JP7621082B2 - SPATIAL EVALUATION SUPPORT DEVICE, SPATIAL EVALUATION SUPPORT SYSTEM, AND SPATIAL EVALUATION SUPPORT PROGRAM - Google Patents

Description

The present invention relates to a spatial evaluation support device, a spatial evaluation support system, and a spatial evaluation support program.

Conventionally, the following technologies have been used to help users feel attracted to infrastructure and structures in cities, towns, tourist spots, etc.

Patent Document 1 discloses a town structure in which a main street area is provided that runs vertically and/or across a town area, and energy plant facilities and lifeline infrastructure for supplying at least electricity and gas are provided above and/or underground of the main street area. In this town structure, an infrastructure network is constructed between the main street area and urban areas other than the main street area.

Patent Document 2 discloses a street structure in which a pedestrian waiting space is provided along the pedestrian-vehicle boundary that separates the roadway from the sidewalk, where a sidewalk 4m or more wide is provided on at least one side of the roadway. In this street structure, the pedestrian waiting space is provided in a substantially rectangular area in plan view with the long side in the direction of the roadway, and the width intersecting the direction of the roadway is set to 2m or more and half the width of the sidewalk. In this street structure, a walking floor made of a different material from the paved surface of the sidewalk is provided on the area, and benches, chairs, steps, and other seating areas are provided. In this street structure, the entire long side of the area on the roadway side and one or both short sides connected to the long side that are 1.5m or more on the roadway side are seamlessly surrounded by a vehicle protection body that rises to a height of 1m or more above the surface of the walking floor.

Furthermore, Patent Document 3 discloses an exterior structure for a house in which the main part of the site, where the building and garden are constructed, is separated from the corner part by a passageway, and the corner part is left as an open space.

JP 2019-75042 A JP 2018-53661 A JP 2002-106187 A

However, the technologies disclosed in Patent Documents 1 to 3 were intended to contribute to the construction of attractive infrastructure in the future. For this reason, these technologies were not able to contribute to making existing infrastructure such as cities, townscapes, and tourist destinations more attractive and attractive so that people would want to visit those infrastructure structures.

The present invention has been made in consideration of the above circumstances, and aims to provide a spatial evaluation support device, a spatial evaluation support system, and a spatial evaluation support program that enable users to quantitatively judge the attractiveness of existing infrastructure and structures.

The spatial evaluation support device of the present invention described in claim 1 comprises an acquisition unit that acquires a photographed image of a space to be evaluated, an extraction unit that extracts components in the space from the photographed image acquired by the acquisition unit, and a derivation unit that applies a predetermined weighting to the components extracted by the extraction unit for each type of component, and derives an evaluation value related to at least one of safety and security for the space, wherein the components are components classified according to each of the safety and security, and the derivation unit derives the evaluation value related to safety using an arithmetic formula that divides the actual amount of the corresponding component in the space to be evaluated by the maximum value of the component, and derives the evaluation value related to security using an arithmetic formula that subtracts the value obtained by dividing the actual amount of the corresponding component in the space to be evaluated by the maximum value of the component from 1 .

According to the spatial evaluation support device of the present invention described in claim 1, a photographed image of the space to be evaluated is obtained, components of the space are extracted from the obtained photographed image, and the extracted components are weighted in advance for each type of component to derive an evaluation value for at least one of safety and security for the space, the components are classified into components for each of the safety and security, the evaluation value for safety is derived using an arithmetic formula in which the actual amount of the corresponding component in the space to be evaluated is divided by the maximum value of the component, and the evaluation value for security is derived using an arithmetic formula in which the actual amount of the corresponding component in the space to be evaluated is divided by the maximum value of the component and the value obtained by subtracting from 1 , thereby allowing a user to quantitatively judge the attractiveness of existing infrastructure.

The spatial evaluation support device according to the present invention described in claim 2 is the spatial evaluation support device described in claim 1, further comprising a presentation unit that combines the evaluation value derived by the derivation unit with a map image and presents it.

According to the spatial evaluation support device of the present invention described in claim 2, the derived evaluation value is synthesized and presented on a map image, so that the user can more effectively quantitatively judge the attractiveness of existing infrastructure and structures.

The spatial evaluation support device according to the present invention described in claim 3 is the spatial evaluation support device described in claim 2, in which the map image is a social heat map image.

According to the spatial evaluation support device of the present invention described in claim 3, by making the map image into a social heat map image, it is possible to more effectively allow users to quantitatively judge the attractiveness of existing infrastructure and structures.

The spatial evaluation support device according to the present invention described in claim 4 is the spatial evaluation support device described in any one of claims 1 to 3, in which the extraction unit extracts the components using a segmentation model trained using the photographed image acquired by the acquisition unit.

According to the spatial evaluation support device of the present invention described in claim 4, the above components are extracted using a segmentation model trained using the acquired photographic images, making it possible to extract the above components with higher accuracy.

The spatial evaluation support device according to the present invention described in claim 5 is a spatial evaluation support device described in any one of claims 1 to 4, in which the components include at least one of a sidewalk, a roadway, a pedestrian bridge, a flyer, graffiti, and a parked vehicle on the street.

The spatial evaluation support device according to the present invention described in claim 5 can derive an evaluation value according to the included components.

