3D model generation method, construction method, and 3D model generation device

Abstract

To improve the measurement accuracy of the shape of an upstream surface of a dam embankment.SOLUTION: A 3D model generation method includes: an image acquisition step of acquiring multiple images in which the shape of an upstream surface of a dam embankment is captured by an imaging device mounted on a mobile body; a 3D model generation step of generating a 3D model representing the shape by determining a three-dimensional shape corresponding to the shape shown in the images, based on the parallax of the multiple images acquired in the image acquisition step; and an output step of outputting the result of the generated 3D model.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a 3D model generation method, a construction method, and a 3D model generation device.

Background Art

[0002] For "dam redevelopment" aimed at strengthening or maintaining the dam function, existing dams are being repaired, such as modifying the dam embankment and enhancing the discharge function. When performing the work to enhance the discharge function, it is often the case that the work is carried out while operating the reservoir of the dam, and in many cases, it is impossible to lower the water level to the elevation of the facility to be added. Therefore, after installing a temporary cutoff by underwater work and draining the water, the work is often carried out (see, for example, Patent Document 1). However, while the standard value (allowable range in the work) of the width (thickness) of the dam embankment is from -30 [mm] to +50 [mm], the size of the unevenness (undulation) that can be accommodated in the work of installing the temporary cutoff is about ±10 [mm]. Therefore, a gap may occur between the temporary cutoff and the surface of the dam embankment, and water may flow in. Therefore, when installing the temporary cutoff, divers measure the size of the unevenness of the portion where the temporary cutoff is to be installed on the upstream surface of the dam embankment and perform the work of flattening the unevenness in order to keep it within the range that can be accommodated in the work of installing the temporary cutoff.

[0003] In addition, Patent Document 2 discloses a technique for measuring the size and shape of unevenness using a sonar.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, when manually measuring the shape of the surface on the upstream side of the dam embankment, since a diver dives underwater in the dam and measures at predetermined intervals along the height direction of the dam embankment, it is impossible to obtain measurement results for the intervals between the measurement points. In addition, the measurement of the shape of the surface on the upstream side of the dam embankment by a diver places a heavy burden on the diver and requires management to prevent decompression sickness. Also, the measurement using a sonar has a large measurement error and is difficult to use when measuring the size of the unevenness of the dam embankment.

[0006] Therefore, the present invention has been made in view of the above points, and an object thereof is to improve the measurement accuracy of the shape of the surface on the upstream side of the dam embankment.

Means for Solving the Problems

[0007] [1] One aspect of the present invention is an image acquisition step of acquiring a plurality of images of the shape of the surface on the upstream side of the dam embankment captured by an imaging device provided in a moving body, and based on the parallax of the plurality of images acquired in the image acquisition step, a 3D model generation step of generating a 3D model representing the shape by obtaining a three-dimensional shape corresponding to the shape represented in the image, and an output step of outputting the result of the generated 3D model, which is a 3D model generation method.

[0008] [2] Further, one aspect of the present invention is a specifying step of specifying a target position for unevenness correction of the surface on the upstream side of the dam embankment based on the 3D model generated by the 3D model generation method described in [1], a construction step of performing unevenness correction at the specified target position, and after the unevenness correction is performed, an image acquisition step of acquiring a plurality of images such that at least one of the imaging position and the imaging direction is different from the shape of the surface on the upstream side of the dam embankment, and a 3D model generation step of newly generating the 3D model based on the plurality of images captured in the image acquisition step. When it is determined that there is a target position for unevenness irregularity based on the newly generated 3D model, it is a construction method that repeatedly performs from the construction step.

[0009] [3] Further, one aspect of the present invention includes an image acquisition unit that acquires a plurality of images of the shape of the surface on the upstream side of the dam embankment captured by an imaging device provided in the moving body, and based on the parallax of the plurality of images acquired by the image acquisition unit, obtains a three-dimensional shape corresponding to the shape represented in the images, thereby generating a 3D model representing the shape, and an output unit that outputs the result of the generated 3D model. It is a 3D model generation device comprising:

Effect of the Invention

[0010] According to the present invention, the measurement accuracy of the shape of the surface on the upstream side of the dam embankment can be improved.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Modes for Carrying Out the Invention

[0012] [Embodiment] Regarding the 3D model generation method, construction method, and 3D model generation device according to this embodiment, preferred embodiments will be described in detail below with reference to the accompanying drawings. Note that this embodiment is not limited to these embodiments, but also includes those with various modifications or improvements. That is, the components described below include those that can be easily assumed by those skilled in the art and substantially identical ones, and the components described below can be combined as appropriate. Also, in this embodiment, various omissions, substitutions, or changes of the components can be made without departing from the gist of the present invention.

