Engineering firm information service system

Abstract

To provide a technique allowing a user to use an image of an internal appearance or an external appearance of a building to find an engineering firm capable of easily and accurately constructing a building similar thereto.SOLUTION: An engineering firm information service system includes: an image acquisition part configured to acquire a user image provided from a user with regard to a building; an extraction model storage part configured to store an extraction model for extracting an image similar to the user image; a building image management part configured to associate and manage a building image of a building constructed by an engineering firm and engineering firm information including at least construction area information of the engineering firm; a similar image extraction part configured to use the prescribed extraction model acquired from the extraction model storage part and extract a similar image similar to the user image from a plurality of building images managed by the building image management part; and an information output part configured to output an extraction result of the similar image extraction part. The information output part outputs the similar image and engineering firm information of an engineering firm associated to the similar image.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a construction company information providing system that can be used when a user obtains information about construction companies. [Background technology]

[0002] Users who wish to build a house or other building want to find a construction company that can turn their vague image of a building with the design, scale, and specifications they desire into a concrete building. However, in reality, it is difficult to thoroughly examine all of the numerous construction companies that exist in each area, making it difficult for users to efficiently select a construction company that matches the image they have in mind. As a prior art document in this field, Patent Document 1 discloses a housing design and estimate support device that can turn a vague image of a building with the design, scale, and specifications that a user desires into a concrete building, and then easily find out how much the building will cost during the consideration stage. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-306632 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the housing design and estimate support device described in Patent Document 1 requires users to input plan search criteria, such as the size and layout of the house to be built, making it difficult to perform a visual search using images of the building's interior and exterior, and the only information available is a rough estimate of costs, making it impossible to find a construction company that meets users' needs. Furthermore, local construction companies are often rooted in their local communities and support customers' home building in a way that is closely tied to the local area, and are often limited in the areas in which they can work. Due to these circumstances, it has been difficult for users to efficiently gather information, taking into account the local characteristics, about construction companies that can build a building that matches their image.

[0005] As such, users who want to build a house or other building are often unfamiliar with searching for construction companies and are unable to easily and efficiently find a construction company that suits them. The purpose of this application is to provide a system that enables users to easily and efficiently select a construction company that takes into account regional characteristics, based on the image of the interior and exterior of the building they desire. [Means for solving the problem]

[0006] The present invention can be realized as the following aspects.

[0007] (1) One aspect of the present invention can be realized as a construction company information provision system including an image acquisition unit that acquires user images of buildings provided by users, an extraction model storage unit that stores extraction models that extract images similar to the images acquired by the image acquisition unit, a building image management unit that manages building images of buildings constructed by construction companies in association with construction company information including at least construction area information of the construction companies, a similar image extraction unit that extracts similar images similar to the user images from multiple building images managed by the building image management unit using a predetermined extraction model acquired from the extraction model storage unit, and an information output unit that outputs the extraction results of the similar image extraction unit. The information output unit outputs the similar images and the construction company information of the construction companies associated with the similar images. With this type of builder information system, users can obtain information on suitable builders who handle similar buildings simply by providing an image of the desired image. This builder information also includes information on the construction areas in which the builder can carry out their work, making it easy to select a builder that also takes regional characteristics into account.

[0008] (2) In the construction company information provision system of the above form, an interior / exterior determination unit may be provided that determines whether the user image is the interior or exterior of a building, and the similar image extraction unit may use an extraction model corresponding to the determination result of the interior / exterior determination unit to extract similar images that are similar to the user image from multiple building images managed by the building image management unit. By using an extraction model corresponding to the determination result of the interior and exterior determining unit, the user can search for a construction company based on either the interior or exterior of the building.

