Learned model, information processing apparatus, information processing method, and information processing program

Abstract

To provide a learned model, an information processing apparatus, an information processing method, and an information processing program which can estimate a building.SOLUTION: A learned model according to the present invention mechanically learned data including a construction site and the feature quantity of the construction site included in the data as teacher data. The feature quantity includes the feature quantity of objects related to the construction site, and the objects are two or more, and at least include one or more selected from an iron plate(s), a construction vehicle(s), a construction heavy machine(s) (including a construction crane), construction materials, a strut(s), a structural stand(s), and a fall prevention net(s).SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a trained model for estimating the age of a building, an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] The age of a building is an important piece of data when it comes to rebuilding, renovating, selling, etc. For this reason, the age of a building has traditionally been estimated. Methods for estimating the age of a building include, for example, the following: (1) Estimate from the difference in map information from two different points in time. (2) Estimated from the difference between satellite images taken at two different times. (3) Estimates are based on the Housing and Land Statistics Survey and the National Political Survey conducted by the Statistics Bureau of the Ministry of Internal Affairs and Communications. (4) Obtained from real estate registration information. However, the above methods each have their own problems, which will be explained below. The method (1) of estimating from differences in map information requires the creation of map information, which requires a great deal of work. Regarding the method of (2) estimating from the difference in satellite images, it is difficult to detect when a building of a similar shape has been rebuilt, and satellite images from two different points in time are required. Patent Document 1 discloses an invention that detects new construction, remodeling, or demolition of buildings by comparing old and new geographic images obtained by photographing the same area at different times using an artificial satellite or the like. (3) The method of estimating based on the Housing and Land Statistics Survey and the National Political Survey conducted by the Statistics Bureau of the Ministry of Internal Affairs and Communications can only estimate housing. (4) The method of obtaining information from real estate registration information requires the effort and expense of obtaining real estate registration information for each property. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-241886 Summary of the Invention [Problem to be solved by the invention]

[0004] As described above, the conventional methods have the problem that estimation requires a great deal of effort and cost.

[0005] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a trained model, an information processing device, an information processing method, and an information processing program capable of estimating buildings. [Means for solving the problem]

[0006] In order to solve the above problems, the trained model of the present invention is characterized by having been machine-learned using data including construction sites and the features of the construction sites contained in the data as training data. [Effects of the Invention]

[0007] According to the present invention, it is possible to provide a trained model, an information processing device, an information processing method, and an information processing program capable of estimating buildings. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a diagram illustrating an example of a schematic configuration of an information processing system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of the configuration and functions of a server according to the embodiment. [Figure 3] FIG. 10 is a diagram showing an example of image data for learning of the information processing system according to the embodiment. [Figure 4] FIG. 2 is a diagram illustrating an example of the configuration and functions of a user terminal according to the embodiment. [Figure 5] 10 is a flowchart illustrating an example of processing of the information processing system according to the embodiment. [Figure 6]10 is a flowchart illustrating an example of processing of the information processing system according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] [Embodiment] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. First, the configuration of an information processing system 1 will be described with reference to Fig. 1. As shown in Fig. 1, the information processing system 1 includes a server 2 (information processing device) and a user terminal 3 connected to the server 2 via a network 4. The information processing system 1 may include any number of servers 2 and user terminals 3. The user terminal 3 may be a desktop PC (Personal Computer), a tablet PC, a smartphone, or the like.

[0010] (Server 2) Fig. 2 is a configuration diagram of the server 2. Note that Fig. 2(a) shows the main hardware configuration of the server 2, and the server 2 has a configuration in which a communication IF 200A, a storage device 200B, and a CPU 200C are connected via a bus or the like. Note that although not shown in Fig. 2(a), the server 2 may also include an input device (for example, a mouse, a keyboard, a touch panel, etc.) and a display device (CRT (Cathode Ray Tube), a liquid crystal display, an organic EL display, etc.).

[0011] The communication IF 200A is an interface for communicating with other devices (for example, the user terminal 3).