The spatial evaluation support device according to the present invention described in claim 6 is the spatial evaluation support device described in claim 5, in which the weighting is performed on at least one type of the components.

According to the spatial evaluation support device of the present invention described in claim 6, by weighting at least one type of component, the weighted component can be reflected in the evaluation value to a degree corresponding to the weighting.

The spatial evaluation support device according to the present invention described in claim 7 is the spatial evaluation support device described in any one of claims 1 to 6, in which the captured image is an omnidirectional image.

According to the spatial evaluation support device of the present invention described in claim 7, by making the captured image an omnidirectional image, it is possible to derive a more effective evaluation value that takes into account not only the forward situation but also the surrounding situation.

The spatial evaluation support system according to the present invention as described in claim 8 includes a terminal including the spatial evaluation support device as described in any one of claims 1 to 7, a receiving unit that receives from the spatial evaluation support device an evaluation value derived by the derivation unit of the spatial evaluation support device, and a display control unit that controls a display unit to display information indicating the evaluation value received by the receiving unit.

According to the spatial evaluation support system of the present invention described in claim 8, by displaying information showing the evaluation value derived by the spatial evaluation support device of the present invention, it is possible to allow a user to quantitatively judge the attractiveness of existing infrastructure and structures.

The spatial evaluation support program of the present invention described in claim 9 is a process of acquiring an image of a space to be evaluated, extracting components of the space from the acquired image, applying a predetermined weighting to the extracted components for each type of component, and deriving an evaluation value for at least one of safety and security for the space, wherein the components are components classified according to each of the safety and security, and the computer is caused to execute a process of deriving the evaluation value for safety using an arithmetic formula that divides the actual amount of the corresponding component in the space to be evaluated by the maximum value of the component, and deriving the evaluation value for security using an arithmetic formula that subtracts from 1 the value obtained by dividing the actual amount of the corresponding component in the space to be evaluated by the maximum value of the component .

According to the spatial evaluation support program of the present invention as described in claim 9, a photographed image of the space to be evaluated is obtained, components of the space are extracted from the obtained photographed image, and the extracted components are weighted according to a predetermined weighting for each type of component to derive an evaluation value for at least one of safety and security for the space, the components are classified into components for each of the safety and security , the evaluation value for safety is derived using an arithmetic formula that divides the actual amount of the corresponding component in the space to be evaluated by the maximum value of the component, and the evaluation value for security is derived using an arithmetic formula that subtracts the value obtained by dividing the actual amount of the corresponding component in the space to be evaluated by the maximum value of the component from 1 , thereby allowing a user to quantitatively judge the attractiveness of existing infrastructure.

As described above, the present invention allows users to quantitatively judge the attractiveness of existing infrastructure and structures.

1 is a block diagram showing an example of a hardware configuration of a space evaluation support system according to an embodiment. FIG. 1 is a diagram for explaining a segmentation model according to an embodiment, in which the left diagram shows an example of an image obtained by shooting, and the right diagram shows an example of a classification result of the image shown in the left diagram. 1 is a block diagram showing an example of a functional configuration of a space evaluation support system according to an embodiment. FIG. 2 is a schematic diagram showing an example of a configuration of an image information database according to the embodiment. FIG. 4 is a schematic diagram showing an example of a configuration of a weighting information database according to the embodiment. FIG. 2 is a schematic diagram showing an example of a configuration of an evaluation value information database according to the embodiment. 11 is a flowchart illustrating an example of a space evaluation support process according to the embodiment. 11 is a flowchart illustrating an example of an evaluation value presentation process according to the embodiment. 11 is a flowchart illustrating an example of an evaluation value display process according to the embodiment. FIG. 4 is a front view showing an example of a configuration of an initial screen according to the embodiment. FIG. 13 is a front view showing an example of a configuration of an evaluation result screen according to the embodiment.

Hereinafter, an example of an embodiment of the present invention will be described in detail with reference to the drawings. In this embodiment, the present invention will be described as being applied to a spatial evaluation support system that includes a spatial evaluation support device configured with a server computer or the like, and multiple subject terminals that are terminals used individually by each subject.

First, the configuration of the spatial evaluation support system 90 according to this embodiment will be described with reference to Figures 1 to 3. Figure 1 is a block diagram showing an example of the hardware configuration of the spatial evaluation support system 90 according to this embodiment. Also, Figure 2 is a diagram used to explain the segmentation model according to this embodiment, with the left diagram showing an example of an image obtained by photography, and the right diagram showing an example of the classification results of the image shown in the left diagram. Furthermore, Figure 3 is a block diagram showing an example of the functional configuration of the spatial evaluation support system 90 according to this embodiment.

As shown in FIG. 1, the spatial evaluation support system 90 according to this embodiment includes a spatial evaluation support device 10 and multiple subject terminals 30, each of which can access a network 80. Examples of the spatial evaluation support device 10 include information processing devices such as personal computers and server computers. Examples of the subject terminals 30 include portable terminals such as smartphones, tablet terminals, and PDAs (Personal Digital Assistants, mobile information terminals).

