JP2022134114A - Method and system for generating hd map based on aerial image captured by unmanned flight body or aircraft - Google Patents

Abstract

To provide a method and a computer system for generating an HD map of a target area by, based on aerial images obtained by capturing the same target area from different view points, obtaining a three-dimensional position for a point in the area.SOLUTION: The method includes the steps of: selecting a reference figure from a first aerial image obtained by capturing a target area from a first viewpoint for viewing the target area; discriminating a corresponding figure corresponding to the reference figure from a second aerial image obtained by capturing the target area from a second viewpoint different from the first viewpoint; matching the reference figure and the corresponding figure with each other to determine a three-dimensional position (altitude or a three-dimensional position including altitude) of the reference figure; and generating an HD map of the target area by using the determined three-dimensional position.SELECTED DRAWING: Figure 1

Description

Embodiments relate to a method for generating an HD (High Definition) map based on aerial images, and more particularly, based on aerial images of the same target area captured from different viewpoints, three-dimensional positions of points in the area are calculated. It relates to a method of generating an HD map of a region of interest by obtaining (eg, altitude or 3D position including altitude).

An HD (High Definition) map is a high-definition map used in fields such as autonomous driving, and has higher accuracy than conventional maps. For example, the HD map represents three-dimensional information about the surrounding environment of the target area, such as roads, elevation of the terrain, and rhythm, and is more than ten times more accurate than conventional maps. An HD map may have an error of less than 10 cm, for example.

HD maps are generated based on data collected by MMS (Mobile Mapping System) vehicles traveling on roads and data surveyed on the ground. As a result, the HD map is affected by errors due to the position of the GPS installed in the MMS and the sea level, and requires large-scale survey data, resulting in huge costs.

On the other hand, after generating a True Ortho image based on a Digital Surface Model (DSM) and a Digital Elevation Model (DEM) generated based on the aligned aerial image, In the case of the method of generating the HD map by using the method, there is also the problem that the error generated when generating the DSM, DEM, and true ortho images is also reflected in the HD map.

Patent Document 1 (published on August 7, 2020) describes an HD map by reconstructing a 3D space using depth prediction information of each object and class information of each object obtained by V2X information fusion technology. A learning method and a learning device for updating are disclosed.

The above information is merely an aid to understanding and may contain material that does not form part of the prior art.

Korean Patent Publication No. 10-2020-0094644

One embodiment determines the three-dimensional position (altitude or three-dimensional position including altitude) of facilities included in the target area based on aerial images of the same target area taken from different viewpoints, and determines the determined three-dimensional It is an object of the present invention to provide a method for generating an HD map of a region of interest based on location.

In one embodiment, a reference figure selected from a first aerial image obtained by photographing the target area from a first viewpoint viewing the target area and a second image obtained by photographing the target area from a second viewpoint different from the first image viewpoint. It is another object to provide a tool for determining the 3D position (elevation or 3D position with elevation) of a reference feature by matching corresponding features identified from aerial images.

According to one aspect, a computer system implemented method for generating a High Definition (HD) map based on aerial imagery, comprising: selecting a reference figure from an image; identifying a corresponding figure corresponding to the reference figure from a second aerial image obtained by photographing the target area from a second viewpoint different from the first aerial image; A method of generating an HD map, comprising the steps of: determining a three-dimensional position of said reference figure by matching corresponding figures to each other; and using said determined three-dimensional position to create an HD map of said region of interest. I will provide a.

The step of determining includes: a three-dimensional position and a shooting direction of a camera that captured the first aerial image corresponding to the first viewpoint; and a three-dimensional position of a camera that captured the second aerial image corresponding to the second viewpoint. and a photographing direction, a first angle from the first viewpoint to the reference graphic, and a second angle from the second viewpoint to the corresponding graphic, to determine the three-dimensional position including the altitude of the reference graphic. you can

The reference graphic includes a vertex of any one of the facilities included in the target area of the first aerial image, or the vertex and another vertex of the one facility, or the A reference line, which is a line connecting vertices of other facilities among facilities, may be included.

The step of selecting includes selecting a reference facility from facilities included in the target area of the first aerial image, and using the reference line as at least one guideline intersecting with the reference facility. The step of identifying includes identifying a corresponding line corresponding to the guideline from the second aerial image as the corresponding graphic, and the step of determining includes matching the guideline and the corresponding line. Determining a three-dimensional position of the guideline; and determining a three-dimensional position of an intersection point between the guideline and the reference facility based on the three-dimensional position of the guideline.