[0012] The storage device 200B is, for example, a hard disk drive (HDD) or a semiconductor storage device (solid state drive (SSD)). The storage device 200B stores various data such as trained models and programs. The trained model stored in the storage device 200B is obtained by machine learning using data including a construction site and the features of the construction site contained in the data as training data. The features of the construction site include, for example, features of objects related to the construction site, such as steel plates (primarily steel plates laid (temporarily) at construction sites to ensure access routes for materials, as work floors, to protect the ground, to secure footing for heavy construction equipment on soft ground with poor footing, and as load distribution plates to distribute weight across the ground), construction vehicles, heavy construction equipment (including construction cranes), construction materials, struts, platforms, and fall prevention nets. The storage device 200B also stores the acquisition date of the data including the construction site (which may be the date when the detection unit 204, described later, detected the construction site (hereinafter also referred to as the detection date)) and information that can identify the building to be constructed at the detected construction site, such as the lot number, address, longitude, and latitude, in association with each other. In this embodiment, the acquisition date of the data including the construction site refers to the date when the data including the construction site itself was acquired, rather than the date when the system acquired the data including the construction site, such as the date of photography if the data including the construction site is image data. Furthermore, the information that can identify the building may be used if it is included in the attribute data, or may be registered by the user. Furthermore, the use zone (e.g., Type 1 Exclusive Residential Zone), estimated site area, estimated building area, or estimated total floor area (which may be estimated from an image, may be associated with information from another database or website, or may be entered by the user. In the following, the site area, building area, or total floor area may also be referred to as site area, etc.) in association with each other and stored in the storage device 200B. The zoning may be obtained from open data released by the national or local government. The estimated site area may be calculated by estimating the extent of the construction site using semantic segmentation of deep learning satellite information, or, if the data including the construction site is image data, the site area may be calculated based on the scale of the image and the number of pixels of the construction site. Segmentation, also known as region classification, is a method of labeling each pixel in an image. There are several types of segmentation, and representative methods include semantic segmentation, instance segmentation, and panoptic segmentation. Furthermore, if it is determined that the location is a construction site, objects other than those mentioned above may be included as objects related to the construction site. Furthermore, various data including the construction site can be used, such as aerial photographs, satellite images, and point cloud data acquired by LIDAR (Light Detection and Ranging, Laser Imaging Detection and Ranging).

[0013] FIG. 3 is a diagram showing an example of image data for learning of the information processing system 1 according to the embodiment. Figure 3(a) is an example of an image for learning a construction crane. Figure 3(b) is an example of an image for learning an iron plate. Note that the numerical values ​​in Figure 3 (0.92, 0.75, 0.85, etc.) represent the probability (%) that the detected object is correct (for example, if a construction crane is detected as an object, this is the probability that the detection result is correct (that the object is actually a construction crane)). In addition, multiple types (for example, types of objects such as steel plates, construction vehicles, heavy construction machinery (including construction cranes), building materials, struts, platforms, and fall prevention nets) or multiple objects may be learned using a single image, or one type or one object may be learned using a single image.

[0014] Some or all of the various data stored in the storage device 200B may be stored in an external storage device such as a USB (Universal Serial Bus) memory or an external HDD, or in a storage device of another information processing device connected via the network 4. In this case, the server 2 refers to or acquires the various data stored in the external storage device or the storage device of the other information processing device.

[0015] The CPU 200C controls the server 2 according to this embodiment, and includes a ROM and a RAM (not shown).

[0016] (Server 2 function) Fig. 2(b) is a functional block diagram of the server 2. As shown in Fig. 2(b), the server 2 has functions such as a receiving unit 201, a transmitting unit 202 (output unit), a storage device control unit 203, a detecting unit 204, and an estimating unit 205. The functions shown in Fig. 2(b) are realized by the CPU 200C executing an information processing program stored in the storage device 200B.

[0017] The receiving unit 201 receives data, for example, from the user terminal 3, via the network 4. At this time, attribute data of the data (for example, the acquisition date of data including the construction site (for example, the date of image capture in the case of a captured image) and information that can identify the building to be constructed at the construction site (for example, the lot number, address, longitude and latitude, etc.)) may also be received.

[0018] The transmitting unit 202 (output unit) transmits (outputs) data to, for example, the user terminal 3 via the network 4.

[0019] The storage device control unit 203 controls the storage device 200B and writes and reads data to and from the storage device 200B. The storage device control unit 203 stores, for example, the acquisition date of data including the construction site (which may be the date when the detection unit 204 described later detected the construction site) in the storage device 200B in association with information that can identify the building to be constructed at the detected construction site, such as the lot number, address, longitude and latitude.