[0009] (3) In the construction company information provision system of the above-mentioned form (1) or (2), the extraction model may be tuned using a parameter that quantifies the hue in the building image as a feature, and the similar image extraction unit may be configured to use the feature that quantifies the hue in the user image as an input to the extraction model. By extracting similar images using feature quantities that quantify the hues in the user image, it becomes possible to extract building images that are similar to the user image with high accuracy and efficiency. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating an example of the configuration of a construction company information providing system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram showing an example of the content output from an information output unit in the construction company information providing system according to one embodiment of the present invention. [Figure 3] FIG. 10 is a diagram showing an example of a screen on which a user of the construction company information providing system according to one embodiment of the present invention operates an image on an information communication terminal. [Figure 4] FIG. 10 is a diagram illustrating an example of the configuration of a construction company information providing system according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] An embodiment will be described below with reference to the drawings. Figure 1 is a schematic diagram showing the entire construction company information system. This system can be configured by combining multiple dedicated or general-purpose devices connected via telecommunications lines such as a LAN line or the Internet, or by installing all the programs that execute each function on a single device. Users of the construction company information system 1 can access this system using an information and communication terminal such as a computer or tablet.

[0012] The image acquisition unit 10 acquires user images provided by users of the construction company information provision system 1. The users prepare user images by operating their own information communication terminals, and these user images are sent to the image acquisition unit 10. The user images are provided as interior images taken of the interior of a building or exterior images taken of the building from the outside.

[0013] The similar image extraction unit 20 extracts building images similar to the user image acquired by the image acquisition unit 10. Here, "similar" does not refer to similarity in the composition of the building images, but rather to similarity in evaluation based on information such as the hue of the images, and refers to a similarity in the atmosphere of the buildings as perceived by humans. Building images similar to the user image extracted by the similar image extraction unit 20 are called similar images. The extraction model required to perform this extraction is stored in the extraction model storage unit 30, and the similar image extraction unit 20 acquires a predetermined extraction model from the extraction model storage unit 30. The building image management unit 40 manages multiple building images, and the similar image extraction unit 20 extracts building images similar to the user image from this multiple building images.

[0014] The building images managed by the building image management unit 40 are images of buildings constructed by construction companies with which the user has a specified contract, and these building images allow the user to learn about the track record and characteristics of each construction company. The building images can include either or both interior images of the building's interior and exterior images of the building taken from the outside.

[0015] The building image management unit 40 can manage multiple building images provided by construction companies as a building image database 41. Alternatively, the building images of each construction company may be stored in a database managed by the respective construction company, and the building image management unit 40 may access the building images via the Internet as needed.

[0016] The building image management unit 40 manages building images in association with the contractor information of the contractor that constructed the building. This contractor information includes at least construction area information. Here, construction area information is information indicating the area in which the contractor can perform construction, and is expressed in prefecture or city units such as "Shizuoka Prefecture" or "Numazu City," for example. The contractor information also includes information that allows the user to contact the contractor if desired, such as the contractor's name, address, and access information.

[0017] The building image management unit 40 can manage the construction company information as a construction company information database 42. Alternatively, the information on each construction company may be stored in a database managed by each construction company, and the building image management unit 40 may access the construction company information via the Internet as needed.

[0018] The information output unit 50 outputs the extraction results of the similar image extraction unit 20. The information output here is output in a form that can be used by the user who provided the user image to the image acquisition unit 10, and the information includes the extracted similar images and the construction company information of the construction company linked to the similar images. The user who has acquired this information can use their own information communication terminal to extract and display the information they need from this information as appropriate.

[0019] 2 shows an example of information output from the information output unit 50. In this example, multiple similar images similar to the user image provided by the user are output, and construction company information linked to each similar image, such as construction area information, name, and contact information, is output. Here, the contact information may be in a form that is easy for each construction company to use, such as a phone number, email address, or website URL.

[0020] Below, we will explain an example of the operation screen displayed on the information communication terminal of a user who uses the construction company information provision system 1. Figure 3(a) is a diagram showing an example of the operation screen of the information communication terminal when the user selects a user image to provide to the construction company information provision system 1. When searching for a construction company, the user selects an image of the interior or exterior that they desire. This selection can be made by selecting the image that most closely resembles their image from a random array of photos. It is also possible to activate the camera function and select an image taken on the spot, or to select from an image album saved in the information communication terminal.

[0021] 3(b) is a diagram showing an example of an operation screen on the user's information communication terminal, on which information output from the information output unit 50 is displayed. The user can view the extracted multiple similar images and select any building image that matches their image. Furthermore, the user can access information about the construction company that constructed the building related to the selected building image on a screen not shown.