The subject terminal 30 according to this embodiment is a terminal held by each of a plurality of subjects who are to use the space evaluation support system 90. The subject terminal 30 includes a CPU (Central Processing Unit) 31, a memory 32 as a temporary storage area, a non-volatile storage unit 33, an input unit 34 such as a touch panel, a display unit 35 such as a liquid crystal display, and a media read/write device (R/W) 36. The subject terminal 30 also includes a camera 38, a microphone 39, a GPS (Global Positioning Systems) 40, and a wireless communication unit 42. The CPU 31, the memory 32, the storage unit 33, the input unit 34, the display unit 35, the media read/write device 36, the camera 38, the microphone 39, the GPS 40, and the wireless communication unit 42 are connected to each other via a bus B1. The media read/write device 36 reads information written in the recording medium 37 and writes information to the recording medium 37.

The storage unit 33 is realized by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, or the like. The storage unit 33, which serves as a storage medium, stores an evaluation value display program 33A. The evaluation value display program 33A is stored in the storage unit 33 when a recording medium 37 to which the evaluation value display program 33A has been written is set in the medium reading and writing device 36, and the medium reading and writing device 36 reads the evaluation value display program 33A from the recording medium 37. The CPU 31 reads the evaluation value display program 33A from the storage unit 33, expands it in the memory 32, and sequentially executes the processes of the evaluation value display program 33A.

On the other hand, the spatial evaluation support device 10 is a device that comprehensively stores and manages various information handled by the spatial evaluation support system 90. The spatial evaluation support device 10 comprises a CPU 11, a memory 12 as a temporary storage area, a non-volatile storage unit 13, an input unit 14 such as a keyboard and mouse, a display unit 15 such as a liquid crystal display, a media reading and writing device 16, and a communication interface (I/F) unit 18. The CPU 11, memory 12, storage unit 13, input unit 14, display unit 15, media reading and writing device 16, and communication interface (I/F) unit 18 are connected to each other via a bus B2. The media reading and writing device 16 reads information written in the recording medium 17 and writes information to the recording medium 17.

The storage unit 13 is realized by a HDD, SSD, flash memory, etc. The storage unit 13 as a storage medium stores a space evaluation support program 13A and an evaluation value presentation program 13B. The space evaluation support program 13A is stored in the storage unit 13 by setting the recording medium 17 in which the space evaluation support program 13A is written to the medium reading and writing device 16, and the medium reading and writing device 16 reading the space evaluation support program 13A from the recording medium 17. The evaluation value presentation program 13B is stored in the storage unit 13 by setting the recording medium 17 in which the evaluation value presentation program 13B is written to the medium reading and writing device 16, and the medium reading and writing device 16 reading the evaluation value presentation program 13B from the recording medium 17. The CPU 11 reads the space evaluation support program 13A and the evaluation value presentation program 13B from the storage unit 13, expands them in the memory 12, and sequentially executes the processes that the space evaluation support program 13A and the evaluation value presentation program 13B each have.

The storage unit 13 also stores a segmentation model 13C. The segmentation model 13C according to this embodiment is a segmentation model in which image information showing a captured image of a space to be evaluated by the space evaluation support system 90 is used as input information, and a segmentation image in which regions of predetermined types of components from the captured image are classified by type is used as output information. In this embodiment, the types of the above-mentioned components are flyers (mainly used as advertising media) pasted on sidewalks, roadways, pedestrian bridges, walls, utility poles, etc., graffiti written on fences and building shutters, and cars parked on the street. However, the types are not limited to these, and in addition to these components, any one or a combination of moving bicycles, brightness of the space, garbage, visibility of the road, etc. may be used as the types of the above-mentioned components.

In the spatial evaluation support system 90 according to this embodiment, a model using a semantic segmentation technique using deep learning technology is applied as the segmentation model 13C. FIG. 2 shows examples of images before and after classification by element using image segmentation technology. The left diagram in FIG. 2 is an example of an image before classification (referred to as "original image" in FIG. 2), and the right diagram in FIG. 2 is an example of an image after classification for that image (referred to as "segmented image" in FIG. 2). In the example shown in FIG. 2, vegetation, sidewalks, roadways, buildings, and sky are applied as elements, and each element is represented with a different density.

In addition, in this embodiment, training data is created using omnidirectional images collected by Google for Google Street View, and the segmentation model 13C is trained using the training data. However, this is not limited to this form, and an image obtained by capturing an image using the camera 38 on the subject terminal 30 may be applied as the omnidirectional image. Furthermore, the captured image is not limited to an omnidirectional image, and an image obtained by capturing an image of only the front in general may be applied.

For details of semantic segmentation using deep learning technology, see "ICUC10_Fang-Ying GONG_GSV-ViewFactorMaps_6-10 August 2018," "Using Google Street View to quantify street elements in densely populated urban environments for climate research," "Badrinarayanan, V., Kendall, A., & Cipolla, R. (2017). Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE transactions on pattern analysis and machine intelligence, 39(12), 2481-2495." "Chen, L. C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2017). Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE transactions on pattern analysis and machine intelligence, 40(4), 834-848." and is a widely known technique, so further explanation will be omitted here.

In this way, in the spatial evaluation support system 90 according to this embodiment, a model using a semantic segmentation technique using deep learning technology is applied as the segmentation model 13C, but this is not the only possible form. For example, a model using other image segmentation techniques such as instance-aware segmentation may be applied as the segmentation model 13C.

