficient data indicating the reflection coefficient of the radio waves of the material. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of the radio waves of concrete. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of the radio waves of dry concrete. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of the radio waves of wet concrete. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of the radio waves of glass. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of the radio waves of brick. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of the radio waves of plasterboard. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of the radio waves of wood. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of the radio waves of metal. The reflection coefficient data storage unit 122 stores, for example, reflection coefficient data indicating the reflection coefficient of radio waves from branches and leaves. The reflection coefficient data storage unit 122 may store reflection coefficient data indicating the reflection coefficient of radio waves from one or more other arbitrary materials.

[0112] The simulation unit 124 executes various simulations using the three-dimensional map data. The simulation unit 124 executes various simulations using, for example, the three-dimensional map data generated by the map data generation unit 120. The simulation unit 124 may execute various simulations using the three-dimensional map data generated by the map data generation device 600. In this case, the texture data acquisition unit 110 may receive, from the map data generation device 600, three-dimensional map data including a 3D model that takes into account the material of the outer surface of the model object, via the network 20. The texture data acquisition unit 110 may receive, from the map data generation device 600, three-dimensional map data including a 3D model that takes into account the reflection coefficient of radio waves of the outer surface of the model object, via the network 20.

[0113] The simulation unit 124 executes various simulations of areas corresponding to the three-dimensional map data, for example. The simulation unit 124 executes various simulations of urban areas corresponding to the three-dimensional map data, for example. The simulation unit 124 executes various simulations of city block areas corresponding to the three-dimensional map data, for example. The simulation unit 124 executes various simulations of urban areas corresponding to the three-dimensional map data, for example. The simulation unit 124 executes various simulations of business district areas corresponding to the three-dimensional map data, for example. The simulation unit 124 executes various simulations of residential areas corresponding to the three-dimensional map data, for example. The simulation unit 124 executes various simulations of suburban areas corresponding to the three-dimensional map data, for example. The simulation unit 124 may execute various simulations of any other areas corresponding to the three-dimensional map data.

[0114] The simulation unit 124, for example, executes a simulation of radio wave propagation in an area corresponding to the three-dimensional map data. The simulation unit 124 executes a simulation of radio wave propagation in an area corresponding to the three-dimensional map data, for example, based on material data associated with a 3D model of an object included in the three-dimensional map data and reflection coefficient data stored in the reflection coefficient data storage unit 122. The simulation unit 124 determines, for example, from the reflection coefficient data stored in the reflection coefficient data storage unit 122, radio wave reflection coefficient data of a material corresponding to a class of texture data of the object indicated by the material data, and executes a simulation of radio wave propagation in an area corresponding to the three-dimensional map data, based on the reflection coefficient of the radio wave of the material indicated by the determined reflection coefficient data. The simulation unit 124 may execute a simulation of radio wave propagation in an area corresponding to the three-dimensional map data, based on the reflection coefficient of the radio wave associated with a 3D model of an object included in the three-dimensional map data.

[0115] The simulation unit 124 executes a simulation of a damage state of an area corresponding to the three-dimensional map data when a disaster occurs in the area. The simulation unit 124 executes a simulation of a damage state of a building located in the area corresponding to the three-dimensional map data when a disaster occurs in the building.

[0116] The simulation unit 124 executes a simulation of a scattering area into which broken glass will scatter if the glass of the building is broken, based on, for example, the position and size of an area corresponding to the class estimated to be "glass" in the 3D model of the building. The simulation unit 124 executes a simulation of the amount of glass scattering per unit area in the scattering area if the glass of the building is broken, based on, for example, the position and size of an area corresponding to the class estimated to be "glass" in the 3D model of the building.

[0117] The simulation unit 124 may perform a simulation of the amount of solar radiation irradiated inside a building located in an area corresponding to the three-dimensional map data. The simulation unit 124 performs a simulation of the amount of solar radiation irradiated inside the building based on, for example, the position and size of an area corresponding to a class that transmits sunlight, such as a "window" class or an "entrance" class, in the 3D model of the building, and the solar altitude.

[0118] The simulation unit 124 may further perform a simulation of the temperature inside the building based on the simulation result of the amount of solar radiation irradiated inside the building. The simulation unit 124 may further perform a simulation of the temperature distribution inside the building based on the simulation result of the amount of solar radiation irradiated inside the building.

[0119] The information processing device 100 may transmit various types of information. The information processing device 100 transmits various types of information via the network 20, for example.

[0120] For example, the classification model generation unit 106 transmits the generated classification model to the classification device 300. For example, the estimation model generation unit 108 transmits the generated estimation model to the estimation device 500. For example, the classification unit 114 transmits the classification result to the estimation device 500. For example, the estimation unit 118 transmits the estimation result to the map data generation device 600. For example, the estimation unit 118 transmits the estimation result to the simulation device 700. For example, the map data generation unit 120 transmits the generated three-dimensional map data to the simulation device 700.