[0022] This allows users to use any information and communication terminal to obtain images of buildings and information about construction companies that they like, and also allows them to take a picture of a building they like on the spot and instantly search for construction companies that can construct a building similar to it.The construction company information that users can obtain here includes at least the construction area information of the construction company, so users can efficiently search for construction companies that can do the work in the area where they want to build a building.

[0023] A user image provided by a user typically represents either the interior or exterior of a building. As shown in FIG. 4, the construction company information provision system 1 may include an interior / exterior determination unit 11 that determines whether the user image acquired by the image acquisition unit 10 represents the interior or exterior of a building. This allows the similar image extraction unit 20 to use a more optimized extraction model depending on whether the user image represents the interior or exterior when extracting similar images. Specifically, if the interior / exterior determination unit 11 determines that the user image represents the interior, the similar image extraction unit 20 retrieves an interior model 31 from the extraction model storage unit 30 to extract similar images. If the interior / exterior determination unit 11 determines that the user image represents the exterior, the similar image extraction unit 20 retrieves an exterior model 32 from the extraction model storage unit 30 to extract similar images. The interior / exterior determination unit 11 outputs a determination result so that the similar image extraction unit 20 can operate in this manner. The interior / exterior determination unit 11 can determine whether the user image represents the interior or exterior, based on information provided by the user when submitting the user image. Alternatively, the determination can be automatically performed using artificial intelligence or the like based on the user image itself. This allows the similar image extraction unit 20 to use an extraction model optimized according to the attributes of the user image, and as a result, it becomes possible to extract similar images with higher accuracy.

[0024] The extraction models stored in the extraction model storage unit 30 can be created using similar image search techniques commonly used in the field. For example, when searching for images using an image as a query, a known method is to convert the image into a vector and use the vector to calculate the similarity of the images. To calculate similarity with higher accuracy, feature extraction techniques can be used, which extract and vectorize useful features from an image. In recent years, convolutional neural networks have become increasingly used for feature extraction in the image processing field. Converting images into vectors makes it possible to calculate similarity using Euclidean distance or cosine similarity between vectors. By appropriately adopting such techniques and using machine learning, it becomes possible to extract similar images.

[0025] In similar image search technology, the RGB values ​​of an image or the RGB histogram obtained from the entire image are used as features, but in the extraction model of the present invention, it is preferable that parameters that quantify the hues in the building image are tuned as features. In this case, the similar image extraction unit 20 uses the feature that quantifies the hues in the user image as input to the extraction model. Furthermore, if the building image is an interior view, the building image can be divided into regions such as ceiling, wall, and floor as a preprocessing step before extracting the feature values, and similar images can be extracted based on the feature values ​​of these regions. This makes it possible to extract similar images with higher accuracy, specializing in the interior view.

[0026] The above-described embodiments are presented as examples and do not limit the scope of the invention. The present invention can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. [Explanation of symbols]

[0027] 1. Construction company information system 10 Image acquisition unit 11 Interior and exterior inspection department 20 Similar image management department 30 Extraction model storage unit 31 Introspection Model 32 Appearance Model 40 Building Image Management Department 41 Building Image Database 42 Construction company information database 50 Information output section

Claims

1. an image acquisition unit that acquires a user image of a building provided by a user; an extraction model storage unit that stores an extraction model for extracting an image similar to the image acquired by the image acquisition unit; a building image management unit that manages building images of buildings constructed by a construction company in association with construction company information including at least construction area information of the construction company; a similar image extraction unit that extracts a similar image that is similar to the user image from the plurality of building images managed by the building image management unit using a predetermined extraction model acquired from the extraction model storage unit; an information output unit that outputs the extraction result of the similar image extraction unit; The information output unit outputs the similar image and the construction company information of the construction company associated with the similar image. Construction company information system.

2. an interior / exterior determining unit that determines whether the user image is an interior or exterior view of a building; the similar image extraction unit extracts similar images that are similar to the user image from a plurality of building images managed by the building image management unit, using an extraction model corresponding to the determination result of the interior / appearance determination unit; The construction company information providing system according to claim 1.

3. the extraction model is tuned using a parameter that quantifies a hue in the building image as a feature, the similar image extraction unit uses a feature amount obtained by quantifying a hue in the user image as an input to the extraction model; 3. The construction company information providing system according to claim 1 or 2.

Images