JP2023033991A - Autonomous flight survey system and method by unmanned flying body - Google Patents

Abstract

To provide an autonomous flight survey system and a method by an unmanned flying body that are able to reduce a time required for preliminary preparation and improve autonomous flight survey.SOLUTION: An autonomous flight survey system includes: an unmanned flying body (10) that autonomously flies over a to-be-photographed area; and an information processor (30) that processes image data acquired by the unmanned flying body. The unmanned flying body has: a camera (101) capable of photographing the to-be-photographed area (A) from a vertical direction or a photographing direction inclined by a predetermined angle; and control units (200, 207) that execute control such that a plurality of images of the to-be-photographed area are captured by inclining the photographing direction of the camera by a predetermined angle from a direction different from a vertical downward direction. Based on the plurality of images photographed by the camera, the information processor calculates topographic data of the to-be-photographed area.SELECTED DRAWING: Figure 4

Description

The present invention relates to a surveying technique by autonomous flight of an unmanned air vehicle.

Unmanned Aerial Vehicles (UAVs) called drones are equipped with photogrammetry cameras and scanners, so that construction sites, disaster sites, and the like, which are inaccessible to humans, can be surveyed or monitored from the air. In particular, by equipping the unmanned aircraft with GPS (Global Positioning System) and GNSS (Global Navigation Satellite System) functions, it is possible to accurately measure the position of the unmanned aircraft. autonomous flight is becoming possible (see Patent Documents 1 and 2).

JP 2021-004035 A JP 2020-126663 A

However, in order to fly an unmanned aerial vehicle autonomously for surveying, it is necessary to set control points in the survey target area in advance and measure the position of each control point, which requires a lot of preparation time. In particular, in order to realize fully automatic surveying by autonomous flight, advance preparation becomes a bottleneck in efficiency.

SUMMARY OF THE INVENTION The present invention has been made in view of such circumstances, and its object is to provide an autonomous flight survey system and method using an unmanned flying object that can shorten the time required for advance preparation and improve the efficiency of autonomous flight survey. be.

To achieve the above object, an autonomous flight survey system according to one aspect of the present invention includes an unmanned flying object that autonomously flies over an imaging target area, information for processing image data acquired by the unmanned flying object, a processing device, wherein the unmanned flying object has a camera capable of photographing the photographing target area from a vertical direction or from a photographing direction tilted by a desired angle; and a photographing direction of the camera vertically downward. a control unit configured to capture a plurality of images of the target area while being tilted by a predetermined angle from different directions, wherein the information processing device captures the plurality of images captured by the camera; and calculating terrain data of the imaging target area based on the above.
According to one aspect of the present invention, it is desirable that the predetermined angle is 10° or more and 30° or less.
According to one aspect of the present invention, a plurality of control points are arranged in advance at predetermined intervals in the imaging target area, each control point is equipped with a positioning function, and the information processing device obtains the position information of the control points. and the plurality of images captured by the camera, topography data of the imaging target area can be calculated.
According to one aspect of the present invention, the information processing device reduces the image size of each of the plurality of captured images, extracts a plurality of feature points from each of the reduced image sizes, The terrain data can be calculated from the plurality of feature points.
According to one aspect of the present invention, it is desirable to limit the number of feature points extracted from one image to a predetermined number or less.
An autonomous flight survey method according to one aspect of the present invention is an autonomous flight survey method using an unmanned flying object that autonomously flies over an imaging target region, wherein the unmanned flying object moves the imaging target region vertically or at a desired angle. a photographing step of setting the photographing direction of the camera and photographing a plurality of images including the survey target area by the camera; and a terrain calculation step of calculating terrain data of the survey target area based on the image of , wherein in the photographing step, the photographing direction of the camera is tilted by a predetermined angle from a different direction with respect to the vertical downward direction. and a plurality of images of the imaging target area.
According to one aspect of the present invention, it is desirable that the predetermined angle is 10° or more and 30° or less.
According to one aspect of the present invention, a plurality of control points are arranged in advance at predetermined intervals in the imaging target area, each control point is equipped with a positioning function, and in the terrain calculation step, the position information of the control points is and the plurality of images captured by the camera, topography data of the imaging target area can be calculated.
According to one aspect of the present invention, in the terrain calculation step, the image size of each of the plurality of images captured in the photographing step is reduced, and a plurality of feature points are calculated from each of the reduced image sizes. The terrain data can be calculated from the plurality of feature points.
According to one aspect of the present invention, it is desirable to limit the number of feature points extracted from one image to a predetermined number or less.