The storage unit 13 also stores an image information database 13D, a weighting information database 13E, and an evaluation value information database 13F. The image information database 13D, the weighting information database 13E, and the evaluation value information database 13F will be described in detail later.

Next, referring to FIG. 3, the functional configuration of the spatial evaluation support device 10 and the subject terminal 30 according to this embodiment will be described.

As shown in FIG. 3, the spatial evaluation support device 10 includes an acquisition unit 11A, an extraction unit 11B, a derivation unit 11C, and a presentation unit 11D. The CPU 11 of the spatial evaluation support device 10 executes the spatial evaluation support program 13A and the evaluation value presentation program 13B, thereby functioning as the acquisition unit 11A, the extraction unit 11B, the derivation unit 11C, and the presentation unit 11D.

The acquisition unit 11A according to this embodiment acquires a captured image of the space to be evaluated (hereinafter referred to as the "evaluation target space"). In this embodiment, as the captured image, an omnidirectional image collected by Google for Google Street View is applied, similar to the captured image used when training the segmentation model 13C. However, this is not limited to this form, and the omnidirectional image may be an image obtained by capturing an image using the camera 38 of the subject terminal 30, or instead of the omnidirectional image, an image obtained by capturing only the normal forward view may be applied, similar to the captured image applied in training the segmentation model 13C.

The extraction unit 11B according to this embodiment extracts components in the evaluation target space from the captured image acquired by the acquisition unit 11A. Here, the extraction unit 11B according to this embodiment extracts components in the evaluation target space using a segmentation model 13C trained using the captured image acquired by the acquisition unit 11A. This allows the above components to be extracted with higher accuracy than when a segmentation model trained using a captured image other than the captured image acquired by the acquisition unit 11A is used. However, this is not limited to this form, and the above components may be extracted from the captured image of the evaluation target space using a segmentation model trained using a captured image other than the captured image acquired by the acquisition unit 11A. Also, each component may be extracted from the captured image of the evaluation target space using a conventionally known image recognition technology without using a segmentation model.

The derivation unit 11C according to this embodiment then assigns a predetermined weighting to each type of component extracted by the extraction unit 11B, and derives an evaluation value for both safety and security for the space to be evaluated. In this way, the derivation unit 11C according to this embodiment derives evaluation values for both safety and security, but this is not limited thereto, and the derivation unit 11C may derive an evaluation value for only one of safety and security.

The presentation unit 11D according to this embodiment combines the evaluation value derived by the derivation unit 11C with a map image and presents it. The presentation unit 11D according to this embodiment presents the evaluation value by transmitting information indicating the evaluation value to the requested subject terminal 30 and displaying it on the display unit 35 of the subject terminal 30, but this is not limited to the presenting. For example, the presentation of the evaluation value may be performed by displaying it on the display unit 15 of the spatial evaluation support device 10. The presentation of the evaluation value by the presentation unit 11D is not limited to displaying it on the display unit, and may be presented by voice or by printing it on an image forming device (a so-called printer).

In addition, in this embodiment, a social heat map image is applied as the map image. The social heat map image is an image showing a map that highlights locations where there is a lot of information that matches the target's category by displaying areas with different densities or colors over the map image that is normally displayed for the corresponding area. That is, in this embodiment, each target is asked to answer multiple questions in advance, and the answer results are analyzed and classified to determine the category of each target in advance. Then, the social heat map image according to this embodiment is displayed over the map image so that the density increases in locations where there is a lot of information that matches the target's category (in this embodiment, information posted on SNS (Social Networking Service)). However, this is not limited to changing the density, and the color may be changed in order from high density to low density, such as red → yellow → green.

In this embodiment, the spatial evaluation support device 10 is connected via a network 80 or the like to a server that provides the latest social heat map images of the areas (hereinafter referred to as "target areas") that are the subject of the spatial evaluation support system 90. The spatial evaluation support device 10 then obtains the latest social heat map images from this server, and sequentially updates the social heat map images stored in the image information database 13D (see also FIG. 4) described below. However, this is not the only possible form, and the spatial evaluation support device 10 itself may sequentially update the social heat map images corresponding to each subject.

On the other hand, the subject terminal 30 according to this embodiment includes a receiving unit 31A and a display control unit 31B. The CPU 31 of the subject terminal 30 executes the evaluation value display program 33A, thereby functioning as the receiving unit 31A and the display control unit 31B.

The receiving unit 31A according to this embodiment receives information indicating the evaluation value derived by the derivation unit 11C of the spatial evaluation support device 10 from the spatial evaluation support device 10. The display control unit 31B also controls the display of the information indicating the evaluation value received by the receiving unit 31A on the display unit 35.

Next, the image information database 13D according to this embodiment will be described with reference to FIG. 4. FIG. 4 is a schematic diagram showing an example of the configuration of the image information database 13D according to this embodiment.

The image information database 13D according to this embodiment is a database in which omnidirectional images and the like used for learning the segmentation model 13C and as captured images of the evaluation target space are registered. As shown in FIG. 4, the image information database 13D according to this embodiment stores information such as the target area name, social heat map image, omnidirectional image, target district position, and target district ID (Identification).

The target area name is information indicating the name of each of the target areas, and the social heat map image is information indicating the above-mentioned social heat map image for each subject in the target area indicated by the corresponding target area name. The omnidirectional image is information indicating the above-mentioned omnidirectional image taken within the corresponding target area, and the target district location is information indicating the location of the district where the corresponding omnidirectional image was taken (hereinafter referred to as the "target district").