The reference facility may be the centerline of the road in the target area, and the guideline may be a line connecting vertices of two lanes centered on the centerline.

The step of selecting selects, as the reference line, a line connecting vertices of a reference facility among the facilities included in the target area of the first aerial image, and the step of identifying includes selecting a line connecting vertices of the reference facility as the reference line. The step of identifying a corresponding line corresponding to the reference line from the aerial image as the corresponding graphic and determining may determine the three-dimensional position of the baseline by matching the reference line and the corresponding line. .

The method for generating the HD map comprises the steps of: determining a three-dimensional position of the reference facility; The step of determining the position may also be included.

In the step of determining the three-dimensional position of the peripheral facility, an altitude value corresponding to the three-dimensional position of the reference facility is adjusted according to a gradient existing between the reference facility and the peripheral facility. and determining the three-dimensional location of the surrounding facility based on the adjusted altitude value.

The reference facility may be a centerline of a road within the target area, and the peripheral facilities may be lanes centered on the centerline.

The method for generating the HD map further includes determining a three-dimensional position of a facility within the target area including the reference figure based on the determined three-dimensional position. may be determined by calculating the 3D positions of the points constituting the facility by interpolation based on the determined 3D positions of the reference figure.

The determining step includes projecting the first aerial image including the selected reference figure or the selected reference figure onto the second aerial image, and projecting the projected reference figure and the second aerial image. moving the reference graphic so that the identified corresponding graphic from the image matches; and moving the height value of the corresponding reference graphic when the identified corresponding graphic matches the reference graphic. Determining as a three-dimensional position may be included.

The step of determining includes the step of outputting an altitude value corresponding to a position moved by the movement of the reference figure, and the altitude value output when the identified corresponding figure and the reference figure match. It may be determined as the 3D position of the figure.

The moving reference graphic may be a layer corresponding to the reference graphic.

At least one of a Digital Surface Model (DSM) and a Digital Elevation Model (DEM) for the target area using the determined three-dimensional position. may further include the step of generating

The method of generating the HD map may further comprise generating a True Ortho image of the region of interest based on the generated DSM and DEM.

According to another aspect, a computer system includes at least one processor configured to execute computer readable instructions contained in a memory, the at least one processor performing a first A reference figure is selected from a first aerial image obtained by photographing the target area from a viewpoint, and corresponding to the reference figure from a second aerial image obtained by photographing the target area from a second viewpoint different from the first aerial image. determining a three-dimensional position of the reference figure by identifying a corresponding figure to match, matching the reference figure and the corresponding figure with each other, and using the determined three-dimensional position to generate an HD map of the region of interest. , providing computer systems.

A process of generating a DSM or DEM by determining the 3D position (altitude or 3D position including altitude) of facility(s) included in the aerial image from aerial images of the target area taken from different viewpoints. An HD map of the region of interest can be generated without going through In addition, since an HD map is generated using aerial images, a huge amount of surveying work and data acquisition work using MMS vehicles are not required.

By utilizing aerial images of the target area captured from different viewpoints, it is possible to accurately determine the three-dimensional position including the altitude of facilities that are hidden in specific aerial images.

In one embodiment, the three-dimensional position (altitude or three-dimensional position including altitude) of facilities included in the target area is determined based on aerial images of the same target area taken from different viewpoints, and an HD map is generated. Fig. 3 shows a method; 1 is a block diagram illustrating the structure of a computer system for generating HD maps based on aerial images in one embodiment; FIG. In one embodiment, the three-dimensional position (altitude or three-dimensional position including altitude) of facilities included in the target area is determined based on aerial images of the same target area taken from different viewpoints, and an HD map is generated. 4 is a flow chart illustrating a method; In one example, by selecting a guideline from the first aerial photograph and determining the 3D position (altitude or 3D position including altitude) of the reference facility, 4 is a flow chart showing a method for determining . In one example, by selecting a guideline from the first aerial photograph and determining the 3D position (altitude or 3D position including altitude) of the reference facility, 4 is a flow chart showing a method for determining . 10 is a flow chart showing a method of determining the three-dimensional position (the altitude or the three-dimensional position including the altitude) of surrounding facilities based on the altitude of a reference facility in one example. In one example, a method of determining the three-dimensional position (altitude or three-dimensional position including altitude) of a reference figure by matching a reference figure selected from a first aerial image with a corresponding figure identified from a second aerial image. It is a flow chart showing. FIG. 4 illustrates a method of generating a DSM and DEM of a region of interest according to the determined elevation and further generating a true ortho image of the region of interest in one example; In one example, a method of determining the three-dimensional position (altitude or three-dimensional position including altitude) of a reference figure by matching a reference figure selected from a first aerial image with a corresponding figure identified from a second aerial image. It is a figure showing. In one example, by selecting a guideline from the first aerial photograph and determining the 3D position (altitude or 3D position including altitude) of the reference facility, FIG. 2 illustrates a method for determining . 1 shows a method of determining the three-dimensional position (altitude or three-dimensional position including altitude) of surrounding facilities based on the altitude of a reference facility in one example.