[0013] FIG. 1 is a diagram for explaining an example of a temporary cut-off installed on a dam embankment. Among the "dam redevelopment" works aimed at enhancing or maintaining the dam function, for the works to enhance the discharge function, in addition to the works to renovate the existing discharge facilities, there are works to add discharge pipes or overflow-type discharge facilities to the dam embankment 2, works to install tunnel-type flood discharges in the natural ground, etc. With reference to this figure, this embodiment will be described using an example of the work to add a discharge pipe to the dam embankment 2. Note that the 3D model generation device 1 described later may be used not only in the case of the work to add a discharge pipe to the dam embankment 2, but also in the "dam redevelopment" works when measuring the shape of an object underwater.

[0014] In the construction for enhancing the discharge capacity, it is often the case that the construction is carried out while the reservoir is in operation, and it is often impossible to lower the water storage level to the elevation of the facility to be added. In the construction for newly adding a discharge pipe, it is necessary to drill a hole (hereinafter sometimes referred to as the dam body drilling 4) through from a predetermined location on the downstream surface of the dam body 2 to a predetermined location on the upstream surface of the dam body 2 (hereinafter sometimes referred to as the upstream surface 3). When performing the dam body drilling 4 so that the water stored in the dam does not flow into the working area from the hole formed in the upstream surface 3 due to the dam body drilling 4, a temporary cutoff 5 is installed on the upstream surface 3 in advance by underwater construction. The temporary cutoff 5 is a temporary structure that forms a water cutoff space on the upstream surface 3 to block the intrusion of water from the upstream side of the dam body 2 during the "dam redevelopment" construction. Note that the construction of installing the temporary cutoff 5 on the upstream surface 3 is referred to as the temporary cutoff work.

[0015] In the temporary cutoff work, a door stop 6 is installed to connect the upstream surface 3 and the temporary cutoff 5 for the purpose of preventing water intrusion. At this time, while the standard value (allowable range in the construction) of the width (thickness) of the dam body 2 is from -30 [mm] to +50 [mm], the size of the unevenness (irregularity) that can be accommodated by the door stop 6 is often about ±10 [mm]. When the unevenness of the upstream surface 3 is larger than ±10 [mm], even if the temporary cutoff work is carried out, water may intrude through the gap between the door stop 6 and the upstream surface 3, and it may be difficult to carry out construction such as the dam body drilling 4. Therefore, before performing the temporary cutoff work, by performing a construction to flatten the unevenness of the location where the temporary cutoff work is to be carried out in advance, it is possible to prevent water from intruding after the temporary cutoff work.

[0016] The construction method for flattening is specifically as follows: measure the shape of the upstream surface 3, and from the measurement results, identify the position to be the construction target for flattening the unevenness of the upstream surface 3 (hereinafter, may be referred to as the target position). It has a specific process of identification and a construction process of performing the work of flattening the unevenness at the identified target position. The work of flattening the unevenness is, for example, an operation of shaving the convex part at the target position (chipping) and re-filling the concave part at the target position with putty or the like. After the construction process, in order to confirm whether the unevenness can be flattened, the shape of the upstream surface 3 is measured again. When the unevenness of the upstream surface 3 is within the range that can be accommodated by temporary fastening (for example, within ±10 [mm]), temporary fastening work is performed. When the unevenness of the upstream surface 3 is greater than ±10 [mm], the specific process and the construction process are repeated until the unevenness of the upstream surface 3 is within ±10 [mm]. Conventionally, the specific process and the construction process were carried out manually by a diver. This embodiment relates to the specific process. Instead of a diver manually measuring the shape of the upstream surface 3, a 3D model representing the shape of the upstream surface is generated based on the image captured by the camera of the underwater drone, and by referring to the generated 3D model, the shape of the upstream surface 3 is grasped.