In this embodiment, the target area is divided into areas surrounded by a rectangle of a predetermined size in a plan view, and each divided area is applied as the target district. In addition, in this embodiment, the target district position is defined as the coordinate position in a two-dimensional coordinate system of a pair of diagonal corners of the rectangle indicating the corresponding target district. However, this is not limited to this form, and for example, other shapes such as a circle or an ellipse may be applied instead of the rectangle, and latitude and longitude may be applied instead of the coordinate position in the two-dimensional coordinate system.

Furthermore, the above-mentioned target area ID is information that is assigned in advance and is different for each target area in order to individually identify the corresponding target area.

Next, the weighting information database 13E according to this embodiment will be described with reference to FIG. 5. FIG. 5 is a schematic diagram showing an example of the configuration of the weighting information database 13E according to this embodiment.

The weighting information database 13E according to this embodiment stores information for specifying the weighting of each component, which is used when deriving the evaluation value described above. As shown in FIG. 5, the weighting information database 13E according to this embodiment stores information on type, component, maximum value, and weight value.

The above-mentioned type is information indicating each type of safety and security, and the above-mentioned component is information indicating the above-mentioned component in the space to be evaluated. The above-mentioned maximum value is information indicating the maximum value expected for the corresponding component, and the above-mentioned weight value is information indicating the importance of the corresponding component.

In the space evaluation support system 90 according to this embodiment, the evaluation value of each component (hereinafter referred to as "component-specific evaluation value") is derived individually by the following formulas (1) and (2). In formula (1), F _m represents the component-specific evaluation value of the mth type of component related to safety, fv _m represents the actual amount of the component in the evaluation target space, and fmax _m represents the maximum value of the component. In formula (2), R _n represents the component-specific evaluation value of the nth type of component related to security, rv _n represents the actual amount of the component in the evaluation target space, and rmax _n represents the maximum value of the component.

F _m =fv _m /fmax _m (1)

R _n =1-rv _n /rmax _n (2)

In the spatial evaluation support system 90 according to this embodiment, as described below, the evaluation value for safety (hereinafter referred to as "safety level") and the evaluation value for security (hereinafter referred to as "security level") are derived by calculating the weighted average of the evaluation values for each element. For this reason, the weighting information database 13E according to this embodiment stores maximum values and weights for each category of safety and security and for each type of component. Note that, although the example shown in FIG. 5 shows a case where the maximum weight value is "10", it goes without saying that this is not limited to this.

Next, the evaluation value information database 13F according to this embodiment will be described with reference to FIG. 6. FIG. 6 is a schematic diagram showing an example of the configuration of the evaluation value information database 13F according to this embodiment.

The evaluation value information database 13F according to this embodiment registers the evaluation values obtained by executing the spatial evaluation support program 13A. As shown in FIG. 6, the evaluation value information database 13F according to this embodiment stores information on the target area name, the target district ID, and the evaluation value.

The target area name and the target district ID are the same information as the target area name and the target district ID in the image information database 13D. The evaluation value is information indicating the evaluation value of the corresponding target district, and as shown in FIG. 6, three types of evaluation values are applied in this embodiment: safety level, security level, and overall level.

In the spatial evaluation support system 90 according to this embodiment, the safety level, security level, and overall level are derived by calculation using the following formulas (4) to (6). In formula (1), FD represents the safety level, F ₁ to F _m represent element-specific evaluation values of each component related to safety, and FW ₁ to FW _m represent weight values of the corresponding component. In formula (2), RD represents the security level, R ₁ to R _n represent element-specific evaluation values of each component related to security, and RW ₁ to RW _n represent weight values of the corresponding component. Furthermore, in formula (3), TD represents the overall level.

FD=(F ₁ ×FW ₁ +F ₂ ×FW ₂ +...+F _m ×FW _m )
/(FW ₁ +FW ₂ +...+FW _m ) (4)

RD=(R ₁ ×RW ₁ +R ₂ ×RW ₂ +...+R _n ×RW _n )
/(RW ₁ +RW ₂ +...+RW _n ) (5)

TD=(FD+RD)/2 (6)

As shown in formulas (4) and (5), in the spatial evaluation support system 90 according to this embodiment, a weighted average value using element-specific evaluation values corresponding to each component is applied as the safety level FD and the security level RD. However, this is not limited to this form, and for example, an arithmetic mean value of element-specific evaluation values corresponding to each component may be applied as the safety level FD and the security level RD. Also, as shown in formula (5), in the spatial evaluation support system 90 according to this embodiment, an arithmetic mean value of the safety level FD and the security level RD is applied as the overall level TD. However, this is not limited to this form, and for example, a weighted average value of the safety level FD and the security level RD may be applied as the overall level TD.

Next, the operation of the spatial evaluation support system 90 according to this embodiment will be described with reference to Figs. 7 to 11. Fig. 7 is a flowchart showing an example of the spatial evaluation support process according to this embodiment. Fig. 8 is a flowchart showing an example of the evaluation value presentation process according to this embodiment. Fig. 9 is a flowchart showing an example of the evaluation value display process according to this embodiment. Fig. 10 is a front view showing an example of the configuration of an initial screen according to this embodiment. Furthermore, Fig. 11 is a front view showing an example of the configuration of an evaluation result screen according to this embodiment.

