3D rendering system, 3D rendering method and program

Abstract

A three-dimensionalization system according to the present disclosure comprises: an acquisition means that acquires a plurality of images captured by imaging devices each installed on each of a plurality of moving bodies; a selection means that selects at least two images capturing a road surface of a road or a structure on the road from among the plurality of acquired images on the basis of a predetermined condition; and a generation means that uses the selected images to generate a three-dimensional model of the captured road surface of the road or the structure on the road.

Description

[Technical Field]

[0001] The present disclosure relates to a three-dimensional rendering system and the like. [Background technology]

[0002] Roads and structures located around them are subject to deterioration over time and damage caused by accidents, making repair work necessary. 3D data of roads and structures is sometimes used to support repair work planning.

[0003] Patent Document 1 discloses a 3D model construction system that collects image data captured by image capture devices provided on multiple moving objects and generates a 3D model. In Patent Document 1, the image capture area of ​​an existing 3D model is identified using supplementary information related to the image capture included in the image capture data, and the existing 3D model is updated using new image capture data for the image capture area. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-177317 [Patent Document 2] Japanese Patent Application Publication No. 2019-164018 [Patent Document 3] Japanese Patent Application Publication No. 2018-119927 Summary of the Invention [Problem to be solved by the invention]

[0005] According to Patent Document 1, it is possible to generate 3D models of large fixed objects such as roads, buildings, bridges, etc. However, simply collecting images of the same area makes it difficult to generate 3D models that can identify the deterioration state of small objects such as road surfaces or structures on roads.

[0006] The present disclosure aims to provide a 3D rendering system and the like that enables the generation of highly accurate 3D models. [Means for solving the problem]

[0007] The 3D system according to the present disclosure includes an acquisition means for acquiring a plurality of images captured by imaging devices installed on each of a plurality of moving bodies, a selection means for selecting at least two images of a road surface or a structure on the road from the acquired plurality of images based on predetermined conditions, and a generation means for generating a 3D model of the photographed road surface or structure on the road using the selected images.

[0008] The three-dimensionalization method disclosed herein acquires multiple images captured by imaging devices installed on multiple moving bodies, selects at least two images of the road surface or structures on the road from the acquired multiple images based on predetermined conditions, and generates a three-dimensional model of the photographed road surface or structures on the road using the selected images.

[0009] A program according to the present disclosure causes a computer to execute a process of acquiring a plurality of images captured by imaging devices installed on a plurality of moving bodies, selecting at least two images of a road surface or a structure on a road from the acquired plurality of images based on predetermined conditions, and generating a 3D model of the captured road surface or structure on the road using the selected images. The program may be stored in a computer-readable non-transitory recording medium. [Effects of the Invention]

[0010] According to the present disclosure, it is possible to generate a highly accurate three-dimensional model. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram showing an overview of devices connected to a 3D rendering system. [Figure 2]1 is a block diagram showing an example of the configuration of a three-dimensional conversion system according to a first embodiment. [Figure 3] 4 is a flowchart showing an example of the operation of the three-dimensional conversion system according to the first embodiment. [Figure 4] 10 is a table showing an example of information included in the imaging data. [Figure 5] FIG. 2 is a block diagram illustrating an example of a hardware configuration of a computer. DETAILED DESCRIPTION OF THE INVENTION

[0012] The surface of a paved road can deteriorate due to factors such as vehicle traffic and rainfall, resulting in cracks, potholes, ruts, and other damage. Road structures, such as signs, lighting, guardrails, and curbs, can also deteriorate and become damaged. Therefore, road conditions are analyzed to understand the deterioration of roads and structures and to plan repairs for them.

[0013] The 3D modeling system of the present disclosure generates a 3D model of a road surface or a structure on a road using images selected based on predetermined conditions from images captured by imaging devices installed on multiple mobile objects.

[0014] The roads targeted by the 3D rendering system according to the present disclosure are not limited to roads on which vehicles pass, but also include roads on which people pass. Furthermore, the range targeted by the 3D rendering system for 3D rendering is not limited to the road itself, but also includes road slopes and land on which structures necessary for road management exist.