[0020] The detection unit 204 inputs the data including the construction site received by the receiving unit 201 into the trained model and detects the construction site from the data including the construction site. Note that the data including the construction site may include multiple construction sites. When the data including the construction site includes multiple construction sites, the detection unit 204 detects each of the included construction sites.

[0021] The estimation unit 205 estimates the age of the building based on the acquisition date of data including the construction site (which may be the date when the detection unit 204 detected the construction site) and the presented date of the age. Specifically, the estimation unit 205 estimates the age W of the building (the number of years since the building was completed (constructed)) based on the following formula (1): W=XYZ (1) X: Presentation date (date at the time of presentation) Y: Date of acquisition of data including the construction site (date of detection is also acceptable) Z: The number of years from the acquisition date of the data including the construction site (or the detection date) to the completion date (1 to several years) That is, the estimation unit 205 estimates that the building will be completed at the construction site one to several years after the acquisition date (or the detection date) of the data including the construction site, and calculates the age W by subtracting one to several years (= Z) from the difference XY between the presentation date X and the acquisition date Y (or the detection date) of the data including the construction site of the construction site. That is, the age of the building estimated by the estimation unit 205 has a range of several years, but the rebuilding cycle of buildings is usually several decades (often more than 50 years), and a few years is within the acceptable range in the scope of use such as rebuilding, renovating, and selling the building.

[0022] (User terminal 3) The user terminal 3 is a terminal used by the first user. Fig. 4 is a configuration diagram of the user terminal 3. Note that Fig. 4(a) shows the main hardware configuration of the user terminal 3, and the user terminal 3 has a configuration in which a communication IF 300A, a storage device 300B, an input device 300C, a display device 300D, and a CPU 300E are connected via a bus or the like.

[0023] The communication IF 300A is an interface for communicating with other devices (for example, the server 2).

[0024] The storage device 300B is, for example, an HDD (Hard Disk Drive) or a semiconductor storage device (SSD (Solid State Drive)). The storage device 300B stores a terminal identifier, an information processing program, etc. The terminal identifier is an identifier for identifying the user terminal 3. By assigning the terminal identifier to data transmitted from the user terminal 3, the server 2 can determine which user terminal 3 transmitted the received data. The terminal identifier may be an IP (Internet Protocol) address, a MAC (Media Access Control) address, etc., or may be assigned to the user terminal 3 by the server 2.

[0025] The input device 300C is, for example, an input device such as a keyboard, a mouse, or a touch panel, but may be any other device or equipment that allows input. It may also be a voice input device.

[0026] The display device 300D is, for example, a liquid crystal display, a plasma display, an organic EL display, or the like, but may be any other device or equipment (for example, a CRT: Cathode Ray Tube) as long as it is capable of displaying.

[0027] The CPU 300E controls the user terminal 3 according to this embodiment, and includes a ROM and a RAM (not shown).

[0028] Fig. 4(b) is a functional block diagram of the user terminal 3. As shown in Fig. 4(b), the user terminal 3 has functions such as a receiving unit 301, a transmitting unit 302, a storage device control unit 303, an input accepting unit 304, and a display device control unit 305. Note that the functions shown in Fig. 4(b) are realized by the CPU 300E executing an information processing program stored in the storage device 300B.

[0029] The receiving unit 301 receives data transmitted from the server 2, for example.

[0030] The transmitting unit 302 transmits, for example, data corresponding to the input operation received by the input receiving unit 304 to the server 2.

[0031] The storage device control unit 303 controls, for example, the storage device 300B to write and read data.

[0032] The input receiving unit 304 receives, for example, an input operation from the input device 300C.

[0033] The display device control unit 305 controls, for example, the display device 300D, and displays the data received by the receiving unit 301 on the display device 300D.

[0034] (Information Processing) Figures 5 and 6 are flowcharts showing an example of information processing of the information processing system 1. Below, the information processing of the information processing system 1 will be described with reference to Figures 5 and 6. In the following description, the same components as those described with reference to Figures 1 to 4 will be assigned the same reference numerals, and duplicated description will be omitted.

[0035] First, the construction site detection process will be described with reference to FIG. (Step S101) The receiving unit 201 of the server 2 receives data including the construction site.

[0036] (Step S102) The detection unit 204 of the server 2 inputs the data including the construction site received by the receiving unit 201 into the trained model and detects the construction site from the data including the construction site.