First, referring to FIG. 7, the operation of the spatial evaluation support device 10 according to this embodiment when executing the spatial evaluation support process will be described. When a user of the spatial evaluation support device 10 (for example, an administrator of the spatial evaluation support system 90) inputs an instruction to start the execution of the spatial evaluation support process via the input unit 14, the CPU 11 of the spatial evaluation support device 10 executes the spatial evaluation support program 13A, thereby executing the spatial evaluation support process shown in FIG. 7. Note that, in order to avoid confusion, a case will be described here in which the image information database 13D and the weighting information database 13E have already been constructed. Also, in order to avoid confusion, a case will be described here in which the segmentation model 13C has already been trained using some of the omnidirectional images registered in the image information database 13D. Furthermore, in order to avoid confusion, a case will be described here in which the omnidirectional image to be evaluated (hereinafter referred to as the "evaluation target image") in the image information database 13D is specified in advance.

In step 100 of FIG. 7, the CPU 11 reads one of the images to be evaluated (hereinafter referred to as the "image to be processed") from the image information database 13D together with the corresponding target area ID. In step 102, the CPU 11 reads the trained segmentation model 13C from the memory unit 13.

In step 104, the CPU 11 inputs the image to be processed into the segmentation model 13C. In response to the input of the image to be processed, a corresponding segmentation image is output from the segmentation model 13C, and in step 106, the CPU 11 acquires the segmentation image output from the segmentation model 13C.

In step 108, the CPU 11 extracts each of the above-mentioned components from the acquired segmentation image, and uses the extracted components to derive the safety level FD, the security level RD, and the overall level TD according to the above-mentioned equations (1) to (5).

In step 110, the CPU 11 stores the derived safety level FD, security level RD, and overall level TD in the evaluation value information database 13F together with the target area ID read in the processing of step 100.

In the next step 112, the CPU 11 determines whether the above processing has been completed for all images to be evaluated. If the determination is negative, the process returns to step 100, whereas if the determination is positive, the spatial evaluation support process ends. Note that when the processes of steps 100 to 112 are repeatedly executed, in step 100, evaluation target images that have not been processed up to that point become images to be processed.

The spatial evaluation support process described above results in the construction of an evaluation value information database 13F, as shown in FIG. 6 as an example.

Next, referring to FIG. 8, the operation of the spatial evaluation support device 10 according to this embodiment when executing the evaluation value presentation process will be described.

In the spatial evaluation support system 90 according to this embodiment, when a subject wishes to refer to the above-mentioned evaluation values of various places in a target area, the subject uses the subject's own terminal 30 to execute the evaluation value display process described below. In this evaluation value display process, the subject transmits reference request information including information indicating the target area for which the subject wishes to refer to the evaluation value (hereinafter referred to as "designated target area information") to the spatial evaluation support device 10. When this reference request information is received, the CPU 11 of the spatial evaluation support device 10 executes the evaluation value presentation program 13B, thereby executing the evaluation value presentation process shown in FIG. 8. Note that, in order to avoid confusion, a case will be described here where the evaluation value information database 13F has already been constructed.

In step 150 of FIG. 8, the CPU 11 extracts the specified target area information from the received reference request information. In step 152, the CPU 11 reads the target area IDs and evaluation values (safety level FD, security level RD, and overall level TD) of all target areas included in the target area indicated by the specified target area information (hereinafter referred to as the "processing target area") from the evaluation value information database 13F.

In step 154, the CPU 11 reads out from the image information database 13D the target area location corresponding to the read target area ID, the processing target area, and the social heat map image corresponding to the target person who is the access source.

In step 156, the CPU 11 uses the retrieved social heat map image, the evaluation value of each target district, and the information on the target district location to create information (hereinafter referred to as "evaluation result screen information") that shows an evaluation result screen with a predetermined configuration. In step 158, the CPU 11 transmits the created evaluation result screen information to the target terminal 30 that originated the access, and then ends this evaluation value presentation process.

Next, referring to FIG. 9, the operation of the subject terminal 30 according to this embodiment when executing the above-mentioned evaluation value display process will be described. The evaluation value display process shown in FIG. 9 is executed by the CPU 31 of any of the subject terminals 30 executing the evaluation value display program 33A. The evaluation value display process shown in FIG. 9 is executed, for example, when an instruction to execute the evaluation value display process is input from any of the subjects (hereinafter referred to as the "subject") via the input unit 34 of his/her subject terminal 30.

In step 200 of FIG. 9, the CPU 31 controls the display unit 35 to display an initial screen having a predetermined configuration, and in step 202, the CPU 31 waits until the specified information is input.

As an example, as shown in FIG. 10, the initial screen according to this embodiment displays a message prompting input of the target area name, and also displays an input area 35A in which the target area name of the target area for which evaluation values are to be referenced is input. When the initial screen shown in FIG. 10 is displayed on the display unit 35, the subject uses the input unit 34 to input the name of the target area for which evaluation values are to be referenced in the input area 35A, and then selects the end button 35B. In response to this, step 202 is judged to be positive, and the process proceeds to step 204. Note that, in this embodiment, the input of the target area name on the initial screen is in the form of directly inputting the target area name, but this is not limited thereto, and it is also possible to display all target area names for which evaluation values can be referenced in a pull-down format, and to select the desired target area name from the displayed target area names.