Embodiments will be described in detail below with reference to the accompanying drawings. Identical reference numerals presented in the drawings indicate identical parts.

"Altitude" and "determining altitude" in the drawings and detailed description set forth below may be interchanged with "three-dimensional position" and "determining three-dimensional position."

FIG. 1 shows an embodiment in which the three-dimensional positions (altitudes or three-dimensional positions including altitudes) of facilities included in a target area are determined based on aerial images of the same target area taken from different viewpoints, and HD FIG. 4 illustrates a method of generating a map;

A method of creating an HD (High Definition) map 50 of a target area 30 corresponding to at least a portion of the target site will be described with reference to FIG. The HD map 50 is a high definition map and may be a map with higher accuracy than a conventional map. For example, the HD map 50 may three-dimensionally represent surrounding environmental information of the target area 30, such as roads, terrain elevation, and tune, as shown in the figure. That is, on the HD map 50, information about facilities included in the target area 30 may be expressed in three dimensions. Facilities included within the region of interest 30 may include, for example, roads, center lines included (drawn) in roads, and lanes included (drawn) in roads. HD map 50 may include, for example, a map that an autonomous vehicle uses when autonomously driving through target area 30 .

The area of interest 30 may be the area distributed to the workers for construction of the HD map 50 .

In the embodiment, a computer system (hereinafter referred to as the computer system 100 in FIG. 2) uses the aerial images 10 and 20 of the target area 30 to capture the three-dimensional position (elevation or three-dimensional position including altitude) of facilities within the target area 30. position) is determined, and an HD map 50 of the target area 30 is generated based on the determined three-dimensional position of the facility.

The "altitude" of a facility (or a specific point) may be altitude or elevation. Alternatively, "altitude" may indicate the absolute height of the facility (or a specific point). Alternatively, 'altitude' may indicate depth information of the facility (or a specific point) with respect to a predetermined standard. The reference at this time may be the sea level or the ground.

The three-dimensional location of a particular point described in this disclosure means the altitude of the particular point, or a three-dimensional location including the altitude of the particular point (e.g., three-dimensional coordinates including altitude and horizontal coordinates). you can

The computer system 100 uses a first aerial image 10 that captures the target region 30 and a second aerial image 20 that captures the target region 30 from a different viewpoint than the first aerial image 10 to create three images associated with the target region 30 . A dimensional position or altitude (ie, a three-dimensional position or altitude of a facility (or a particular point) included in the region of interest 30) may be determined. The first aerial image 10 and the second aerial image 20 may be registered images. That is, the first aerial image 10 and the second aerial image 20 may be images whose orientations are determined. Images described in this disclosure may include pictures or images.

The first aerial image 10 and the second aerial image 20 are images captured by an unmanned flying object such as a drone or an aircraft (for target capturing) flying over a target area including the target area 30, or the captured images are It can be a registered image.

For example, the computer system 100 may specify the same point from the target area 30 of the first aerial image 10 and the target area 30 of the second aerial image 20, and identify the exact three-dimensional position of the point. As an example, computer system 100 may determine the precise three-dimensional location of points of interest by triangulation.

The computer system 100 calculates the altitude (height) of a specific point based on an aerial photograph by a plotter, or calculates the altitude of a point for floating contour lines. A three-dimensional position (elevation or three-dimensional position including elevation) may be determined for a point in the region of interest 30 .

A more specific method by which the computer system 100 determines the three-dimensional position (the altitude or the three-dimensional position including the altitude) with respect to the point of the target area 30 will be described in more detail with reference to FIGS. 2 to 11. FIG.

FIG. 2 is a block diagram illustrating the structure of a computer system for generating HD maps based on aerial images in one embodiment.

A computer system 100 that performs the method of determining the three-dimensional position of the target area 30 and generating the HD map 50 according to the embodiments described below may be implemented in the computer system 100 having the structure shown in FIG. .