[0037] (Step S103) The storage device control unit 203 of the server 2 stores the data including the construction site in the storage device 200B in association with the acquisition date of the data (which may be the date the detection unit 204 detected the construction site) and information that can identify the building to be constructed at the detected construction site, such as the lot number, address, longitude and latitude.

[0038] Next, the building age presentation process will be described with reference to FIG. (Step S201) The receiving unit 201 of the server 2 receives from the user terminal 3 information that identifies the building (for example, a lot number, an address, and latitude and longitude).

[0039] (Step S202) The estimation unit 205 of the server 2 references the storage device 200B and estimates the age of the building identified by the information received by the receiving unit 201. Note that the method for estimating the age by the estimation unit 205 has already been described, so a duplicated description will be omitted. Furthermore, if the information of the building identified by the information received by the receiving unit 201 is not stored in the storage device 200B, the estimation unit 205 instructs the transmission unit 202 to transmit a message to that effect (that the age cannot be presented because there is no data) to the user terminal 3, and the transmission unit 202 transmits a message to that effect to the user terminal 3.

[0040] (Step S203) The transmitting unit 202 of the server 2 transmits (outputs) the estimation result (building age) from the estimating unit 205 to the user terminal 3.

[0041] In the above description, the information identifying a building may be information about a specific area. In this case, the ages of buildings included in the specific area and stored in the storage device 200B are presented. When presenting the ages of buildings within the specific area, the ages may be displayed in different colors (for example, in the form of a heat map) according to the ages. The estimated ages may be displayed on the current map in a manner that makes them discernible (such as by overlaying the age numbers or by color). Information such as the lot number, address, and longitude and latitude information stored in association with the storage device 200B may be transmitted, and this information may be displayed as listable information together with the ages. The aforementioned zoning and estimated site area may also be displayed as listable information.

[0042] As described above, the server 2 according to this embodiment includes a receiving unit 201 that receives data including a construction site, a detection unit 204 that inputs the data received by the receiving unit 201 into a trained model stored in the memory device 200B and detects a construction site from the data, and an estimation unit 205 that estimates the age of a building constructed at the construction site based on the acquisition date and time of the data including the construction site or the time when the detection unit 204 detected the construction site. In this embodiment, the age of a building is estimated from data including the construction site at a single point in time, eliminating the need for data from two different points in time as in the past, allowing for easy estimation of the age of a building. Furthermore, if the data including the construction site is image data (e.g., aerial photographs or satellite images), there is no need to create map information. Furthermore, unlike methods that estimate the age based on the Housing and Land Statistics Survey and National Political Survey conducted by the Statistics Bureau of the Ministry of Internal Affairs and Communications, this method can estimate the age of buildings other than residential buildings. Furthermore, unlike methods that obtain the age from real estate registration information, this method eliminates the effort and expense of obtaining real estate registration information for each property.

[0043] Furthermore, as described above, multiple types or multiple objects may be learned using a single image, or one type or one object may be learned using a single image. In this embodiment, the feature of the construction site includes the feature of objects related to the construction site. There are two or more objects, and they include at least one or more of the following: steel sheets, construction vehicles, heavy construction machinery (including construction cranes), construction materials, struts, gantry, and fall prevention nets. In this way, having two or more objects to learn is expected to enable more accurate learning of the construction site.

[0044] [Modification of the embodiment] In the above embodiment, the estimation unit 205 may estimate the age of the building constructed at the construction site based on the size (site area, etc.) of the construction site detected by the detection unit 204. For example, when the data including the construction site is image data, the estimation unit 205 may calculate the site area, etc. of the construction site based on the scale of the image and the number of pixels of the construction site, and change the value of Z in the above-mentioned equation (1) based on the calculated site area, etc. of the construction site. Since larger construction sites usually require longer construction periods, the value of Z may be increased according to the site area, etc. of the construction site. The size (site area, etc.) of the construction site may be estimated by segmenting satellite information using deep learning and calculating the site area, or information from an external database or website may be referenced. Alternatively, the user may input the size. In this way, by estimating the age of a building based on the size of the construction site (site area, etc.), a more accurate estimation of the age of the building can be expected.