[0121] 5 is an explanatory diagram for explaining an example of a processing flow of the information processing device 100. Here, an example of a processing flow will be described in which the information processing device 100 generates three-dimensional map data including a 3D model that takes into account the material of the outer surface of a model object from three-dimensional map data including a 3D model of LOD2. Note that the description will be given assuming that the information processing device 100 is in a start state in which it has not yet acquired three-dimensional map data including a 3D model of LOD2.

[0122] In step (sometimes abbreviated to S) 102, the texture data acquisition unit 110 acquires three-dimensional map data including a 3D model of LOD 2. In S104, the classification unit 114 uses the classification model generated by the classification model generation unit 106 to classify the texture data of the objects included in the three-dimensional map data acquired by the texture data acquisition unit 110 in S102 into classes.

[0123] In S106, the estimation unit 118 estimates a material corresponding to the class of the texture data of the object classified into classes by the classification unit 114 in S104, from the texture data of the object, using the material estimation model generated by the estimation model generation unit 108. In S108, the map data generation unit 120 associates material data indicating a material corresponding to the class of the texture data of the object with the 3D model of the object, based on the estimation result by the estimation unit 118 in S108.

[0124] In S110, the map data generating unit 120 judges whether or not the material data has been associated with the 3D models of all objects in the three-dimensional map data including the 3D model of LOD2. If the map data generating unit 120 judges that the material data has not been associated with the 3D models of all objects in the three-dimensional map data including the 3D model of LOD2, the process returns to S102, and the material data is associated with the 3D models of objects to which the material data has not been associated. If the map data generating unit 120 judges that the material data has been associated with the 3D models of all objects in the three-dimensional map data including the 3D model of LOD2, the process of the information processing device 100 for generating three-dimensional map data including a 3D model taking into account the material of the outer surface of the model object from the three-dimensional map data including the 3D model of LOD2 is then terminated.

[0125] 6 illustrates an example of a functional configuration of the classification model generating device 200. The classification model generating device 200 includes a learning data acquiring unit 202, a learning data storage unit 204, a preprocessing unit 205, a classification model generating unit 206, and a transmitting unit 226. Note that it is not essential that the classification model generating device 200 includes all of these components.

[0126] The learning data acquiring unit 202 acquires second learning data. For example, the learning data acquiring unit 202 acquires the second learning data by receiving the second learning data from an external device via the network 20. The learning data acquiring unit 202 may acquire the second learning data by accepting an input from a user of the classification model generation device 200 via an input unit included in the classification model generation device 200. The learning data acquiring unit 202 may store the acquired second learning data in the learning data storage unit 204.

[0127] The learning data storage unit 204 stores the second learning data. The learning data storage unit 204 stores, for example, the second learning data acquired by the learning data acquisition unit 202. The learning data storage unit 204 may be an example of a second learning data storage unit.

[0128] The preprocessing unit 205 performs preprocessing on the second learning data stored in the learning data storage unit 204. The preprocessing unit 205 may store, in the learning data storage unit 204, the second learning data after the preprocessing.

[0129] The preprocessing unit 205 may perform preprocessing on the second learning data in the same manner as the preprocessing unit 105 performs preprocessing on the second learning data. The preprocessing unit 205 may have the same functions as the preprocessing unit 105.

[0130] The classification model generation unit 206 generates a classification model. For example, the classification model generation unit 206 uses a plurality of second learning data stored in the learning data storage unit 204 as second teacher data to generate the classification model by machine learning.

[0131] The classification model generation unit 206 may generate a classification model in the same manner as when the classification model generation unit 106 generates a classification model. The classification model generation unit 206 may have the same functions as the classification model generation unit 106.

[0132] The transmission unit 226 transmits the classification model generated by the classification model generation unit 206. The transmission unit 226 transmits the classification model via the network 20, for example.

[0133] The transmission unit 226 transmits the classification model to, for example, the classification device 300. The transmission unit 226 may transmit the classification model to the information processing device 100.

[0134] 7 illustrates an example of a functional configuration of the classification device 300. The classification device 300 includes a texture data acquisition unit 310, a texture data storage unit 311, a classification model acquisition unit 312, a classification unit 314, and a transmission unit 326. It is not essential that the classification device 300 includes all of these components.

[0135] The texture data acquisition unit 310 acquires various texture data. For example, the texture data acquisition unit 310 acquires various texture data by receiving various texture data from an external device via the network 20. The texture data acquisition unit 310 may acquire the various texture data by accepting an input from a user of the classification device 300 via an input unit included in the classification device 300. The texture data acquisition unit 310 may store the acquired various texture data in the texture data storage unit 311.

[0136] The texture data acquisition section 310 may acquire various texture data similar to the various texture data acquired by the texture data acquisition section 110. The texture data acquisition section 310 may have the same functions as the texture data acquisition section 110.