According to the present invention, since depth is created by oblique photography at a predetermined angle with respect to the vertical downward direction, it is possible to perform surveying with sufficient accuracy without setting control points, shorten the time required for advance preparation, and autonomously perform surveying. Flight survey can be made more efficient. In particular, by setting the predetermined angle to 10° or more and 30° or less, the survey accuracy can be improved.
Furthermore, by arranging multiple ground control points equipped with positioning functions, it is possible to simplify and shorten the series of tasks from preparation to data analysis.
Also, by reducing the data size of the captured image and limiting the number of feature points, the time required for data analysis can be greatly shortened.

1 is a schematic diagram showing a schematic configuration of an autonomous flight system for an unmanned air vehicle according to an embodiment of the present invention; FIG. 1 is a diagram showing a schematic configuration of an example of an unmanned air vehicle used in this embodiment; FIG. FIG. 4 is a diagram showing a schematic configuration of another example of an unmanned air vehicle used in this embodiment; It is a figure which shows the structure of the autonomous flight survey system used with the autonomous flight survey system by this embodiment. 1 is a schematic diagram showing an example of an autonomous flight survey system according to a first embodiment of the present invention; FIG. 1 is a schematic diagram showing an example of a flight route of the autonomous flight survey system according to the first embodiment; FIG. FIG. 5 is a schematic diagram showing an example of a flight route of the autonomous flight survey system according to the second embodiment of the present invention; 9 is a flow chart showing a processing procedure in the autonomous flight survey system according to the second embodiment; . FIG. 11 is a schematic diagram showing an example of an imaging method in the autonomous flight survey system according to the second embodiment; FIG. 11 is a side view schematically showing the imaging direction of the unmanned flying object in the autonomous flight survey system according to the second embodiment;

Preferred embodiments of the autonomous flight survey system according to the present invention will be described in detail below with reference to the accompanying drawings.

1. Autonomous Flight Survey System As illustrated in FIG. 1, an autonomous flight survey system according to an embodiment of the present invention includes an unmanned flying object 10 equipped with a camera for photographing and capable of autonomous flight and a departure and arrival point for the unmanned flying object 10. base 11; It is assumed that the unmanned air vehicle 10 has a flight course and a shooting target area A set in advance, as will be described later. After taking off from the base 11 , the unmanned flying object 10 flies over an imaging target area A (for example, a cutting and filling construction site 15 ) along a predetermined course, takes an image of the imaging target area A, holds the imaging data, and returns to the base 11 . return.

The base 11 is not particularly limited, but is a hangar having an opening and closing type hatch that can take off and land. For example, openable domes 12a and 12b are installed on base 13 so as to be movable in the direction of arrow C, and base 13 is provided with take-off and landing ports 14. As shown in FIG. Preferably, the take-off/landing port 14 is provided with facilities for charging the battery of the unmanned air vehicle 10 upon landing. The unmanned air vehicle 10 takes off and lands with the retractable domes 12a and 12b open, and after landing, the retractable domes 12a and 12b are closed and stored.

2. Unmanned flying object As shown in FIG. 2, an unmanned flying object 10 is provided with a plurality of flight motors M at predetermined positions on the periphery of a housing 100, and a propeller Pr is connected to the rotating shaft of each motor M. . As will be described later, the unmanned air vehicle 10 is capable of at least one of remote-controlled flight and autonomous flight.

A visible light camera 101 is connected to a support member 102 so as to be able to pass through, and three or more legs 103 are fixed to the bottom surface of a housing 100 of the unmanned air vehicle 10 . The visible light camera 101 can be a normal RGB camera for color photography, and is capable of photographing moving images or still images.