[0015] 1 is a diagram showing an overview of devices communicably connected to a three-dimensional rendering system 100 via a communication network 30, either wired or wirelessly. The three-dimensional rendering system 100 is connected to, for example, an imaging device 10, a display 20, an input device 21, and a database 40.

[0016] The imaging device 10 is installed on a moving object 11 and captures images including roads or structures on the roads. The imaging device 10 is realized, for example, by a drive recorder mounted on an automobile. However, the type of imaging device 10 is not limited to this, and cameras mounted on various types of moving objects 11 may be used. For example, images may be captured by a camera mounted on another moving object such as a bicycle or a drone.

[0017] The images captured by the imaging device 10 may be still images or moving images captured while the moving object 11 is moving. The images may be captured at a location designated by a person, or may be captured automatically at any interval.

[0018] 1 shows one imaging device 10 and one moving body 11. However, the 3D generation system 100 may be connected to multiple imaging devices 10-1, ..., 10-n installed on multiple moving bodies 11-1, ..., 11-n, respectively. Here, n is a natural number greater than or equal to 2. The multiple moving bodies 11 may be of the same type or different types. The multiple imaging devices 10 may be of the same model or different models.

[0019] The imaging data including the images captured by the imaging device 10 is stored in the database 40. The imaging device 10 may also transmit the imaging data including the images to the three-dimensional rendering system 100.

[0020] The imaging data may further include the following imaging conditions for the image. For example, the imaging data may include an identifier for identifying the imaging device 10 that captured the image. The imaging data may also include location information for the location where the image was captured. The location information may include, for example, latitude and longitude, location information from a Global Navigation Satellite System (GNSS) or a Global Positioning System (GPS), or a location on a map. The imaging data may also include time information regarding the date and time when the image was captured.

[0021] Furthermore, the imaging data may include the imaging direction in which the image was captured. The imaging direction includes, for example, the orientation or elevation / depression angle of the imaging device 10. The imaging direction can be acquired by a sensor provided in the imaging device 10. Furthermore, when the installation direction of the imaging device 10 relative to the moving object 11 is specified, the imaging direction can be acquired based on the traveling direction of the moving object 11.

[0022] The display 20 displays information to the user. The display 20 includes, for example, a display, a tablet, etc. The information displayed will be described later.

[0023] The input device 21 receives operations from a user. The input device 21 includes, for example, a mouse, a keyboard, etc. If the display 20 is a touch panel display, the display 20 may be configured as the input device 21.

[0024] The database 40 stores image data including images captured by the image capture device 10 .

[0025] [First embodiment] 2 is a block diagram showing an example of the configuration of a three-dimensional rendering system 100 according to the first embodiment. The three-dimensional rendering system 100 includes an acquisition unit 110, a selection unit 120, and a generation unit 130. The three-dimensional rendering system 100 further includes an output unit 140 as necessary.

[0026] The acquisition unit 110 acquires a plurality of images including images of predetermined points captured by the imaging devices 10 installed on each of the plurality of moving bodies 11. The predetermined points are points from which a three-dimensional model is to be generated. The range of points from which the acquisition unit 110 acquires images can be set as appropriate.

[0027] In one example, the acquisition unit 110 may acquire a video captured at a predetermined location. Alternatively, the acquisition unit 110 may extract and acquire a still image captured at the predetermined location from each of a plurality of videos.

[0028] The acquisition unit 110 may acquire shooting data including an image. That is, the acquisition unit 110 may acquire, along with the image, location information of the location where the image was captured, the date and time of capture, model information of the imaging device 10, and the like.

[0029] The selection unit 120 selects at least two images of the road surface or structures on the road from the plurality of acquired images based on predetermined conditions.

[0030] The predetermined conditions are conditions for selecting images suitable for generating a 3D model of the road surface or a structure on the road. A 3D model is data that represents the 3D shape and size of an object. The 3D model is, for example, 3D point cloud information. Images suitable for generating a 3D model are at least two images that can sufficiently obtain the information required to calculate the 3D shape and size of the object. Images suitable for generating a 3D model include images that are expected to have parallax relative to the object.

[0031] The road surface may have depressions and protrusions due to road deterioration. The selection unit 120 may select images suitable for estimating the depth of road deterioration as images suitable for generating a 3D model of the road surface. Road deterioration includes, for example, cracks, potholes, and ruts.