[0137] The texture data storage unit 311 stores various types of texture data, for example, various types of texture data acquired by the texture data acquisition unit 310.

[0138] The texture data storage unit 311 may store various texture data similar to the various texture data stored in the texture data storage unit 111. The texture data storage unit 311 may have the same functions as the texture data storage unit 111.

[0139] The classification model acquisition unit 312 acquires the classification model. The classification model acquisition unit 312 acquires the classification model by receiving the classification model via the network 20, for example.

[0140] The classification model acquisition unit 312 receives the classification model from, for example, the classification model generation device 200. The classification model acquisition unit 312 may receive the classification model from the information processing device 100.

[0141] The classification unit 314 classifies the texture data into classes. The classification unit 314 classifies the texture data into classes using, for example, the classification model acquired by the classification model acquisition unit 312.

[0142] The classification unit 314 may classify the texture data in the same manner as the classification unit 114 classifies the texture data. The classification unit 314 may have the same functions as the classification unit 114.

[0143] The transmission unit 326 transmits the classification result by the classification unit 314. The transmission unit 326 transmits the classification result via the network 20, for example.

[0144] The transmission unit 326 transmits the classification result to, for example, the estimation device 500. The transmission unit 326 may transmit the classification result to the information processing device 100.

[0145] 8 illustrates an example of a functional configuration of the estimation model generating device 400. The estimation model generating device 400 includes a learning data acquiring unit 402, a learning data storage unit 404, a preprocessing unit 405, an estimation model generating unit 408, and a transmitting unit 426. Note that it is not essential that the estimation model generating device 400 includes all of these components.

[0146] The learning data acquiring unit 402 acquires various types of learning data. For example, the learning data acquiring unit 402 acquires various types of learning data by receiving various types of learning data from an external device via the network 20. The learning data acquiring unit 402 may acquire various types of learning data by accepting input from a user of the estimation model generating device 400 via an input unit included in the estimation model generating device 400. The learning data acquiring unit 402 may store the acquired various types of learning data in the learning data storage unit 404.

[0147] The learning data acquiring unit 402 acquires, for example, first learning data. The learning data acquiring unit 402 may acquire third learning data.

[0148] The learning data storage unit 404 stores various types of learning data. The learning data storage unit 404 stores various types of learning data acquired by the learning data acquisition unit 402, for example.

[0149] The learning data storage unit 404 stores, for example, first learning data. The learning data storage unit 404 may be an example of a first learning data storage unit.

[0150] The learning data storage unit 404 may store the third learning data. The learning data storage unit 404 may be an example of a third learning data storage unit.

[0151] The preprocessing unit 405 performs preprocessing on various types of learning data stored in the learning data storage unit 404. The preprocessing unit 405 performs preprocessing on, for example, the first learning data. The preprocessing unit 405 may perform preprocessing on the third learning data.

[0152] The preprocessing unit 405 may perform preprocessing on various types of learning data stored in the learning data storage unit 404, similar to the case where the preprocessing unit 105 performs preprocessing on various types of learning data. The preprocessing unit 405 may have the same functions as the preprocessing unit 105.

[0153] The estimation model generation unit 408 generates an estimation model. For example, the estimation model generation unit 408 uses a plurality of pieces of training data stored in the training data storage unit 404 as teacher data to generate the estimation model through machine learning.

[0154] The estimation model generation unit 408 generates a material estimation model by machine learning, for example, using a plurality of first learning data stored in the learning data storage unit 404 as first teacher data. The estimation model generation unit 408 may generate a reflection coefficient estimation model by machine learning, using a plurality of third learning data stored in the learning data storage unit 404 as third teacher data.

[0155] The estimation model generation unit 408 may generate an estimation model similar to the estimation model generated by the estimation model generation unit 108. The estimation model generation unit 408 may have the same functions as the estimation model generation unit 108.

[0156] The transmission unit 426 transmits the estimation model generated by the estimation model generation unit 408. The transmission unit 426 transmits the estimation model via, for example, the network 20.

[0157] The transmission unit 426 transmits the estimation model to, for example, the estimation device 500. The transmission unit 426 may also transmit the estimation model to the information processing device 100.

[0158] FIG. 9 schematically shows an example of the functional configuration of the estimation device 500. The estimation device 500 includes an estimation model acquisition unit 516, an estimation unit 518, a transmission unit 526, and a classification result acquisition unit 528. Note that it is not always essential for the estimation device 500 to include all of these configurations.

[0159] The estimation model acquisition unit 516 acquires an estimation model. The estimation model acquisition unit 516 acquires the estimation model by receiving the estimation model via, for example, the network 20.

[0160] The estimation model acquisition unit 516 receives, for example, a classification model from the estimation model generation device 400. The estimation model acquisition unit 516 may also receive the classification model from the information processing device 100.