[0017] Figure 2 is a functional configuration diagram for showing an example of the functional configuration of the 3D model generation device according to the embodiment. An example of the functional configuration of the 3D model generation device 1 will be described with reference to this figure. The 3D model generation device 1 includes an operation reception unit 11, an image acquisition unit 12, a 3D model generation unit 13, an association unit 14, a storage unit 15, and an output unit 16. These functional units are realized, for example, using a computer and software. Also, each functional unit may be realized using an electronic circuit as needed. Furthermore, each functional unit does not have to be included in a single device, and may be in a form that constitutes the 3D model generation device 1 from a plurality of devices. The 3D model generation device 1 is operated by a user. The user may be, for example, any of the person operating the moving body 7 described later, the person giving an instruction to generate a 3D model by operating the 3D model generation device 1, and the person referring to the 3D model generated by the 3D model generation device 1.

[0018] The operation reception unit 11 receives the user's operation. The operation reception unit 11 acquires operation information OI from, for example, the mouse M or the keyboard KB. The operation information OI includes, for example, information regarding the user's operation using the mouse M or the keyboard KB. The operation information OI may include, for example, information for the user to instruct the 3D model generation device 1 to generate a 3D model. The operation reception unit 11 transmits the operation information OI to the 3D model generation unit 13.

[0019] The image acquisition unit 12 acquires image information IMI from the imaging device 8. The image information IMI is information regarding a plurality of images captured by the imaging device 8. The image acquisition unit 12 transmits the acquired image information IMI to the 3D model generation unit 13.

[0020] The imaging device 8 images a predetermined angular field region in front of the imaging device 8. The imaging device 8 transmits the image information IMI including the captured image to the image acquisition unit 12. Note that the imaging device 8 may be, for example, a CCD camera using a CCD (Charge Coupled Devices) image sensor, or a CMOS camera using a CMOS (Complementary Metal Oxide Semiconductor) image sensor.

[0021] FIG. 3 is a diagram for explaining an example of the case where the imaging device images the surface on the upstream side of the dam embankment. With reference to this figure, an example of the case where, in water, the imaging device 8 provided on the moving body 7 images the upstream surface 3 will be described. The imaging device 8 is provided on the moving body 7. The moving body 7 is, for example, any one of an unmanned aerial vehicle (also referred to as “Unmanned aerial vehicle”, “UAV”, “drone”, etc.), an unmanned surface vehicle (also referred to as “Unmanned surface vehicle”, “USV”, etc.), and an unmanned underwater vehicle (also referred to as “Unmanned underwater vehicle”, “UUV”, “underwater drone”, etc.). The moving body 7 is operated using an operation means such as a dedicated controller or a mobile terminal installed with a predetermined application. The user operates the moving body 7 to change the height direction position and posture (imaging direction) of the moving body 7 with respect to the surface on the upstream side of the dam embankment 2, thereby changing the position or angle at which the shape of the upstream surface 3 is imaged, and imaging a plurality of images. Specifically, in the width direction of the upstream surface 3 (the direction orthogonal to the direction from upstream to downstream as viewed from above), while moving the moving body 7 in the height direction of the dam embankment 2 in the region including the position where the temporary cut-off is installed, imaging may be performed while changing the imaging direction. Regarding which position in the width direction of the dam embankment 2 to measure, it may be based on the position where the temporary cut-off is installed.

[0022] Note that the imaging by the imaging device 8 is performed so that the subject (a part of the upstream surface 3) represented in the captured image is again within the angle of view area. That is, imaging is performed so that a part of the subject represented in the image captured for the first time is included in the image captured for the second time. When the same subject is imaged while changing the position and angle, due to the difference in the relative positional relationship between the subject and the imaging device 8, the position, shape, orientation, etc. of the subject represented in the image change (hereinafter sometimes referred to as parallax). By calculating the parallax between the image captured at a specific timing and the image captured thereafter, the 3D model generation unit 13 described later can generate a 3D model. In addition, in the water, floating objects such as pieces of wood, leaves, and aquatic organisms are floating. Since the floating objects do not stay in one place but float or are washed away, by imaging the same subject (a part of the upstream surface 3) multiple times while changing the imaging position and angle, even if there is a part of the upstream surface 3 to be imaged that is shielded by the floating objects, the imaging device 8 can obtain an image in which that part is not shielded by imaging multiple times while changing the imaging position and angle. Therefore, by imaging the same subject multiple times while changing the position and angle, it is possible to reduce the possibility that floating objects such as pieces of wood, leaves, and aquatic organisms floating in the water, which are represented in the captured image, will be generated as part of the 3D model of the upstream surface 3.