[0032] Road structures are objects installed on roads near where vehicles and people pass, and include, for example, signs, lights, guardrails, curbs, etc.

[0033] For example, the selection unit 120 selects at least two images based on the acquired photographic data and predetermined conditions. Details of the predetermined conditions will be described in the second embodiment.

[0034] The selection unit 120 may select images that are taken at the same location and that satisfy the predetermined condition from the images acquired by the acquisition unit 110. For example, the selection unit 120 may extract images that are taken at the same location by referring to location information in the shooting data. Alternatively, the selection unit 120 may extract images that are taken at the same location by estimating the location at which the images were taken based on image features. Then, the selection unit 120 may select at least two images that satisfy the predetermined condition from the extracted images. Alternatively, the selection unit 120 may extract images that satisfy the predetermined condition and then select images that are taken at the same location.

[0035] The generation unit 130 generates a 3D model of the captured road surface or structures on the road using at least two images selected by the selection unit 120. For example, the generation unit 130 acquires parameters necessary for processing the parallax of the selected images. The necessary parameters are, for example, the distance between the image capture devices 10 and the focal length of the image capture devices 10. Then, the generation unit 130 generates the 3D model by calculating the distance from the image capture devices 10 to the captured object based on the acquired parameters and the parallax of the selected images.

[0036] The generating unit 130 may generate a 3D model of the road surface that indicates the depth of road deterioration, including the depth of cracks, the depth of potholes, and the amount of rutting.

[0037] The output unit 140 outputs information to the display 20 based on the generated three-dimensional model. For example, the output unit 140 may display the three-dimensional model on the display 20. The output unit 140 may also output a value of the depth of road deterioration. The depth of road deterioration is calculated, for example, based on the change in depth of a portion of the three-dimensional model cut out in stripes in the road width direction.

[0038] 3 is a flowchart showing an example of the operation of the three-dimensional rendering system 100 according to the first embodiment. The three-dimensional rendering system 100 may start the operation of FIG. 3 in response to a user's operation using the input device 21.

[0039] The acquisition unit 110 acquires a plurality of images captured by the imaging devices 10 installed in the plurality of moving bodies 11 (step S11).

[0040] The selection unit 120 selects at least two images of the road surface or structures on the road from the plurality of images acquired by the acquisition unit 110 based on predetermined conditions (step S12).

[0041] The generating unit 130 generates a three-dimensional model of the photographed road surface or structures on the road, using the image selected by the selecting unit 120 (step S13).

[0042] The output unit 140 outputs information to the display 20 based on the generated three-dimensional model (step S14).

[0043] According to the first embodiment, the acquisition unit 110 acquires a plurality of images captured by the imaging devices 10 installed on each of a plurality of moving bodies 11. Then, the selection unit 120 selects at least two images of a road surface or a structure on a road from the plurality of images acquired by the acquisition unit 110 based on predetermined conditions. The generation unit 130 generates a 3D model of the captured road surface or structure on the road using the images selected by the selection unit 120. Since images suitable for generating a 3D model are selected based on predetermined conditions, the first embodiment enables the generation of a highly accurate 3D model.

[0044] To measure road surface conditions, a stereo camera or a light-section imaging device that emits a slit laser is installed on a mobile object 11. Patent Document 2 discloses an imaging system that creates three-dimensional road surface data based on images captured by the stereo camera. However, the stereo camera and the light-section imaging device are expensive. According to the first embodiment, road conditions can be analyzed using images captured by a monocular camera such as a drive recorder. Therefore, according to the first embodiment, images required for a three-dimensional model can be collected at low cost.

[0045] Patent Document 3 discloses an image processing device that improves the convenience of distance measurement technology using a monocular camera based on the motion stereo method. It has been difficult to estimate the depth of road deterioration from images captured by a monocular camera such as a drive recorder. Estimating the depth of cracks in the road surface is particularly difficult because they are small. To analyze the depth of road deterioration, it is necessary to accurately convert the image into three dimensions. According to a first embodiment, images suitable for estimating the depth of road deterioration are selected from images captured by imaging devices 10 installed on multiple mobile objects 11. Then, using the selected images, a generation unit 130 generates a three-dimensional model representing the depth of road deterioration. Therefore, according to the first embodiment, it is possible to support the creation of repair plans based on the depth of road deterioration, such as cracks, potholes, or ruts, while reducing costs.