[0161] The estimation model acquisition unit 516 acquires, for example, a material estimation model. The estimation model acquisition unit 516 may also acquire a reflection coefficient estimation model.

[0162] The classification result acquisition unit 528 acquires a classification result. The classification result acquisition unit 528 acquires the classification result by receiving the classification result via, for example, the network 20.

[0163] The classification result acquisition unit 528 receives the classification result from, for example, the classification device 300. The estimation model acquisition unit 516 may receive the classification result from the information processing device 100.

[0164] The estimation unit 518 estimates information about the texture data of the object classified into the class. The estimation unit 518 estimates information about the texture data of the object classified into the class, which is indicated by the classification result acquired by the classification result acquisition unit 528, using, for example, the estimation model acquired by the estimation model acquisition unit 516.

[0165] The estimation unit 518 may estimate a material corresponding to the class of the texture data of an object classified into classes from the texture data of the object using a material estimation model, for example. The estimation unit 518 may estimate a radio wave reflection coefficient corresponding to the class of the texture data of the object using a reflection coefficient estimation model from the texture data of the object classified into classes.

[0166] The estimation unit 518 estimates information about the texture data of an object classified into a class in the same manner as the estimation unit 118 estimates information about the texture data of an object classified into a class. The estimation unit 518 may have the same function as the estimation unit 118.

[0167] The transmission unit 526 transmits the estimation result by the estimation unit 518. The transmission unit 526 transmits the estimation result via the network 20, for example.

[0168] For example, the transmission unit 526 transmits the estimation result to the map data generation device 600. For example, the transmission unit 526 transmits the estimation result to the simulation device 700. The transmission unit 526 may transmit the estimation result to the information processing device 100.

[0169] 10 shows an example of a schematic functional configuration of a map data generation device 600. The map data generation device 600 includes a texture data acquisition unit 610, a texture data storage unit 611, a map data generation unit 620, a transmission unit 626, and an estimation result acquisition unit 630. It is not essential that the map data generation device 600 includes all of these components.

[0170] The texture data acquisition unit 610 acquires various texture data. For example, the texture data acquisition unit 610 acquires various texture data by receiving various texture data from an external device via the network 20. The texture data acquisition unit 610 may acquire various texture data by accepting input from a user of the map data generation device 600 via an input unit included in the map data generation device 600. The texture data acquisition unit 610 may store the acquired various texture data in the texture data storage unit 611.

[0171] The texture data acquisition unit 610 acquires, for example, a 3D model of LOD2. The texture data acquisition unit 610 may acquire three-dimensional map data including the 3D model of LOD2.

[0172] The texture data acquisition section 610 may acquire various texture data similar to the various texture data acquired by the texture data acquisition section 110. The texture data acquisition section 610 may have the same functions as the texture data acquisition section 110.

[0173] The texture data storage unit 611 stores various types of texture data, for example, various types of texture data acquired by the texture data acquisition unit 610.

[0174] The texture data storage unit 611 may store various texture data similar to the various texture data stored in the texture data storage unit 111. The texture data storage unit 611 may have the same functions as the texture data storage unit 111.

[0175] The estimation result acquisition unit 630 acquires the estimation result. The estimation result acquisition unit 630 acquires the estimation result by receiving the estimation result via the network 20, for example.

[0176] The estimation result acquisition unit 630 receives the estimation result from, for example, the estimation device 500. The estimation result acquisition unit 630 may receive the estimation result from the information processing device 100.

[0177] The map data generating unit 620 generates three-dimensional map data including a 3D model. The map data generating unit 620 generates three-dimensional map data including a 3D model that takes into account, for example, the material of the outer surface of the model object. The map data generating unit 620 generates three-dimensional map data including a 3D model that takes into account, for example, the radio wave reflection coefficient of the outer surface of the model object. The map data generating unit 620 generates three-dimensional map data including a 3D model in which the outer surface is expressed at a level of detail equivalent to the level of detail of the outer surface expressed in the 3D model of LOD4.

[0178] The map data generation unit 620 generates three-dimensional map data including a 3D model of an object, for example, by associating material data indicating a material corresponding to a class of texture data of the object with a 3D model of the object based on the estimation result acquired by the estimation result acquisition unit 630. The map data generation unit 620 may generate three-dimensional map data including a 3D model of an object, for example, by associating a reflection coefficient of a radio wave corresponding to a class of texture data of the object with the 3D model of the object based on the estimation result acquired by the estimation result acquisition unit 630.

[0179] The map data generation unit 620 may determine the position of the 3D model of the object in the three-dimensional coordinate system of the three-dimensional map data, based on the three-dimensional map data stored in the texture data storage unit 611. The map data generation unit 620 may determine the position of the 3D model of the object in the three-dimensional coordinate system of the three-dimensional map data, based on position data of the 3D model of the object included in the estimation result.