The visible light camera 101 can photograph the ground surface while the unmanned air vehicle 10 is flying. Also, the visible light camera 101 has a swinging mechanism, and can arbitrarily change the photographing direction. For example, the visible light camera 101 can set the photographing direction directly below the unmanned air vehicle 10 (vertical direction) or in an oblique direction inclined at a desired angle. Various mechanisms can be employed as the swing mechanism, and for example, the imaging direction of the visible light camera 101 can be changed by rotating a servomotor.

As illustrated in FIG. 3, an infrared (IR) camera may be provided as an aerial camera in addition to the visible light camera 101 as shown in FIG. An unmanned flying object 10a illustrated in FIG. 3 has a camera unit 111 that integrates a visible light camera 101 and an infrared (IR) camera 101a, and switches between the visible light camera 101 and the infrared camera 101a as necessary. , or both. A combination of the visible light camera 101 and the illumination light source 104 can also be used for nighttime photogrammetry. The illumination light source 104 may be installed on the bottom surface of the housing 100 through the support member 105 . The illumination light source 104 is a light source for illuminating the surface of the earth, which will be described later, and can be a so-called spotlight or searchlight. Other than these, the configuration is the same as that of the unmanned flying vehicle 10 of FIG. 2, so the same reference numerals are given to the same members and the description thereof will be omitted.

The camera unit 111 and the illumination light source 104 are equipped with a swing mechanism like the visible light camera 101, and can arbitrarily change the photographing/illumination direction. The IR camera 101a of the camera unit 111 can capture the temperature distribution of the ground surface in the imaging area Ai as an image, and the topography can be grasped even at night. For example, images captured by both the visible light camera 101 and the infrared camera 101a can be used to survey the disaster site. Since the infrared camera 101a visualizes the temperature distribution, it becomes possible to detect the existence of survivors, for example.

As exemplified in FIG. 4, the unmanned flying object 10 includes the above-described visible light camera 101, and furthermore, a processor 200 for realizing various functions of the unmanned flying object 10, a flight control unit, and a flight control unit are provided in the housing 100 of the unmanned flying object 10. A unit 201, a motor control unit 202, a wireless communication unit 203, an RTK (Real Time Kinematic)-GNSS (Global Navigation Satellite System) unit 204, a memory 205, an interface 206, and an imaging direction control unit 207 are provided. there is

The processor 200 can implement various functions described below by executing a program stored in a memory (not shown). The processor 200 controls the imaging timing, imaging direction, and other imaging parameters (shutter speed, ISO sensitivity, aperture, focus, etc.) of the visible light camera 101 . For example, when the current position of the unmanned flying object 10 calculated by the RTK-GNSS unit 204, which will be described later, reaches a predetermined photographing position, the processor 200 causes the visible light camera 101 to move to the predetermined position through the photographing direction control unit 207. The visible light camera 101 is controlled to face the shooting direction, and the visible light camera 101 is controlled to shoot with predetermined shooting parameters. The photographing direction control unit 207 can set the photographing direction of the visible light camera 101 to an arbitrary direction, for example, by controlling the rotation of the servo motor of the swing mechanism described above. Note that the processor 200, the flight control unit 201, the motor control unit 202, and the imaging direction control unit 207 function as a control unit of the unmanned air vehicle 10, and are realized by the processor 200 executing a program stored in a memory (not shown). can also

A wireless communication unit 203 can communicate with a base station and/or a wireless control terminal, receive various commands, and notify the current state of the unmanned air vehicle 10 . A flight control unit 201 controls the flight state (flight direction, flight speed, etc.) of the unmanned air vehicle 10 . Specifically, the flight control unit 201 adjusts the rotation speeds of the flight motors M1 to M4 through the motor control unit 202 according to a remote control instruction or a predetermined flight route, thereby controlling the flight direction, flight speed, and the like. . In this embodiment, the flight control unit 201 controls the flight direction, flight speed, etc., according to the preset flight route and imaging target area A. FIG.

The RTK-GNSS unit 204 is a positioning unit that calculates the current position (latitude, longitude, height) of the unmanned air vehicle 10 based on signals transmitted from satellites. Using the RTK-GNSS unit 204 enables autonomous flight.