[0045] In addition, the estimation unit 205 can also estimate the age of a building constructed at a construction site based on the type of object detected by the detection unit 204 from the data received by the receiving unit 201 and the estimated size of the construction site (site area, etc.). In this case, the estimation unit 205 estimates the scale (site area, etc.) of the construction site estimated by the above-mentioned method, for example. Next, the estimation unit 205 refers to the relationship between the processes required for the construction of the building, in other words, the processes required until the building is completed, the types of objects required for each process, and the construction period for each process depending on the size of the construction site (site area, etc.) (the relationship may be compiled into a database or quantified based on past performance), and estimates the processes at the construction site based on the types of objects detected by the detection unit 204, and estimates the remaining processes required until the building is completed. Next, the estimation unit 205 may obtain the construction period corresponding to the estimated size of the construction site (site area, etc.) for the remaining processes required until the estimated completion of the building, and estimate the completion date of the building based on the period required until the building is completed. Furthermore, the receiving unit 201 may receive two or more pieces of data at different times that include a construction site, the detection unit 204 may detect objects related to the construction site for each of the two or more pieces of data received by the receiving unit 201, and the estimation unit 205 may estimate the age of a building constructed at the construction site based on the acquisition date and time of the data that includes the construction site and the objects detected by the detection unit 204 from the data that includes the construction site. Specifically, the estimation unit 205 refers to the relationship between the steps required to complete the building and the types of objects required for each step (this relationship may be compiled into a database or quantified based on past performance), and estimates the construction site process for each piece of data received by the receiving unit 201 based on the type of object detected by the detection unit 204. Next, the estimation unit 205 may calculate the construction progress rate for the building based on the acquisition date and time of each piece of data or the time when the detection unit 204 detected the construction site and the construction site process, and estimate the date when the final process of the building will be completed (completion date) based on this calculated progress rate. By estimating the age of a building based on the progress of the construction site in this way, it is possible to expect a more accurate estimation of the age of the building.

[0046] Furthermore, the user terminal 3 may be configured to have some or all of the functions of the server 2 (for example, the acquisition unit 204, the calculation unit 205, the evaluation unit 206, and the recommendation unit 207). [Explanation of symbols]

[0047] 1: Information processing system 2: Server 200A: Communication IF 200B: Storage device 200C: CPU 201: Receiving unit 202: Transmission unit (output unit) 203: Storage device control unit 204: Detection unit 205: Estimation section 3: User terminal 300A: Communication IF 300B: Storage device 300C: Input device 300D:Display device 300E: CPU 301: Receiving unit 302: Sending department 303: Storage device control unit 304: Input reception section 305: Display device control unit 4: Network

Claims

1. Machine learning was performed using data including construction sites and the features of the construction sites contained in the data as training data. A trained model characterized by:

2. the feature amounts include feature amounts of objects related to the construction site; The object is 2 or more, At least steel plates, construction vehicles, heavy machinery for construction work (including construction cranes), construction materials, struts, platforms, fall prevention nets, Including one or more of the following: The trained model according to claim 1 .

3. The data including the construction site is image data. The trained model according to claim 1 .

4. a receiving unit for receiving data including a construction site; a detection unit that inputs the data received by the receiving unit into the trained model according to any one of claims 1 to 3 and detects a construction site from the data; an estimation unit that estimates the age of the building constructed at the construction site based on the acquisition date and time of the data or the time when the detection unit detected the construction site; An information processing device comprising:

5. The estimation unit Estimating the age of the building constructed at the construction site based on the size of the construction site; 5. The information processing apparatus according to claim 4,

6. The estimation unit estimating the age of the building constructed at the construction site based on the type of the object detected by the detection unit from the data and the estimated size of the construction site; 5. The information processing apparatus according to claim 4,

7. a receiving unit receiving data including a construction site; A step in which a detection unit inputs the data received by the receiving unit into a trained model according to any one of claims 1 to 3 to detect a construction site from the data; An estimation step of estimating the age of the building constructed at the construction site based on the acquisition date and time of the data or the time when the detection unit detected the construction site; An information processing method comprising:

8. Computer, a receiving unit for receiving data including the construction site; a detection unit that inputs the data received by the receiving unit into the trained model according to any one of claims 1 to 3 and detects a construction site from the data; an estimation unit that estimates the age of the building constructed at the construction site based on the acquisition date and time of the data or the time when the detection unit detected the construction site; An information processing program characterized by causing the program to function as:

Images