[0180] The map data generating unit 620 may generate three-dimensional map data including a 3D model in the same manner as the map data generating unit 120 generates three-dimensional map data including a 3D model. The map data generating unit 620 may have the same functions as the map data generating unit 120.

[0181] The transmission unit 626 transmits the three-dimensional map data generated by the map data generation unit 620. The transmission unit 626 transmits the three-dimensional map data via the network 20, for example.

[0182] The transmission unit 626 transmits, for example, three-dimensional map data to the simulation device 700. The transmission unit 626 may transmit the three-dimensional map data to the information processing device 100.

[0183] 11 shows an example of a schematic functional configuration of a simulation device 700. The simulation device 700 includes a reflection coefficient data storage unit 722, a simulation unit 724, an estimation result acquisition unit 730, and a map data acquisition unit 732. It is not essential that the simulation device 700 includes all of these components.

[0184] The reflection coefficient data storage unit 722 stores various reflection coefficient data. The reflection coefficient data storage unit 722 may store various reflection coefficient data similar to the various reflection coefficient data stored in the reflection coefficient data storage unit 122. The reflection coefficient data storage unit 722 may have the same function as the reflection coefficient data storage unit 122.

[0185] The estimation result acquisition unit 730 acquires the estimation result. The estimation result acquisition unit 730 acquires the estimation result by receiving the estimation result via the network 20, for example.

[0186] The estimation result acquisition unit 730 receives the estimation result from, for example, the estimation device 500. The estimation result acquisition unit 730 may receive the estimation result from the information processing device 100.

[0187] The map data acquisition unit 732 acquires three-dimensional map data. The map data acquisition unit 732 acquires the three-dimensional map data by receiving the three-dimensional map data via the network 20, for example.

[0188] The map data acquisition unit 732 receives, for example, the three-dimensional map data generated by the map data generation unit 620 from the map data generation device 600. The map data acquisition unit 732 may receive the three-dimensional map data generated by the map data generation unit 120 from the information processing device 100.

[0189] The simulation unit 724 executes various simulations using the three-dimensional map data acquired by the map data acquisition unit 732. The simulation unit 724 executes, for example, a simulation of radio wave propagation in an area corresponding to the three-dimensional map data. The simulation unit 724 executes, for example, a simulation of radio wave propagation in an area corresponding to the three-dimensional map data based on material data associated with a 3D model of an object included in the three-dimensional map data and reflection coefficient data stored in the reflection coefficient data storage unit 722. The simulation unit 724 may execute a simulation of radio wave propagation in an area corresponding to the three-dimensional map data based on a reflection coefficient of radio waves associated with a 3D model of an object included in the three-dimensional map data.

[0190] The simulation unit 724 may execute a simulation of a damage situation in an area corresponding to the three-dimensional map data when a disaster occurs in the area. The simulation unit 724 may execute a simulation of the amount of solar radiation irradiated into a building located in the area corresponding to the three-dimensional map data.

[0191] The simulation unit 724 may execute various simulations in the same manner as the simulation unit 124 executes various simulations. The simulation unit 724 may have the same functions as the simulation unit 124.

[0192] 12 is an explanatory diagram for explaining an example of a processing flow of the system 10. Here, an example of a processing flow will be described in which the system 10 generates three-dimensional map data including a 3D model that takes into account the material of the outer surface of the model object from three-dimensional map data including a 3D model of LOD2. Note that the following description will be given assuming that the classification model generation device 200 and the estimation model generation device 400 are in a start state in which they have not yet acquired any learning data.

[0193] In S202, the learning data acquisition unit 202 acquires a plurality of second learning data. In S204, the classification model generation unit 206 generates a classification model by machine learning using the plurality of second learning data acquired by the learning data acquisition unit 202 in S202 as second teacher data.

[0194] In S206, the transmission unit 226 transmits the classification model generated by the classification model generation unit 206 in S204 to the classification device 300 via the network 20. The classification model acquisition unit 312 receives the classification model from the classification model generation device 200 via the network 20.

[0195] In S208, the learning data storage unit 404 acquires a plurality of first learning data. In S210, the estimation model generation unit 408 generates a material estimation model by machine learning, using the plurality of first learning data acquired by the learning data acquisition unit 402 in S208 as first teacher data.

[0196] In S212, the transmission unit 426 transmits the material estimation model generated by the estimation model generation unit 408 in S210 to the estimation device 500 via the network 20. The estimation model acquisition unit 516 receives the material estimation model from the estimation model generation device 400 via the network 20.

[0197] In S214, the texture data acquisition unit 310 acquires texture data of the building. In S216, the classification unit 314 classifies the texture data of the building acquired by the texture data acquisition unit 310 in S214 into classes, using the classification model received by the classification model acquisition unit 312 in S206.