The memory 205 can store data such as images captured by the visible light camera 101, capturing positions, capturing parameters at the time of capturing, and position information of the autonomous flight route and the capturing target area A. The memory 205 may be a storage medium such as a memory card that can be attached to and detached from the unmanned air vehicle 10 . Further, the memory 205 may be built in the housing 100 of the unmanned air vehicle 10 and connected to an external system through the interface 206 so that image data and the like can be transmitted and received.

3. Information Processing System The information processing system of the autonomous flight survey system using the unmanned flying object 10 equipped with the visible light camera 101 will be described below as an example.

As illustrated in FIG. 4, an autonomous flight survey system 20 according to this embodiment includes an unmanned flying object 10 and an information processing device (computer) 30, and the unmanned flying object 10 has the configuration described above.

The computer 30 has an interface 210 and a processor 211 for transmitting and receiving image data to and from the unmanned air vehicle 10, further includes a ROM for storing control programs and the like, a RAM as an operation area for the control programs, various data , an interface with a peripheral circuit, and the like. The computer 30 is installed, for example, at a location away from the location where photogrammetry is performed, and executes processing such as terrain data analysis based on images captured by the unmanned air vehicle 10 . For example, the processor 211 can implement a terrain calculation function by executing SfM (Structure from Motion) analysis software, and can generate a three-dimensional model of the imaging target area A from the images captured by the unmanned air vehicle 10 .

The terrain calculation function realized by the processor 211 includes image preprocessing, feature point extraction, terrain model generation, and terrain data calculation functions.

The image preprocessing function performs preprocessing such as noise removal, color correction, and image size change on a plurality of images captured by the unmanned air vehicle 10 in preparation for subsequent processing. The feature point extraction function extracts a plurality of feature points from a plurality of preprocessed images. A feature point is a point in an image that has a feature different from surrounding areas (for example, a difference in gradation from surrounding pixels is equal to or greater than a predetermined value).

The terrain model generation function associates feature points between a plurality of images to generate a three-dimensional model of the imaging target area A. FIG. More specifically, matching between images is performed by associating feature points extracted from each image to generate a sparse point cloud. By calculating the 3D coordinates of each pixel using a rougher point cloud, generating a denser point cloud, creating mesh data from this point cloud, and pasting the original image Generate the final 3D model.

The terrain data calculation function calculates a terrain index value (for example, the elevation of an arbitrary point) from the three-dimensional model of the imaging target area A as numerical data (topography data). It should be noted that the processing of the SfM analysis software described above is shown as an example and simplified, and it goes without saying that various conventionally known analysis methods can be applied to the present invention. Note that the terrain data calculation function will be described using a specific example.

4. First Embodiment The autonomous flight survey system according to the first embodiment of the present invention uses GPS-equipped control points for reference surveying, thereby omitting a series of complicated operations such as installation of control points, surveying, and input of coordinates. It is possible to greatly reduce the man-hours for surveying and data processing at the site. In particular, when it is necessary to repeat the installation and removal of ground control points in accordance with the progress of construction work, there is a remarkable reduction in man-hours.

As shown in FIG. 5, GPS-equipped control points L1 to L7 are arranged at intervals of 100 m or less at the cutting and banking construction site 15, and the coordinate data of each control point can be automatically acquired by the GPS function. After daily earthwork work is completed, the unmanned air vehicle 10 is flown, and the visible light camera 101 photographs a predetermined photographing target area A for each predetermined photographing area Ai.

As illustrated in FIG. 6, the unmanned flying object 10 flies along a predetermined flight route R so that the entire photographing target area A set in advance is photographed. An image is captured by (imaging area Ai). Therefore, only by changing the setting of the imaging target area A, a desired area of the construction site can be photographed. As described above, the processor 211 can execute the SfM analysis software to generate a three-dimensional model of the imaging target area A from the images captured by the unmanned air vehicle 10 . Since the coordinate data of each control point can be automatically acquired by the GPS function, it is possible to save labor and shorten the time required for setting up the control points after surveying.

The captured images are stored in the memory 205 without being compressed, but in this embodiment, the image size of each image is reduced (for example, about 1/4). Reducing the image size reduces the number of pixels in the image, making it easier to extract feature points and reducing the load of data processing. In particular, it is difficult to obtain significant feature points when the surface of the ground to be photographed has little undulation and the geology is uniform. Therefore, in this embodiment, the image size is reduced before extracting the feature points to make it easier to extract the feature points. Such reduction can be performed, for example, on SfM analysis software.