In step 204, the CPU 31 transmits the above-mentioned reference request information to the spatial evaluation support device 10. In response to this, the spatial evaluation support device 10 executes the evaluation value presentation process as described above, and transmits evaluation result screen information to the target user terminal 30 from which the user accessed the device.

Therefore, in step 206, the CPU 31 waits until evaluation result screen information is received from the spatial evaluation support device 10. In step 208, the CPU 31 controls the display unit 35 to display the evaluation result screen indicated by the received evaluation result screen information, and in step 210, the CPU 31 waits until predetermined information is input, and then ends this evaluation value display process.

As an example, as shown in FIG. 11, the evaluation result screen according to this embodiment displays three types of evaluation values, namely, safety level FD, security level RD, and overall level TD, of the target area included in the processing target area at corresponding positions on the social heat map image of the processing target area. Note that the corresponding positions are the positions indicated by the target area positions described above.

Therefore, by referring to the evaluation result screen, participants can understand the evaluation values of each area within the processing area along with the social heat map image.

As described above, this embodiment includes an acquisition unit 11A that acquires a photographed image of the space to be evaluated, an extraction unit 11B that extracts components of the space from the photographed image acquired by the acquisition unit 11A, and a derivation unit 11C that assigns a predetermined weighting for each type of component to the components extracted by the extraction unit 11B and derives an evaluation value for at least one of safety and security for the space. Therefore, it is possible to allow users to quantitatively judge the attractiveness of existing infrastructure and structures.

In addition, according to this embodiment, the derived evaluation value is synthesized and presented on a map image. This allows users to more effectively quantitatively judge the attractiveness of existing infrastructure and structures.

Furthermore, according to this embodiment, the map image is a social heat map image. Therefore, it is possible to more effectively allow users to quantitatively judge the attractiveness of existing infrastructure and structures.

In addition, according to this embodiment, the above components are extracted using a segmentation model trained using the captured image acquired by the acquisition unit 11A. Therefore, the above components can be extracted with higher accuracy.

Furthermore, according to this embodiment, the above components include sidewalks, roadways, pedestrian bridges, flyers, graffiti, and parked vehicles on the street. Therefore, it is possible to derive an evaluation value according to the included components.

Furthermore, according to this embodiment, the weighting is applied to at least one type of component. Therefore, the weighted component can be reflected in the evaluation value to a degree according to the weighting.

Furthermore, according to this embodiment, the captured image is an omnidirectional image. Therefore, it is possible to derive a more effective evaluation value that takes into account not only the forward view but also the surrounding conditions.

In the above embodiment, the evaluation result screen is described as displaying only the evaluation values of the target areas in the processing target area superimposed on the social heat map image, but this is not limited to the above. For example, in addition to each evaluation value, the omnidirectional image from which the evaluation value was derived may be displayed in the vicinity of the evaluation value.

In the above embodiment, the subject himself/herself specifies the processing target area, but this is not limiting. For example, a GPS 40 built into the subject terminal 30 held by the subject may be used to automatically apply an area including the location of the subject terminal 30 as the processing target area. In addition, information indicating the subject's preference trends may be acquired in advance by the spatial evaluation support device 10, and information on the target area according to the subject's preferences may be provided to the subject at any time.

In the above embodiment, the case where three types of evaluation values, safety level FD, security level RD, and overall level TD, are applied as evaluation values is described, but this is not limited to this. For example, only one type or a combination of two types of these may be applied as evaluation values.

In addition, in the above embodiment, the spatial evaluation support process is executed in the spatial evaluation support device 10, but this is not limited to the above. For example, the spatial evaluation support process may be executed by each subject terminal 30. In this case, the spatial evaluation support device 10 of the present invention is included in the subject terminal 30.

In the above embodiment, a case has been described in which an image showing a map that highlights locations with a lot of information that matches the target person's category is applied as the social heat map image, but the present invention is not limited to this. For example, an image in which the redder the area is, the higher at least one of the safety level FD, the security level RD, and the overall level TD, may be applied as the social heat map image.

In the above embodiment, the input of the processing target area is performed on the initial screen using the input unit 34, but this is not limited to the above. For example, the processing target area may be input as voice information using the microphone 39.

In addition, in the above embodiment, no particular mention is made of what happens after the subject refers to the evaluation result screen, but if the evaluation value displayed on the evaluation result screen is different from the subject's own subjective evaluation, the subject's own subjective evaluation may be reflected in the evaluation value information database 13F using the subject's terminal 30.

In the above embodiment, a case has been described in which components in a captured image are extracted using a segmentation model, but this is not limiting. For example, components in a captured image may be extracted by applying an object detection model other than a segmentation model. Note that the following can be given as examples of the method of constructing the model in this case.

First, through experiments and workshops with test subjects, object categories such as trash cans that subjectively compromise safety and security are determined. Next, a large number of examples of the determined object categories are collected through surveys or other methods. The collected data is then used as training data to train an object detection model.