[0023] In addition, the user operates the moving body 7 to image one or more reference parts existing on the upstream surface 3. Here, the reference part is provided on the upstream surface 3 and enables the user to grasp the actual position in the dam embankment 2. The reference part may be, for example, something that the dam embankment 2 is pre-equipped with, such as the "diamond head" provided on a hollow gravity concrete dam, flood discharge gates, and convex parts existing at both ends of the flood discharge gates. Also, the reference part may be, for example, an object with a shape that is not confused with the unevenness (irregularities) of the upstream surface 3, a printed material (marker) with a two-dimensional code or symbol printed on it, etc., which is installed after the completion of the dam. The association unit 14 described later associates the position represented in the 3D model with the actual position in the dam embankment 2 using the reference part.

[0024] The sonar 9 is provided on the moving body 7. The sonar 9 acquires information about underwater objects using sound waves. The sonar 9 acquires, for example, information regarding the direction of the upstream surface 3 where the temporary cutoff work is to be performed and the distance to the upstream surface 3. With a camera provided for operating the moving body 7, a sufficient field of view cannot be ensured due to turbidity, brightness, etc., and there may be cases where the direction in which the moving body 7 is facing and the direction and distance from the moving body 7 to the upstream surface 3 become unknown. According to the moving body 7 equipped with the sonar 9, the direction and distance from the moving body 7 in the water to the upstream surface 3 can be easily grasped regardless of the underwater environment such as turbidity and brightness.

[0025] Returning to FIG. 2, the 3D model generation unit 13 acquires operation information OI from the operation reception unit 11. Also, the 3D model generation unit 13 acquires image information IMI from the image acquisition unit 12. The 3D model generation unit 13 generates a 3D model of the shape of the upstream surface 3 by calculating the parallax of a plurality of images included in the image information IMI. For example, a technique such as stereo matching may be used for calculating the parallax of the images. The 3D model generation unit 13 transmits 3D model information regarding the generated 3D model to the association unit 14. The 3D model information may include, for example, information on the 3D model generated by the 3D model generation unit 13 and information regarding the dimensions of the shape of the upstream surface 3 based on the 3D model.

[0026] The association unit 14 acquires 3D model information from the 3D model generation unit 13. The association unit 14 generates correspondence model information CMI in which the position shown in the 3D model is associated with the actual position in the dam body 2 based on the position of the reference part shown in the 3D model information and the actual position of the reference part in the dam body 2. Regarding a specific part represented in the generated 3D model, the positional relationship (relative coordinates) with other parts represented in the 3D model can be grasped, but the actual position (absolute coordinates such as latitude, longitude, altitude, etc.) in the dam body 2 cannot be grasped. Since the positional relationship (relative coordinates) between a plurality of parts represented in the 3D model can be grasped, based on a reference point where the actual position (absolute coordinates) in the dam body 2 can be grasped, the actual position in the dam body 2 of a specific part represented in the 3D model can be grasped. Therefore, according to the correspondence model information CMI, the user can specify the target position of the upstream surface 3 from the 3D model. The association unit 14 transmits the correspondence model information CMI to the output unit 16.

[0027] Note that the association unit 14 may associate the position shown in the 3D model with the actual position on the dam embankment 2 based on a 3D model in which two or more reference parts are represented, that is, a 3D model generated by the user operating the moving body 7 and the imaging device 8 and performing imaging for each of the two or more reference parts. When there are two or more reference parts serving as the association criteria, the association accuracy is improved as compared with the case where there is one reference point. For example, when the 3D model is associated with a tilt greater than the actual one, the tilt and position of the 3D model can be corrected using other reference parts. Therefore, according to the 3D model generated by the user operating the moving body 7 and performing imaging for each of the two or more reference parts, the association unit 14 can more accurately associate the position represented in the 3D model with the actual position on the dam embankment 2.

[0028] Note that when the location to be measured is underwater and the reference part is above water, for example, a 3D model of the location to be measured may be generated by a moving body 7 that can move underwater, and a 3D model of the reference part above water may be generated by a moving body 7 that can move in the air (above water). When the reference part and the location to be measured are underwater, a 3D model can be generated by one moving body 7. Further, in order to generate a 3D model based on the reference part underwater and the location to be measured, the measurement of the shape of the upstream surface 3 using the 3D model generation device 1 may be performed when the water storage rate of the dam is high (for example, 100 [%]).