Also, the number of feature points extracted from each image is not limited, and generally, for example, about 10,000 feature points are extracted from one image. For this reason, feature points are forcibly extracted even when there is no change in geographic shape that becomes a feature, which may cause errors. In particular, if the area to be photographed has little undulations and uniform geology, forcibly extracting a large number of feature points may reduce the surveying accuracy. Therefore, in this embodiment, the number of feature points extracted from one image is limited (for example, within 2000). Such restrictions can be implemented, for example, by changing settings on the SfM analysis software. In this way, significant feature points that are commonly extracted from a plurality of images can be extracted.

As described above, by reducing the data size of the captured image and limiting the number of feature points, it is possible to greatly reduce the time required for data analysis.

As an example, assume that the survey area of the imaging target area A is 2 ha, use the control points L1 to L7 equipped with the GNSS function for the basic survey, and the RTK-GNSS function and a 20-megapixel high-resolution visible light camera for the unmanned air vehicle 10. 101 is installed, and SfM analysis software, point cloud editing software, and three-dimensional CAD are used for data analysis, according to this embodiment. The time required for preparation such as score setting can be shortened to 1/4 compared to the conventional method, and by suppressing the data size, the time required for a series of tasks from basic surveying to data analysis can be reduced to 1/3. could be shortened to

5. Second Embodiment An autonomous flight survey system according to a second embodiment of the present invention can improve the accuracy of estimation of the focal length of the visible light camera 101 by oblique photography described in detail below. Even without it, survey results with sufficient accuracy can be obtained. That is, according to the second embodiment, surveying by the unmanned flying object 10 can be performed without setting control points, and the time required for surveying can be further reduced. Moreover, even when combined with the basic survey using the GPS-equipped control points according to the first embodiment described above, it is possible to achieve simplification and time reduction of a series of operations from advance preparation to data analysis.

As exemplified in FIG. 7, the unmanned flying object 10 flies along a predetermined flight route R so as to photograph the entirety of a predetermined photographing target area A at a cutting and banking construction site 15, which will be described below. Each photographing area Ai of the photographing target area A is photographed from different photographing directions (oblique direction photographing). On the flight route R illustrated in FIG. 7, when the unmanned flying object 10 flies in the forward direction D1(1) along one side of the imaging target area A, the visible light camera 101 is tilted at a predetermined angle toward the traveling direction to photograph. Each photographing area Ai of the target area A in the D1(1) direction is photographed. When the unmanned flying object 10 reaches the end of the imaging target area A, it reverses its flight direction, and changes the imaging direction of the visible light camera 101 to the direction opposite to that during forward movement. In this way, the unmanned flying object 10 flies in the backward direction D2(1), and with the visible light camera 101 tilted at a predetermined angle toward the traveling direction, photographs each photographing area Ai of the photographing target area A in the D2(1) direction. .

In this way, by repeating the reciprocating photographing shown in FIG. 7, a plurality of images of the photographing target area A photographed from different directions for each photographing area Ai can be obtained. By using a plurality of captured images of the entire imaging target area A obtained in this way, the survey work and data analysis work of the desired imaging target area A can be efficiently performed without the work of setting control points, as described below. can.

In FIG. 8, a plurality of images of the imaging target area A are captured while controlling the imaging direction of the visible light camera 101 mounted on the unmanned air vehicle 10, and the captured image data is stored in the memory 205 (step 301: imaging process). ). Subsequently, the image captured in the image capturing process (the image stored in the memory 205) is read into the computer 30 through the interfaces 206 and 210 (step 302). The image read into the computer 30 is processed by the image preprocessing function (step 303: image preprocessing step) and the feature point extraction function (step 304: feature point extraction step). are combined to generate a three-dimensional model of the survey target area (step 305). Subsequently, the terrain data is calculated by the terrain data calculation function (step 306). The details of steps 301 to 306 will be described below.