In addition, by using other conventionally known artificial intelligence instead of an object detection model, the safety and security levels may be derived directly from the captured image, rather than extracting components in the captured image. In this case, the learning method of the artificial intelligence may be exemplified as follows.

First, subjects are asked to subjectively evaluate the safety and security levels of a number of existing photographed images taken during experiments, workshops, etc., using multiple scores or multiple ratings such as good, average, and bad. Then, these evaluation results and the corresponding photographed images are used as training data to train the above-mentioned artificial intelligence.

In addition, in the above embodiment, no particular reference was made to the method of determining the weighting values in the weighting information database 13E, but, for example, the weighting values may be determined from the results of a questionnaire survey conducted by multiple people.

It goes without saying that the calculation formulas (1) to (5) are merely examples and can be modified as appropriate without departing from the spirit of the present invention.

In the above embodiment, for example, the hardware structure of the processing unit that executes each process of the acquisition unit 11A, extraction unit 11B, derivation unit 11C, and presentation unit 11D can be the various processors shown below. As described above, the various processors include a CPU, which is a general-purpose processor that executes software (programs) and functions as a processing unit, as well as a programmable logic device (PLD), which is a processor whose circuit configuration can be changed after manufacture, such as an FPGA (Field-Programmable Gate Array), a dedicated electrical circuit, which is a processor with a circuit configuration designed specifically to execute a specific process, such as an ASIC (Application Specific Integrated Circuit), etc.

The processing unit may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (e.g., a combination of multiple FPGAs, or a combination of a CPU and an FPGA). The processing unit may also be configured with a single processor.

As an example of configuring the processing unit as a single processor, first, there is a form in which one processor is configured as a combination of one or more CPUs and software, as typified by computers such as clients and servers, and this processor functions as the processing unit. Secondly, there is a form in which a processor is used to realize the functions of the entire system, including the processing unit, in a single IC (Integrated Circuit) chip, as typified by systems on chips (SoCs). In this way, the processing unit is configured as a hardware structure using one or more of the various processors listed above.

More specifically, the hardware structure of these various processors can be an electrical circuit that combines circuit elements such as semiconductor elements.

10 space evaluation support device 11 CPU
Reference Signs List 11A Acquisition unit 11B Extraction unit 11C Derivation unit 11D Presentation unit 12 Memory 13 Storage unit 13A Spatial evaluation support program 13B Evaluation value presentation program 13C Segmentation model 13D Image information database 13E Weighting information database 13F Evaluation value information database 14 Input unit 15 Display unit 16 Medium reading and writing device 17 Recording medium 18 Communication I/F unit 30 Subject terminal 31 CPU
31A Receiving unit 31B Display control unit 32 Memory 33 Storage unit 33A Evaluation value display program 34 Input unit 35 Display unit 36 Medium reading and writing device 37 Recording medium 38 Camera 39 Microphone 40 GPS
42 Wireless communication unit 80 Network 90 Space evaluation support system

Claims (9)

An acquisition unit that acquires a captured image of a space to be evaluated;
an extraction unit that extracts components in the space from the captured image acquired by the acquisition unit;
a derivation unit that performs a weighting process for each type of the component extracted by the extraction unit, and derives an evaluation value related to at least one of safety and security for the space;
Equipped with
The components are classified into components for each of the safety and the security,
The derivation unit derives the evaluation value regarding safety using an arithmetic expression in which an actual amount of the corresponding component in the space to be evaluated is divided by a maximum value of the component, and derives the evaluation value regarding security using an arithmetic expression in which a value obtained by dividing the actual amount of the corresponding component in the space to be evaluated by a maximum value of the component is subtracted from 1 .
Spatial evaluation support device. a presentation unit that combines the evaluation value derived by the derivation unit with a map image and presents the combined evaluation value;
The spatial evaluation support device according to claim 1 , further comprising: The map image is a social heatmap image.
The space evaluation support device according to claim 2. The extraction unit extracts the components using a segmentation model trained using the captured image acquired by the acquisition unit.
The space evaluation support device according to any one of claims 1 to 3. The components include at least one of a sidewalk, a roadway, a pedestrian bridge, a flyer, graffiti, and a parked vehicle;
The space evaluation support device according to any one of claims 1 to 4. The weighting is performed on at least one type of the components.
The space evaluation support device according to claim 5. The captured image is an omnidirectional image.
The space evaluation support device according to any one of claims 1 to 6. A space evaluation support device according to any one of claims 1 to 7,
a terminal including a receiving unit that receives the evaluation value derived by the derivation unit of the spatial evaluation support device from the spatial evaluation support device, and a display control unit that controls a display unit to display information indicating the evaluation value received by the receiving unit;
A spatial evaluation support system including: Acquire an image of the space to be evaluated,
Extracting components in the space from the acquired photographed image;
A predetermined weight is applied to the extracted components for each type of the components, and an evaluation value regarding at least one of safety and security for the space is derived.
Processing,
The components are classified into components for each of the safety and the security,
The evaluation value regarding safety is derived using an arithmetic expression in which an actual amount of the corresponding component in the space to be evaluated is divided by a maximum value of the component, and the evaluation value regarding peace of mind is derived using an arithmetic expression in which a value obtained by dividing an actual amount of the corresponding component in the space to be evaluated by a maximum value of the component is subtracted from 1 .
A spatial evaluation support program for executing processing on a computer.