[0029] Note that the 3D model generation unit 13 and the association unit 14 transmit control information to the storage unit 15 as necessary. The control information widely includes information for the 3D model generation unit 13 or the association unit 14 to control the storage unit 15. The control information may include, for example, information for controlling the storage unit 15 to transmit a pre-trained learned model to the 3D model generation unit 13. Further, the control information may include, for example, information for controlling the storage unit 15 to transmit a pre-trained learned model to the association unit 14.

[0030] The storage unit 15 may store, for example, a learned model that has been pre-trained to generate a 3D model from the disparities of a plurality of images. The storage unit 15 may also store a learned model that has been pre-trained to be able to associate the reference part represented in the 3D model with the position of the actual reference part in the dam body 2. The storage unit 15 acquires control information from the 3D model generation unit 13 or the association unit 14. The storage unit 15 transmits information about the learned model (hereinafter sometimes referred to as learned model information) to the 3D model generation unit 13 or the association unit 14 that has transmitted the control information, according to the acquired control information. The 3D model generation unit 13 or the association unit 14 acquires the learned model information from the storage unit 15.

[0031] The output unit 16 acquires 3D model information from the 3D model generation unit 13. The output unit 16 transmits the acquired 3D model information to, for example, the display DSP. The user refers to the result of the 3D model displayed on the display DSP, for example. The user identifies the target position based on the result of the 3D model referred to, and performs construction work to flatten the unevenness of the target position as necessary.

[0032] Although not shown in the figure, the image acquisition unit 12 may transmit the image information IMI to the storage unit 15. When the image acquisition unit 12 transmits the image information IMI to the storage unit 15, the 3D model generation unit 13 acquires the image information IMI from the storage unit 15 by transmitting control information.

[0033] FIG. 4 is a flowchart for explaining an example of the flow of the construction method according to the embodiment. The user operates the mobile body 7 and the imaging device 8 to capture a plurality of images of the surface on the upstream side of the dam body 2 (upstream surface 3) (step S201). The 3D model generation device 1 acquires a plurality of images (image information IMI) captured by the imaging device 8 (step S202). The 3D model generation device 1 generates a 3D model from the parallax of the acquired plurality of images (step S203). The 3D model generation device 1 associates the position represented in the 3D model with the actual position on the dam body 2 based on the reference part (step S204). The 3D model generation device 1 outputs the result of the generated 3D model (step S205). The user specifies the target position with reference to the result of the 3D model (step S206). The user performs construction to flatten the unevenness of the specified target position (step S207).

[0034] In addition, in the present embodiment, an example in which the association unit 14 associates the position represented in the 3D model with the actual position on the dam body 2 is shown. However, the present embodiment is not limited to this example. Based on the position of the reference part, the user may associate the position represented in the 3D model by the association unit 14 with the actual position on the dam body 2. By the user making the association, the configuration of the 3D model generation device 1 becomes simpler.

[0035] In addition, in this embodiment, an example is shown in which the user operates the moving body 7 to change the position or angle at which the upstream surface 3 is imaged for imaging. However, this embodiment is not limited to this example. The imaging angle of the imaging device 8 may be changed by the user operating the imaging device 8 or by the imaging device 8 being controlled by a predetermined program. When the imaging angle of the imaging device 8 is constant (for example, in front of the moving body 7), the moving body 7 needs to change its position and the direction it is facing (rotate around) to image a specific subject from various angles when it wants to image the subject from various angles. If the imaging can be performed by changing the imaging angle of the imaging device 8, the moving body 7 does not need to change the direction it is facing (rotate around) to image a specific subject from various angles. Therefore, the operation of the moving body 7 becomes easier.