<Photography process>
As shown in FIG. 9, the imaging direction of the visible light camera 101 is inclined by a predetermined angle with respect to the vertical downward direction (the direction of gravity), and the imaging area Ai is photographed by the visible light camera 101 (oblique direction imaging). When the unmanned air vehicle 10 is positioned at position L1, the imaging area Ai of the visible light camera 101 is an area including points N1 to N4, and the imaging direction is F1. The points N1 to N4 are shown for convenience in order to define the photographing area Ai, and are different from control points. Also, the traveling direction of the unmanned flying object 10 at position L1 is assumed to be the direction of arrow D1.

The unmanned flying object 10 flies while photographing the photographing area Ai in the traveling direction D1, changes the traveling direction at a predetermined point (end of the photographing target area A), and again photographs the previously photographed area Ai from a different angle. . For example, the unmanned flying object 10 captures an image of the area Ai again from the capturing direction F2 at the position L2. It is assumed that the traveling direction of the unmanned flying object 12 at position L2 is the direction of arrow D2. Thus, in this embodiment, one area Ai is photographed from different directions. In FIG. 9, the imaging range at the position L1 and the imaging range at the position L2 are assumed to be the same area Ai. (shooting position) may be set.

Such oblique photography is performed because depth is created in the imaging direction by using an image obtained by oblique photography, and the estimation accuracy of the focal length of the visible light camera 101 can be improved. In conventional vertical photography, all images are taken in the same direction, so it is not possible to uniquely determine the focal length, which is a parameter that indicates the scale in the depth direction. need to be installed. By performing oblique photography as in the present embodiment, it is possible to improve the accuracy of estimating the focal length of the visible light camera 101, and it is possible to obtain sufficiently accurate survey results without setting control points. .

FIG. 10 is a side view schematically showing the imaging direction of the unmanned flying object 10. As shown in FIG. The unmanned flying object 10 performs imaging by tilting the imaging direction F of the visible light camera 101 by a predetermined angle α with respect to the vertically downward direction P, and an image of the imaging area Ai is captured.

Here, the inclination α of the photographing direction F of the visible light camera 101 with respect to the vertically downward direction P is preferably 10° or more and 30° or less. If the inclination α is less than 10°, the photographing direction becomes close to the vertically downward direction, and the effect of oblique photographing cannot be sufficiently obtained. Moreover, if the inclination α exceeds 30°, the dimension per pixel will change greatly between the front side and the back side of the photographed image, resulting in increased distortion and possibly lower analysis accuracy.

<Image preprocessing step>
In general, images shot in the shooting process are uncompressed and stored in the memory 205, but in the present embodiment, the image size of each image is reduced (for example, about 1/4) in the image preprocessing process. Reducing the image size reduces the number of pixels in the image, making it easier to extract feature points. In particular, it is difficult to obtain significant feature points when the surface of the ground to be surveyed has little undulation and the geology is uniform. Therefore, in this embodiment, the image size is reduced before extracting the feature points to make it easier to extract the feature points. Such reduction can be performed, for example, on SfM analysis software.

<Feature point extraction process>
In general, the number of feature points extracted from each image in the feature point extraction process is not limited, and for example, about 10,000 feature points are extracted from one image. On the other hand, in such a method, feature points are forcibly extracted even when there is no change in the geographical shape that becomes a feature, which may cause an error. In particular, in a survey area with poor undulations and uniform geology, forcibly extracting a large number of feature points may reduce survey accuracy. Therefore, in this embodiment, the number of feature points extracted from one image is limited (for example, within 2000). Such restrictions can be implemented, for example, by changing settings on the SfM analysis software. In this way, significant feature points that are commonly extracted from a plurality of images can be extracted.

As described above, by reducing the data size of the captured image and limiting the number of feature points, it is possible to greatly reduce the time required for data analysis.

As described above, according to the autonomous flight survey system 20 according to the second embodiment of the present invention, the unmanned flying object 10 equipped with the visible light camera 101 is used, and the visible light camera 101 performs photogrammetry. In particular, oblique photographing at a predetermined angle with respect to the vertical downward direction produces depth, so that the accuracy of surveying in the height direction can be improved. Therefore, it is possible to carry out surveying with sufficient accuracy without setting control points, and to efficiently and accurately carry out photogrammetry using the unmanned aerial vehicle 10 even in cases where it is not possible to enter, such as at a disaster site. can be done. For example, when photogrammetry is performed on the unmanned air vehicle 10 at an altitude corresponding to a GSD of 20 mm, high accuracy within a survey accuracy of ±50 mm can be achieved.