[0046] [Second embodiment] Next, as a second embodiment, a more detailed description will be given of the three-dimensional rendering system 100. Regarding the configuration of the second embodiment, the description of the same configuration as the first embodiment will be omitted.

[0047] In the second embodiment, the acquisition unit 110 acquires shooting data including an image captured at a predetermined location. FIG. 4 is a table showing an example of information included in the shooting data. The shooting data in FIG. 4 includes an image ID (identifier), an imaging device ID, time information, location information, and shooting direction. The image ID is an identifier that identifies an image. The image ID may be an identifier that identifies one frame of a video. The imaging device ID is an identifier that identifies the imaging device 10 that captured the image. The shooting data does not need to include the information shown in FIG. 4, and may include information other than the information shown.

[0048] Here, examples of predetermined conditions for selecting at least two images suitable for generating a 3D model of a road surface or a structure on a road will be described. The selection unit 120 may select images based on any combination of the conditions described below.

[0049] In one example, the predetermined condition is a condition related to the similarity of the images. The selection unit 120 may select an image from the plurality of images based on the similarity of the images. The similarity of the images may be calculated by any method. The similarity between images of the same object taken from similar viewpoints at the same location is higher than the similarity between images of different objects taken from different viewpoints at the same location.

[0050] For example, the selection unit 120 selects at least two images whose similarity is higher than a threshold (lower limit). Two images whose similarity is higher than the threshold have the same imaging range and are likely to depict the same object. Therefore, it is possible to calculate the distance based on the parallax of the object. If the similarity between images is low, the difference in imaging range may be too large, resulting in insufficient parallax information and making it difficult to calculate the distance.

[0051] Depending on the lighting conditions, the brightness and color of the images may differ. Therefore, in order to estimate the difference between the image capture ranges from the similarity, the brightness and color of the images may be converted before the similarity of the images is calculated. Note that, if it is preferable to select images captured under the same lighting conditions, the similarity is calculated as is without converting the images.

[0052] Alternatively, the selection unit 120 may select images such that at least two images have a similarity lower than the upper threshold. This is because distance calculation becomes difficult when the similarity between images is high and the overlap of the imaging ranges is too large. By the selection unit 120 selecting images with a similarity lower than the upper threshold, images captured by the imaging devices 10 of the multiple moving objects 11 from different viewpoints can be selected.

[0053] In another example, the predetermined condition is a condition related to the imaging device 10 that captured the image. The selection unit 120 may select images to include images captured by different imaging devices 10 installed in different moving objects 11. In this case, the selection unit 120 refers to the imaging device ID of the image capture data, for example. By selecting images captured by imaging devices 10 installed in different moving objects 11, it is expected that a larger parallax can be obtained than images captured by the same moving object 11 passing through a road multiple times. This enables more accurate distance calculation.

[0054] However, the selection unit 120 may select at least two images captured by the imaging device 10 when one moving object 11 travels through a predetermined location multiple times. This is because disparity information can be obtained between at least two images when the traveling positions on the road at the predetermined location are different for the first and second travels. That is, the selection unit 120 may select at least two images captured by the imaging device 10 installed on the same moving object 11 at dates and times that are different by at least a predetermined time. The selection unit 120 may also exclude from the selection targets at least two images captured continuously by the imaging device 10 while one moving object 11 is traveling. Disparity information can only be obtained from two continuously captured images corresponding to one frame of movement of the moving object 11. Therefore, continuously captured images may not be suitable for generating a 3D model.

[0055] In another example, the predetermined condition is a condition related to the installation state of the imaging device 10. The installation state of the imaging device 10 includes the installation height, the installation angle or left / right position relative to the moving body 11, and the type of moving body 11 on which the imaging device 10 is installed. The selection unit 120 may select images based on the installation state of the imaging device 10. For example, the selection unit 120 selects images captured by imaging devices 10 with different installation states. This allows the selection unit 120 to select images with parallax. The installation state of the imaging device 10 is stored in the database 40 in association with the imaging device ID.