[0036] In addition, in this embodiment, an example is shown in which the mobile body 7 is provided with one imaging device 8. However, this embodiment is not limited to this example, and the mobile body 7 may be provided with a plurality of imaging devices 8. When the mobile body 7 is provided with one imaging device 8, the plurality of images captured by the imaging device 8 do not include information regarding the position and direction at which the imaging device 8 captured each image. Also, in the case of underwater, since radio waves are quickly attenuated, it is difficult to grasp the position of the mobile body 7, that is, the position of the imaging device 8, by means of GPS (Global Positioning System) or the like. When generating a 3D model based on an image captured by one imaging device 8, the 3D model generation unit 13 estimates the position and angle of the imaging device 8, the three-dimensional shape of the subject, etc. from a plurality of images in which the same subject (for example, a part of the upstream surface 3) is shown, and thereby generates a 3D model. On the other hand, when the mobile body 7 is provided with a plurality of imaging devices 8, by the user measuring the position and orientation of each of the imaging devices 8 installed on the mobile body 7, the user can obtain information regarding the positional relationship between the plurality of imaging devices 8 and the directions in which each of the imaging devices 8 captures images. When generating a 3D model based on images captured by a plurality of imaging devices 8, the 3D model generation unit 13 can generate a 3D model with higher accuracy by using the information regarding the positional relationship between the plurality of imaging devices 8 obtained by the user and the directions in which each of the imaging devices 8 captures images. Therefore, based on images captured by a plurality of imaging devices 8, the 3D model generation unit 13 can generate a 3D model with higher accuracy than when generating a 3D model by estimating the position and angle of the imaging device 8 using an image captured by one imaging device 8.

[0037] In addition, in this embodiment, an example is shown in which a 3D model of the shape of the upstream surface 3 in water is generated based on a plurality of images of the upstream surface 3 captured in water. However, this embodiment is not limited to this example. The 3D model generation device 1 may generate a 3D model of the shape of the upstream surface 3 including above water by combining a 3D model based on an image of the upstream surface 3 captured in water and a 3D model based on an image of the upstream surface 3 captured on the water surface or in the air. The 3D model generation device 1 generates a 3D model of the surface on the upstream side of the dam body in water (hereinafter, may be referred to as an underwater 3D model) based on a plurality of images of the upstream surface 3 captured in water. Further, the 3D model generation device 1 generates a 3D model of the surface on the upstream side of the dam body 2 at an elevation above the water surface based on an image of the upstream surface 3 captured on the water surface or in the air (hereinafter, may be referred to as an above-water 3D model). The 3D model generation device 1 may combine the generated underwater 3D model and above-water 3D model based on the common part between the underwater 3D model and the above-water 3D model. Here, the common part between the underwater 3D model and the above-water 3D model is, for example, a reference part that exists across water and above water. Specifically, the common part between the underwater 3D model and the above-water 3D model may be a "diamond head" provided in a hollow gravity concrete dam, a flood discharge gate, convex parts existing at both ends of the flood discharge gate, or a marker installed on the dam body 2 across water and above water. It is easier to measure the position of the reference part on the water surface compared to measuring the position of the reference part in water where the visibility is blocked due to the underwater environment and the burden on the diver is high. According to the 3D model in which the underwater 3D model and the above-water 3D model are combined, based on the position of the reference part represented in the above-water 3D model and the actual position of the reference part on the dam body 2 above water, the position shown in the underwater 3D model and the actual position on the dam body 2 in water can be associated. That is, by generating a 3D model in which the underwater 3D model and the above-water 3D model are combined, the 3D model generation unit 13 can more easily associate the position represented in the 3D model with the actual position on the dam body 2 compared to the case of using the reference part in water.

[0038] [Summary of This Embodiment] According to the above-described embodiment, the 3D model generation device 1 includes an image acquisition unit 12 that acquires image information IMI regarding a plurality of images in which the shape of the upstream surface (upstream surface 3) of the dam embankment 2 is imaged by the imaging device 8 provided in the moving body 7, and based on the parallax of the plurality of images included in the image information IMI acquired by the image acquisition unit 12, a 3D model generation unit 13 that generates a 3D model representing the shape of the upstream surface 3 by obtaining a three-dimensional shape corresponding to the shape of the upstream surface 3 represented in the image, and an output unit 16 that outputs the result of the generated 3D model. The 3D model generation device 1 generates a 3D model representing the shape of the upstream surface 3 using the images captured by the imaging device 8 provided in the moving body 7. The user can grasp the shape of the upstream surface 3 by referring to the 3D model generated by the 3D model generation device 1. In the work of measuring the shape of the upstream surface 3 among the works of flattening the unevenness of the target position, since the measurement is performed using the moving body 7, there is no need to manage the physical condition of the diver such as decompression sickness. Also, when the diver manually measures the shape of the upstream surface 3, the diver measures the shape of the upstream surface 3 at predetermined intervals (discretely). On the other hand, according to the 3D model generated by the 3D model generation device 1, the shape of the upstream surface 3 can be continuously measured. Therefore, according to the 3D model generation device 1, the measurement accuracy of the shape of the upstream surface 3 can be improved.