[0198] In S218, the transmission unit 326 transmits the classification result by the classification unit 314 in S216 to the estimation device 500 via the network 20. The classification result acquisition unit 528 receives the classification result from the classification device 300 via the network 20.

[0199] In S220, the estimation unit 518 estimates a material corresponding to the class of the texture data of the building, which is indicated in the classification result received by the classification result acquisition unit 528 in S218 and classified into a class by the classification unit 314 in S216, from the texture data of the building, using the material estimation model acquired by the estimation model acquisition unit 516 in S212. In S222, the transmission unit 526 transmits the estimation result by the estimation unit 518 in S220 to the map data generation device 600 via the network 20. The estimation result acquisition unit 630 receives the estimation result from the estimation device 500 via the network 20.

[0200] In S224, the map data generation unit 620 generates three-dimensional map data including a 3D model of a building by associating material data indicating a material corresponding to the class of the texture data of the building with the 3D model of the building based on the estimation result received by the estimation result acquisition unit 630 in S222. Thereafter, the process of the system 10 that generates three-dimensional map data including a 3D model that takes into account the material of the outer surface of the model target from the three-dimensional map data including the 3D model of LOD2 ends.

[0201] FIG. 13 schematically shows an example of the hardware configuration of a computer 1200 that functions as the information processing apparatus 100, the classification model generation apparatus 200, the classification apparatus 300, the estimation model generation apparatus 400, the estimation apparatus 500, the map data generation apparatus 600, or the simulation apparatus 700. The program installed in the computer 1200 causes the computer 1200 to function as one or more "units" of the apparatus according to the above-described embodiment, or causes the computer 1200 to execute an operation associated with the apparatus according to the above-described embodiment or the one or more "units", and / or causes the computer 1200 to execute the process according to the above-described embodiment or a stage of the process. Such a program may be executed by the CPU 1212 so as to cause the computer 1200 to execute specific operations associated with some or all of the blocks of the flowcharts and block diagrams described herein.

[0202] The computer 1200 according to this embodiment includes a CPU 1212, a RAM 1214, and a graphics controller 1216, which are connected to each other by a host controller 1210. The computer 1200 also includes input / output units such as a communication interface 1222, a storage device 1224, a DVD drive 1226, and an IC card drive, which are connected to the host controller 1210 via an input / output controller 1220. The DVD drive 1226 may be a DVD-ROM drive, a DVD-RAM drive, or the like. The storage device 1224 may be a hard disk drive, a solid state drive, or the like. The computer 1200 also includes legacy input / output units such as a ROM 1230 and a keyboard 1242, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0203] The CPU 1212 operates according to a program stored in the ROM 1230 and the RAM 1214, thereby controlling each unit. The graphic controller 1216 acquires image data generated by the CPU 1212 into a frame buffer or the like provided in the RAM 1214 or into itself, and causes the image data to be displayed on the display device 1218.

[0204] The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD drive 1226 reads programs or data from a DVD-ROM 1227 or the like and provides the programs or data to the storage device 1224. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0205] The ROM 1230 stores therein a boot program or the like that is executed by the computer 1200 upon activation, and / or a program that depends on the hardware of the computer 1200. The input / output chip 1240 may also connect various input / output units to the input / output controller 1220 via a USB port, a parallel port, a serial port, a keyboard port, a mouse port, and the like.

[0206] The programs are provided by a computer-readable storage medium such as a DVD-ROM 1227 or an IC card. The programs are read from the computer-readable storage medium, installed in the storage device 1224, the RAM 1214, or the ROM 1230, which are also examples of computer-readable storage media, and executed by the CPU 1212. Information processing described in these programs is read by the computer 1200, and brings about cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be constructed by implementing operations or processing of information according to the use of the computer 1200.

[0207] For example, when communication is performed between the computer 1200 and an external device, the CPU 1212 may execute a communication program loaded in the RAM 1214 and instruct the communication interface 1222 to perform communication processing based on the processing described in the communication program. Under the control of the CPU 1212, the communication interface 1222 reads transmission data stored in a transmission buffer area provided in the RAM 1214, the storage device 1224, the DVD-ROM 1227, or a recording medium such as an IC card, and transmits the read transmission data to a network, or writes reception data received from the network to a reception buffer area or the like provided on the recording medium.

[0208] Furthermore, the CPU 1212 may cause all or a necessary portion of a file or database stored in an external recording medium such as the storage device 1224, the DVD drive 1226 (DVD-ROM 1227), an IC card, etc. to be read into the RAM 1214, and perform various types of processing on the data on the RAM 1214. The CPU 1212 may then write back the processed data to the external recording medium.