As an example, the survey area of the imaging target area A is assumed to be 2 ha, and the unmanned flying object 10 is equipped with an RTK-GNSS function and a high-resolution visible light camera 101 of 20 million pixels, and SfM analysis software and point cloud editing are used for data analysis. According to this embodiment, when using software and three-dimensional CAD. It eliminates the need for advance preparation such as setting control points, and shortens the work time required for surveying to 1/4 of the conventional time. Moreover, even when combined with the basic survey using the GPS-equipped control points according to the first embodiment, it is possible to achieve simplification and time reduction of a series of operations from advance preparation to data analysis.

10, 10a Unmanned aerial vehicle 11 Retractable base 12a, 12b Retractable dome 13 Base 14 Departure and arrival port 15 Cut and fill construction site 20 Autonomous flight survey system 30 Computer 100 Housing 101 Visible light camera 200, 211 Processor 201 Flight control unit 202 Motor Control unit 203 Wireless communication unit 204 RTK-GNSS unit 205 Memory 206, 210 Interface

Claims (10)

An autonomous flight survey system comprising an unmanned flying object that autonomously flies over an imaging target area and an information processing device that processes image data acquired by the unmanned flying object,
The unmanned flying object has a camera capable of photographing the photographing target area from a vertical direction or from a photographing direction tilted by a desired angle, and the photographing direction of the camera tilted by a predetermined angle from a different direction with respect to the vertical downward direction. a control unit that controls to capture a plurality of images of the target area;
The information processing device calculates terrain data of the shooting target area based on a plurality of images shot by the camera;
An autonomous flight survey system characterized by: 2. The autonomous flight survey system according to claim 1, wherein the predetermined angle is between 10[deg.] and 30[deg.]. A plurality of control points are arranged in advance at predetermined intervals in the imaging target area, and each control point is equipped with a positioning function,
3. The autonomy according to claim 1, wherein said information processing device calculates terrain data of said photographing target area based on position information of said control point and a plurality of images photographed by said camera. flight survey system. The information processing device reduces the image size of each of the plurality of captured images, extracts a plurality of feature points from each of the reduced image sizes, and extracts the terrain data from the plurality of feature points. The autonomous flight survey system using the unmanned flying object according to any one of claims 1 to 3, wherein the system calculates 5. The autonomous flight survey system according to claim 4, wherein the number of feature points extracted from one image is limited to a predetermined number or less. An autonomous flight survey method by an unmanned flying object autonomously flying over an imaging target area,
The unmanned flying object has a camera capable of photographing the object area to be photographed from a vertical direction or from a photographing direction tilted by a desired angle, and a plurality of images including the survey object area are obtained by the camera by setting the photographing direction of the camera. a photographing process of photographing,
a terrain calculation step of calculating terrain data of the survey target area based on the plurality of images captured in the photographing step;
wherein, in the photographing step, a plurality of images of the photographing target area are photographed by tilting the photographing direction of the camera from different directions with respect to the vertically downward direction by a predetermined angle. 7. The autonomous flight survey method according to claim 6, wherein the predetermined angle is 10[deg.] or more and 30[deg.] or less. A plurality of control points are arranged in advance at predetermined intervals in the imaging target area, and each control point is equipped with a positioning function,
8. The autonomy according to claim 6, wherein in said terrain calculation step, terrain data of said photographing target area is calculated based on position information of said control points and a plurality of images photographed by said camera. Flight survey method. In the terrain calculation step, the image size of each of the plurality of images captured in the photographing step is reduced, a plurality of feature points are extracted from each of the reduced image sizes, and from the plurality of feature points The autonomous flight survey method according to any one of claims 6 to 8, wherein the terrain data is calculated. 10. The autonomous flight survey method according to claim 9, wherein the number of feature points extracted from one image is limited to a predetermined number or less.

Images