[0056] The installation height may be expressed as the height from the ground of the imaging device 10 installed on the moving body 11. Alternatively, the installation height may be expressed as the distance of the imaging device from a predetermined member of the moving body 11 as a reference.

[0057] The installation angle indicates the angle at which the imaging device 10 is installed to capture images. The installation angle may be expressed as an elevation angle or depression angle. The installation angle may also be expressed as the orientation of the imaging device 10 relative to the front direction of the moving object 11.

[0058] The type of the installed mobile object 11 may be specified by the size or model of the automobile that is the mobile object 11. The type of the mobile object 11 may also be specified by whether the mobile object 11 is a four-wheeled automobile, a two-wheeled automobile, a bicycle, or a drone. For example, the selection unit 120 may select images captured by imaging devices 10 installed in a standard automobile and a bus, respectively. Since standard automobiles and buses are different sizes, images from different viewpoints can be selected from the imaging devices 10 installed in each.

[0059] Furthermore, the installation state of the imaging device 10 may be expressed as a left-right position relative to the moving body 11. The left-right position indicates, for example, the position in the width direction of the vehicle, which is the moving body 11, at which the imaging device 10 is installed. The left-right position may also be expressed as a distance from the center of the vehicle.

[0060] In another example, the predetermined condition is a condition related to the shooting direction included in the shooting data. The shooting direction is the azimuth or elevation / depression angle of the imaging device 10, which can be acquired by a sensor provided in the imaging device 10. The selection unit 120 may select images based on such shooting directions. For example, the selection unit 120 selects images shot in different directions. This allows the selection unit 120 to select images with parallax.

[0061] As another example, the predetermined condition may be a condition regarding the shooting direction of an object included in the image. The selection unit 120 may select an image based on the shooting direction of the object identified by recognizing the object through image analysis. The shooting direction of the object indicates the positional relationship between the image capture device 10 and the object included in the image. Objects included in the image include, for example, road deterioration such as cracks and potholes, road markings such as lane markings, and structures on the road.

[0062] The selection unit 120 may select images of the crack taken from multiple directions based on the shooting direction relative to the crack. This allows the selection unit 120 to select images suitable for accurately estimating the depth of the crack. The shooting direction relative to the lane markings allows the traveling position of the mobile object 11 within the lane to be determined. By selecting images based on the shooting direction relative to the lane markings, the selection unit 120 can select images taken while traveling at different positions within the lane. Therefore, the selection unit 120 can select images with parallax.

[0063] As another example, the predetermined condition may be a condition related to the bias of the captured image data. It is expected that if the selection unit 120 selects more images, the generation unit 130 will be able to generate a more accurate 3D model using the selected images. However, in the present disclosure, images automatically captured by a drive recorder of a vehicle traveling daily can be used, so many similar images may be captured for a given location. Even if many similar images are selected, the accuracy of the 3D model will not improve. Therefore, the selection unit 120 may select images captured from various viewpoints in a balanced manner based on the bias of the captured image data as follows:

[0064] For example, the selection unit 120 may select images so as to reduce the bias in the similarity between the images. The selection unit 120 may select images so as to reduce the bias between images with high similarity and images with low similarity.

[0065] The selection unit 120 may also select images to reduce bias in installation conditions. For example, when there is a lot of image data captured by an imaging device 10 installed in a standard automobile, there will be a lot of images captured from a low viewpoint. Therefore, the selection unit 120 uniformly selects images captured from a higher viewpoint by an imaging device 10 installed in a bus, a garbage truck, or the like.

[0066] In another example, the predetermined condition is a condition regarding the presence or absence of road deterioration on the road in the image. The selection unit 120 may select images in which road deterioration can be detected. Furthermore, the selection unit 120 may select more images for areas with road deterioration than for areas without road deterioration based on the presence or absence of road deterioration. This allows the generation unit 130 to represent the depth of road deterioration with higher accuracy using more images.