[0209] Various types of information, such as various types of programs, data, tables, and databases, may be stored in the recording medium and undergo information processing. The CPU 1212 may perform various types of processing on the data read from the RAM 1214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequence of the program, and write back the results to the RAM 1214. The CPU 1212 may also search for information in a file, database, etc. in the recording medium. For example, when a plurality of entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored in the recording medium, the CPU 1212 may search for an entry whose attribute value of the first attribute matches a specified condition from among the plurality of entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0210] The above-described programs or software modules may be stored in a computer-readable storage medium on the computer 1200 or in the vicinity of the computer 1200. In addition, a recording medium such as a hard disk or a RAM provided in a server system connected to a dedicated communication network or the Internet can be used as a computer-readable storage medium, thereby providing the programs to the computer 1200 via the network.

[0211] The blocks in the flowcharts and block diagrams in the present embodiment may represent stages of a process in which an operation is performed or "parts" of an apparatus responsible for performing the operation. Particular stages and "parts" may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable storage medium, and / or a processor provided with computer-readable instructions stored on a computer-readable storage medium. The dedicated circuitry may include digital and / or analog hardware circuits, and may include integrated circuits (ICs) and / or discrete circuits. The programmable circuitry may include reconfigurable hardware circuits, such as, for example, field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), and the like, including AND, OR, XOR, NAND, NOR, and other logical operations, flip-flops, registers, and memory elements.

[0212] A computer-readable storage medium may include any tangible device capable of storing instructions that are executed by a suitable device, such that a computer-readable storage medium having instructions stored thereon comprises an article of manufacture that includes instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable storage media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, and the like. More specific examples of computer-readable storage media may include floppy disks, diskettes, hard disks, random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), electrically erasable programmable read-only memories (EEPROMs), static random access memories (SRAMs), compact disk read-only memories (CD-ROMs), digital versatile disks (DVDs), Blu-ray disks, memory sticks, integrated circuit cards, and the like.

[0213] The computer readable instructions may include either assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages ​​such as the “C” programming language or similar programming languages.

[0214] Computer readable instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, or to a programmable circuit, either locally or over a local area network (LAN), a wide area network (WAN), such as the Internet, etc., to cause the processor of the general purpose computer, special purpose computer, or other programmable data processing apparatus, or to a programmable circuit, to execute the computer readable instructions to generate means for performing the operations specified in the flowcharts or block diagrams. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc.

[0215] Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It is clear to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the description of the claims that such modifications and improvements can also be included in the technical scope of the present invention.

[0216] It should be noted that the order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before" or "prior to," and that the process may be performed in any order unless the output of a previous process is used in a later process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, it does not mean that the process must be performed in this order. [Explanation of symbols]

[0217] 10 system, 20 network, 50 building, 52 window, 54 entrance, 100 information processing device, 102 learning data acquisition unit, 104 learning data storage unit, 105 pre-processing unit, 106 classification model generation unit, 108 estimation model generation unit, 110 texture data acquisition unit, 111 texture data storage unit, 112 classification model acquisition unit, 114 classification unit, 116 estimation model acquisition unit, 118 estimation unit, 120 map data generation unit, 122 reflection coefficient data storage unit, 124 simulation unit, 200 classification model generation device, 202 learning data acquisition unit, 204 learning data storage unit, 205 pre-processing unit, 206 classification model generation unit, 226 transmission unit, 300 classification device, 310 texture data acquisition unit, 311 texture data storage unit, 312 classification model acquisition unit, 314 classification unit, 326 Transmitter, 400, estimation model generating device, 402, learning data acquiring unit, 404, learning data storage unit, 405, preprocessing unit, 408, estimation model generating unit, 426, transmitter, 500, estimation device, 516, estimation model acquiring unit, 518, estimation unit, 526, transmitter, 528, classification result acquiring unit, 600, map data generating device, 610, texture data acquiring unit, 611, texture data storage unit, 626, transmitter, 630, estimation result acquiring unit, 700, simulation device, 722, reflection coefficient data storage unit, 724, simulation unit, 730, estimation result acquiring unit, 732, map data acquiring unit, 1200, computer, 1210, host controller, 1212, CPU, 1214, RAM, 1216, graphic controller, 1218, display device, 1220, input / output controller, 1222, communication interface, 1224 storage devices, 1226 DVD drives, 1227 DVD-ROMs, 1230 ROMs, 1240 I / O chips, 1242 keyboards

Claims

1. a classification model acquisition unit that acquires a classification model for classifying texture data of a building into classes; a classification unit that classifies texture data of buildings into classes using the classification model acquired by the classification model acquisition unit; an estimation model acquisition unit that acquires a material estimation model that estimates a material corresponding to a class of the texture data of a building from the texture data of the building classified into classes; an estimation unit that estimates a material corresponding to a class of the texture data of the building from the texture data of the building classified into classes by the classification unit, using the material estimation model acquired by the estimation model acquisition unit; a map data generating unit that generates three-dimensional map data including a 3D model of the building by associating material data indicating a material corresponding to a class of the texture data of the building with the 3D model of the building based on a result of the estimation by the estimation unit; A system comprising:

2. a first learning data storage unit that stores first learning data including texture data of buildings classified into classes and material data indicating a material corresponding to the class of the texture data of the buildings; an estimation model generation unit that generates the material estimation model by machine learning using the plurality of first learning data stored in the first learning data storage unit as first teacher data; Further equipped with the estimation model acquisition unit acquires the material estimation model generated by the estimation model generation unit, the estimation unit estimates a material corresponding to a class of the texture data of the building from the texture data of the building classified into classes by the classification unit, using the material estimation model generated by the estimation model generation unit; The system of claim 1 .