[0039] Also, according to the above-described embodiment, in the 3D model generation device 1, the image acquisition unit 12 further acquires an image in which a reference portion existing on the upstream surface (upstream surface 3) of the dam embankment 2 is imaged, and the 3D model generation device 1 is based on the 3D model generated using the image in which the shape of the upstream surface 3 is imaged and the image in which the reference portion is imaged. The 3D model generation device 1 further includes a correlation unit 14 that correlates the position represented in the 3D model with the actual position on the dam embankment 2. For a specific portion represented in the generated 3D model, the positional relationship (relative coordinates) with other portions represented in the 3D model can be grasped, but the actual position (absolute coordinates such as latitude, longitude, altitude, etc.) on the dam embankment 2 cannot be grasped. Therefore, the 3D model generation device 1 according to the embodiment generates a 3D model based on the image in which the reference portion whose actual position on the dam embankment 2 is known is imaged and the image in which the shape of the upstream surface 3 is imaged. Since the positional relationship (relative coordinates) between a plurality of portions represented in the 3D model can be grasped, based on a reference point where the actual position (absolute coordinates) on the dam embankment 2 can be grasped, the actual position of the portion represented in the 3D model on the dam embankment 2 can be grasped. Therefore, according to the 3D model generation device 1 according to the present embodiment, the measurement accuracy of the shape of the upstream surface 3 can be improved.

[0040] Also, according to the above-described embodiment, in the 3D model generation device 1, the image acquisition unit 12 acquires a plurality of images captured by changing the position or angle at which the shape of the upstream surface 3 is imaged. When the same subject is imaged multiple times with changes in position and angle, due to the difference in the relative position between the subject and the imaging device 8, the position, shape, orientation, etc. of the subject represented in the image change (parallax). By calculating the parallax between the image captured at a specific timing and the image captured later, the 3D model generation unit 13 according to the embodiment can generate a 3D model. Also, by imaging the same subject multiple times with changes in position and angle, the possibility that floating objects such as pieces of wood, leaves of trees, and aquatic organisms floating in the water, which are represented in the captured images, are generated as part of the upstream surface 3 in the 3D model can be reduced. Therefore, the 3D model generation device according to the present embodiment can improve the measurement accuracy of the shape of the upstream surface 3.

[0041] Also, according to the above-described embodiment, among the upstream surfaces 3, the construction method for flattening the unevenness at the target position refers to a specific process of identifying the target position for leveling the surface on the upstream side of the dam body 2 by referring to the 3D model generated by the 3D model generation device 1, a construction process of performing leveling at the identified target position, and after the leveling is performed, an image acquisition process of acquiring a plurality of images such that at least one of the imaging position and the imaging direction is different from the shape of the surface on the upstream side of the dam body 2, and a 3D model generation process of newly generating a 3D model based on the plurality of images captured in the image acquisition process. When it is determined based on the newly generated 3D model that there is a target position for unevenness, it is repeatedly implemented from the construction process. Conventionally, the measurement of the shape of the upstream surface 3 and the construction for flattening the unevenness at the target position were carried out by divers underwater. However, the work underwater imposed a heavy burden on the divers, and there was a limit to the time during which the divers could continuously work underwater. Also, since the divers performed measurements at predetermined intervals, it was difficult to continuously measure the shape of the upstream surface 3. In contrast, the measurement of the upstream surface 3 by the 3D model generation device 1, that is, the generation of the 3D model, is performed based on the images captured by the imaging device 8 provided on the moving body 7 underwater. According to the 3D model generated by the 3D model generation device 1, the shape of the upstream surface 3 can be continuously measured. Also, since the moving body 7 is not at risk of decompression sickness and does not get tired, there is no time limit for underwater work. Furthermore, in the work of measuring the shape of the upstream surface 3 among the construction for flattening the unevenness at the target position, since the measurement is performed using the moving body 7, there is no need to manage the physical condition of the divers such as decompression sickness. Therefore, according to the construction method according to the present embodiment, the shape of the upstream surface 3 can be accurately measured.