[0067] Here, a specific example of a combination of multiple conditions will be described. For example, the selection unit 120 selects multiple images from images taken at a specific location from which road deterioration can be detected. Then, the selection unit 120 further selects images suitable for generating a 3D model of road deterioration based on the similarity of the selected images. Therefore, the selection unit 120 selects at least two similar images from among images that clearly show road deterioration. This allows the generation unit 130 to generate a 3D model of road deterioration of the road surface with high accuracy.

[0068] Alternatively, the selection unit 120 may select images based on various information. For example, the selection unit 120 may select images based on time information of the shooting data. The selection unit 120 may select images taken during the same time period. The selection unit 120 may also select images taken during a time period when there are fewer shadows cast by road deterioration, such as during the day. The selection unit 120 may further select images taken under the same weather conditions based on weather information for the date and time the images were taken. The selection unit 120 may also select images taken by the same model of imaging device 10.

[0069] The above describes the predetermined conditions under which the selection unit 120 selects an image suitable for generating a 3D model of the road surface or a structure on the road. Next, the generation of a 3D model by the generation unit 130 will be described.

[0070] The generation unit 130 performs a matching process for matching corresponding points of the images selected by the selection unit 120. The generation unit 130 uses an arbitrary matching algorithm to find corresponding pixels in one image for a reference pixel in the other image. For example, the generation unit 130 extracts feature points from the images and matches the feature points between the images. The generation unit 130 may perform matching by converting the brightness and color of the images to eliminate the influence of sunlight conditions. The generation unit 130 may use information on the shooting direction and installation status included in the shooting data to match corresponding points.

[0071] The generation unit 130 may detect areas of road deterioration and perform a matching process on the detected areas of road deterioration. The generation unit 130 may detect road deterioration using a known image recognition technology on the image. The generation unit 130 may detect road deterioration using a trained model. The generation unit 130 may determine whether or not each pixel in the image is road deteriorated. Then, the generation unit 130 associates points representing road deterioration between the images. For example, the generation unit 130 may detect crack areas in each of two images and determine corresponding pixels between each detected crack.

[0072] The generation unit 130 may detect road areas and perform matching processing on the detected road areas, thereby preventing the generation unit 130 from matching, for example, a road area in one image with a building area in another image.

[0073] The generation unit 130 then calculates the three-dimensional coordinates of the points associated between the images based on the imaging data. For example, the generation unit 130 calculates the distance using the principle of triangulation based on the focal length of the imaging device 10, the parallax between the matched reference pixel and the corresponding pixel, and the distance between the imaging devices 10 that captured the images. The generation unit 130 generates three-dimensional point cloud information based on the distance calculated for each pixel. The distance between the imaging devices 10 that captured the images is a parameter required for processing the parallax of the images, and is also referred to as the baseline length.

[0074] The generation unit 130 calculates the baseline length based on the shooting conditions included in the shooting data, such as the installation state of the imaging device 10 and the shooting direction of the image. The generation unit 130 may also calculate the baseline length based on the position of the imaging device 10 on the road at the time of shooting, which is estimated from the image. The position on the road is estimated from the appearance of a reference object in the image. The left and right positions on the road are estimated based on the appearance of lane markings, for example.

[0075] The generation unit 130 may convert the 3D point cloud information so that it is easier to view. For example, the generation unit 130 performs texture mapping by attaching an area surrounded by three neighboring feature points of the image to three points of the 3D coordinates formed by the three feature points. This allows the generation unit 130 to generate a 3D model made up of triangular polygon surfaces.

[0076] According to the second embodiment, like the first embodiment, it is possible to generate a highly accurate three-dimensional model.

[0077] [Hardware configuration] In each of the above-described embodiments, each component of the three-dimensional rendering system 100 represents a functional block. Some or all of the components of the three-dimensional rendering system 100 may be realized by any combination of a computer 500 and a program.

[0078] Fig. 5 is a block diagram showing an example of the hardware configuration of a computer 500. Referring to Fig. 5, the computer 500 includes, for example, a processor 501, a read only memory (ROM) 502, a random access memory (RAM) 503, a program 504, a storage device 505, a drive device 507, a communication interface 508, an input device 509, an input / output interface 511, and a bus 512.