3. a second learning data storage unit that stores second learning data including texture data of buildings and a classification result of classifying the texture data of the buildings into classes; a classification model generation unit that generates the classification model by machine learning using the plurality of second learning data stored in the second learning data storage unit as second teacher data; Further equipped with the classification model acquisition unit acquires the classification model generated by the classification model generation unit; The classification unit classifies the texture data of the building into classes using the classification model generated by the classification model generation unit. The system of claim 1 .

4. a second learning data storage unit that stores second learning data including texture data of buildings and a classification result of classifying the texture data of the buildings into classes; a classification model generation unit that generates the classification model by machine learning using the plurality of second learning data stored in the second learning data storage unit as second teacher data; a first learning data storage unit that stores first learning data including texture data of buildings classified into classes and material data indicating a material corresponding to the class of the texture data of the buildings; an estimation model generation unit that generates the material estimation model by machine learning using the plurality of first learning data stored in the first learning data storage unit as first teacher data; Further equipped with the classification model acquisition unit acquires the classification model generated by the classification model generation unit; The classification unit classifies the texture data of the building into classes using the classification model generated by the classification model generation unit; the estimation model acquisition unit acquires the material estimation model generated by the estimation model generation unit, the estimation unit estimates a material corresponding to a class of the texture data of the building from the texture data of the building classified into classes by the classification unit, using the material estimation model generated by the estimation model generation unit; The system of claim 1 .

5. The system according to claim 1 , wherein the estimation unit estimates whether a material corresponding to a class of the texture data of the building is concrete or glass, based on the texture data of the building classified into classes.

6. a simulation unit that uses the three-dimensional map data generated by the map data generation unit to perform a simulation of radio wave propagation in an area corresponding to the three-dimensional map data. The system of claim 1 , further comprising:

7. a simulation unit that uses the three-dimensional map data generated by the map data generation unit to execute a simulation of a damage situation of the building when a disaster occurs to the building located within an area corresponding to the three-dimensional map data. The system of claim 1 , further comprising:

8. a simulation unit that uses the three-dimensional map data generated by the map data generation unit to execute a simulation of the amount of solar radiation irradiated inside the building located within the area corresponding to the three-dimensional map data. The system of claim 1 , further comprising:

9. a classification model acquisition unit that acquires a classification model for classifying texture data of a building into classes; a classification unit that classifies texture data of buildings into classes using the classification model acquired by the classification model acquisition unit; an estimation model acquisition unit that acquires, from texture data of a building classified into classes, a reflection coefficient estimation model that estimates a reflection coefficient of a radio wave corresponding to the class of the texture data of the building; an estimation unit that estimates a reflection coefficient of a radio wave corresponding to a class of the texture data of a building from the texture data of the building classified into classes by the classification unit, using the reflection coefficient estimation model acquired by the estimation model acquisition unit; a map data generating unit that generates three-dimensional map data including a 3D model of the building by associating a reflection coefficient of a radio wave corresponding to a class of the texture data of the building with the 3D model of the building based on the estimation result by the estimation unit; A system comprising:

10. a second learning data storage unit that stores second learning data including texture data of buildings and a classification result of classifying the texture data of the buildings into classes; a classification model generation unit that generates the classification model by machine learning using the plurality of second learning data stored in the second learning data storage unit as second teacher data; a third learning data storage unit that stores third learning data including texture data of buildings classified into classes and a reflection coefficient of radio waves corresponding to the class of the texture data of the buildings; an estimation model generating unit that generates the reflection coefficient estimation model by machine learning using the third learning data stored in the third learning data storage unit as third teacher data; Further equipped with the classification model acquisition unit acquires the classification model generated by the classification model generation unit; The classification unit classifies the texture data of the building into classes using the classification model generated by the classification model generation unit; the estimation model acquisition unit acquires the reflection coefficient estimation model generated by the estimation model generation unit, the estimation unit estimates a reflection coefficient of a radio wave corresponding to a class of the texture data of the building, from the texture data of the building classified into classes by the classification unit, by using the reflection coefficient estimation model generated by the estimation model generation unit; The system of claim 9.

11. A program for causing a computer to function as the system according to claim 1 or 9.

Images