[0042] Also, according to the above-described embodiment, in the 3D model generation device 1, the moving body 7 further includes a sonar 9, and based on the detection result obtained from the sonar 9, it detects the distance and direction from the sonar 9 to the upstream surface 3 of the dam embankment 2. In the camera provided for operating the moving body 7, a sufficient field of view cannot be ensured due to turbidity, brightness, etc., and there may be cases where the direction in which the moving body 7 is facing, the direction and distance from the moving body 7 to the upstream surface 3 become unknown. According to the moving body 7 equipped with the sonar 9, regardless of the underwater environment such as turbidity and brightness, the direction and distance from the moving body 7 in the water to the upstream surface 3 can be easily grasped.

[0043] Note that the entire function or a part of each part included in the 3D model generation device 1 in the above-described embodiment may be realized by recording a program for realizing these functions on a computer-readable recording medium, reading the program recorded on this recording medium into a computer system, and executing it. Here, the "computer system" is assumed to include hardware such as an OS and peripheral devices.

[0044] Also, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, etc., and a recording unit such as a hard disk built into a computer system. Further, the "computer-readable recording medium" also includes something that dynamically holds a program for a short time, like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line, and something that holds a program for a certain period of time, like a volatile memory inside a computer system that becomes a server or a client in that case. Also, the above program may be for realizing a part of the aforementioned functions, and may further be something that can be realized in combination with a program already recorded in the computer system for realizing the aforementioned functions.

[0045] As described above, an embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to the above, and various design changes and the like can be made without departing from the gist of the present invention. Also, the configurations described in the above-described embodiments and each example may be combined.

Explanation of Reference Signs

[0046] 2…Dam body, 3…Upstream surface, 4…Hole drilled in the body, 5…Temporary cutoff, 6…Door contact, 7…Moving body, 8…Imaging device, 9…Soner, 1…3D model generation device, 11…Operation reception unit, 12…Image acquisition unit, 13…3D model generation unit, 14…Association unit, …Storage unit 15…, 16…Output unit, OI…Operation information, IMI…Image information, CMI…Corresponding model information

Claims

1. An image acquisition step of acquiring a plurality of images of the shape of the surface on the upstream side of the dam embankment imaged by an imaging device provided in a moving body; A 3D model generation step of generating a 3D model representing the shape by obtaining a three-dimensional shape corresponding to the shape represented in the image based on the parallax of the plurality of images obtained in the image acquisition step; An output step of outputting the result of the generated 3D model; A 3D model generation method having the above.

2. The image acquisition step further acquires an image of a reference portion existing on the surface on the upstream side of the dam embankment, Based on the 3D model generated using the image of the shape of the surface on the upstream side of the dam embankment and the image of the reference portion, the position represented in the 3D model and the actual position in the dam embankment are further associated with each other. The 3D model generation method according to claim 1.

3. The image acquisition step acquires a plurality of images taken by changing the position or angle of imaging the shape. The 3D model generation method according to claim 1 or claim 2.

4. A specifying step of specifying a target position for unevenness correction of the surface on the upstream side of the dam embankment based on the 3D model generated by the 3D model generation method according to claim 1 or claim 2; A construction step of performing unevenness correction on the specified target position; After the unevenness correction is performed, an image acquisition step of acquiring a plurality of images of the shape of the surface on the upstream side of the dam embankment such that at least one of the imaging position and the imaging direction is different; A 3D model generation step of newly generating the 3D model based on the plurality of images taken in the image acquisition step; including When it is determined that there is a target position of unevenness irregularity based on the newly generated 3D model, it is repeatedly carried out from the construction process Construction method.

5. The moving body further includes a sonar, and based on the detection result obtained from the sonar, the distance and direction from the moving body to the surface on the upstream side of the dam embankment are detected The 3D model generation method according to claim 1 or claim 2.

6. An image acquisition unit that acquires a plurality of images of the shape of the surface on the upstream side of the dam embankment captured by an imaging device provided in the moving body, A 3D model generation unit that generates a 3D model representing the shape by obtaining a three-dimensional shape corresponding to the shape represented in the image based on the parallax of the plurality of images acquired by the image acquisition unit, An output unit that outputs the result of the generated 3D model, A 3D model generation device comprising.

Images