[0079] The processor 501 controls the entire computer 500. The processor 501 may be, for example, a CPU (Central Processing Unit). The number of processors 501 is not particularly limited, and there may be one or more processors 501.

[0080] The program 504 includes instructions for implementing each function of the 3D rendering system 100. The program 504 is stored in advance in the ROM 502, RAM 503, or storage device 505. The processor 501 executes the instructions included in the program 504 to implement each function of the 3D rendering system 100. The RAM 503 may also store data to be processed in each function of the 3D rendering system 100. For example, captured images may be stored in the RAM 503 of the computer 500.

[0081] The drive device 507 reads and writes data from and to the recording medium 506. The communication interface 508 provides an interface with a communication network. The input device 509 is, for example, a mouse or a keyboard, and receives information input from an administrator or the like. The output device 510 is, for example, a display, and outputs (displays) information to an administrator or the like. The input / output interface 511 provides an interface with peripheral devices. The bus 512 connects these hardware components. The program 504 may be supplied to the processor 501 via a communication network, or may be stored in advance on the recording medium 506, read by the drive device 507, and supplied to the processor 501.

[0082] It should be noted that the hardware configuration shown in FIG. 5 is an example, and other components may be added, or some components may not be included.

[0083] There are various variations in the method for realizing the 3D rendering system 100. For example, the 3D rendering system 100 may be realized by any combination of different computers and programs for each component. Furthermore, multiple components included in the 3D rendering system 100 may be realized by any combination of a single computer and program.

[0084] Furthermore, at least a part of the three-dimensional rendering system 100 may be provided in a SaaS (Software as a Service) format. That is, at least a part of the functions for realizing the three-dimensional rendering system 100 may be executed by software executed via a network.

[0085] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, the configurations in the respective embodiments can be combined with each other without departing from the scope of the present disclosure. [Explanation of symbols]

[0086] 100 3D System 110 Acquisition Department 120 Selection Section 130 Generation part 140 Output section 10. Imaging device 20 Display 21 Input Devices 30 Communication Network 40 databases

Claims

1. an acquisition means for acquiring a plurality of images captured by imaging devices installed in each of a plurality of moving bodies; a selection means for selecting at least two images of the road surface from the plurality of acquired images based on predetermined conditions for selecting images to be used for generating a three-dimensional model showing the depth of road deterioration; a generating means for generating a three-dimensional model showing the depth of road deterioration of the road surface of the photographed road using the selected image; Equipped with The predetermined conditions include a condition for selecting images so as to include images captured by imaging devices respectively installed on different moving bodies, and a condition for selecting images in which road deterioration can be detected. 3D system.

2. The selection means selects the images based on the similarity of the images. The three-dimensionalization system according to claim 1 .

3. The selection means selects the image based on an installation state of the imaging device.

3. The three-dimensionalization system according to claim 1.

4. The selection means selects the image based on a photographing direction of an object included in the image.

3. The three-dimensionalization system according to claim 1.

5. The selection means selects the image based on the type of the moving body in which the imaging device is installed.

3. The three-dimensionalization system according to claim 1.

6. The selection means selects the images so as to reduce bias in the similarity of the images. The three-dimensionalization system according to claim 2 .

7. Acquire a plurality of images captured by imaging devices installed in each of a plurality of moving bodies; selecting at least two images of the road surface from the plurality of acquired images based on predetermined conditions for selecting images to be used in generating a three-dimensional model showing the depth of road deterioration; generating a three-dimensional model showing the depth of road deterioration of the road surface of the photographed road using the selected images; The predetermined conditions include a condition for selecting images so as to include images captured by imaging devices respectively installed on different moving bodies, and a condition for selecting images in which road deterioration can be detected. 3D method.

8. Acquire a plurality of images captured by imaging devices installed in each of a plurality of moving bodies; selecting at least two images of the road surface from the plurality of acquired images based on predetermined conditions for selecting images to be used in generating a three-dimensional model showing the depth of road deterioration; The selected images are used to generate a three-dimensional model showing the depth of road deterioration of the road surface of the photographed road. Have the computer execute the process, The predetermined conditions include a condition for selecting images so as to include images captured by imaging devices respectively installed on different moving bodies, and a condition for selecting images in which road deterioration can be detected. program.

Images