Computer system 100 may be a system for executing a program that performs a method for determining the three-dimensional position of target area 30 and generating HD map 50 . Such programs may be installed in computer system 100 .

Computer system 100 may be a client terminal used by a user to process tasks for generating HD map 50 . Computer system 100 may be an electronic device capable of installing and executing a program that performs the method of determining three-dimensional position and generating HD map 50 . For example, the computer system 100 may be a personal computer (PC), a laptop computer, a laptop computer, a smart phone, a tablet, an Internet Of Things (IOT) device, or a wearable computer. ) and so on.

As shown in FIG. 2, computer system 100 may include memory 110, processor 120, communication interface 130, and input/output interface 140 as components.

The memory 110 is a computer-readable storage medium and may include random access memory (RAM), read only memory (ROM), and permanent mass storage devices such as disk drives. Here, a permanent mass storage device such as a ROM or disk drive may be included in computer device 100 as a separate permanent storage device separate from memory 110 . Also stored in memory 110 may be an operating system and at least one program code. Such software components may be loaded into memory 110 from a computer-readable medium separate from memory 110 . Such other computer-readable recording media may include computer-readable recording media such as floppy drives, disks, tapes, DVD/CD-ROM drives, memory cards, and the like. In other embodiments, software components may be loaded into memory 110 through communication interface 130 that is not a computer-readable medium. For example, software components may be loaded into memory 110 of computer system 100 based on a computer program installed by files received over network 160 .

Processor 120 may be configured to process computer program instructions by performing basic arithmetic, logic, and input/output operations. Instructions may be provided to processor 120 by memory 110 or communication interface 130 . For example, processor 120 may be configured to execute received instructions according to program code stored in a storage device, such as memory 110 .

Communication interface 130 may provide functionality for computer system 100 to communicate with other electronic devices over network 160 . As an example, processor 120 of computer system 100 may transmit requests, instructions, data, files, etc. generated according to program code recorded in a recording device such as memory 110 to other devices via network 160 under the control of communication interface 130 . device. Conversely, signals, instructions, data, files, etc. from other devices may be received by computer system 100 through communication interface 130 of computer system 100 via network 160 . Signals, instructions, data, etc., received through communication interface 130 may be transmitted to processor 120 and memory 110, and files, etc., may be stored in a recording medium (permanent recording device described above) that computing device 100 may further include. may be recorded.

The communication method by the communication interface 130 is not limited. It may also include short-range wireless communication between For example, the network 160 includes a PAN (personal area network), a LAN (local area network), a CAN (campus area network), a MAN (metropolitan area network), a WAN (wide area network), a BBN (broadband network), and the Internet. Any one or more of the networks may be included. Additionally, network 160 may include any one or more of network topologies including, but not limited to, bus networks, star networks, ring networks, mesh networks, star-bus networks, tree or hierarchical networks, and the like. will not be

Input/output interface 140 may be a means for interfacing with input/output device 150 . For example, input devices may include devices such as microphones, keyboards, cameras, or mice, and output devices may include devices such as displays, speakers, and the like. As another example, the input/output interface 140 may be a means for interfacing with a device that integrates functions for input and output, such as a touch screen. Input/output device 150 may be one device with computer system 100 .

Also, in other embodiments, computer system 100 may include fewer or more components than the components of FIG. However, most prior art components need not be explicitly shown in the figures. For example, computer system 100 may be implemented to include at least some of the input/output devices 150 described above, and may also include other components such as transceivers, cameras, various sensors, databases, and the like. It's okay.

The processor 120 of the computer system 100 determines the three-dimensional positions (altitudes or three-dimensional positions including altitudes) of facilities included in the target area 30 based on the aerial images 10, 20, as described below. It may be arranged to perform the steps for performing the method of generating the HD map 50 .

On the other hand, computer system 100 may be a server that communicates with a client terminal used by a user to process the work for generating HD map 50 . That is, the HD map 50 is generated by determining the three-dimensional positions (altitudes or three-dimensional positions including altitudes) of the facilities included in the target area 30 based on the aerial images 10 and 20 as described below. The steps for performing the method may be performed by computer system 100, which is a server in communication with client terminals.

Alternatively, the HD map 50 is generated by determining the three-dimensional positions (altitudes or three-dimensional positions including altitudes) of facilities included in the target area 30 based on the aerial images 10 and 20 as described below. The client terminal and server may be configured such that at least some of the steps for performing the method are performed at the server.

In the following detailed description, for convenience of explanation, the steps described above are described as being performed by computer system 100 (which can be either a server and/or a client terminal).

On the other hand, operations performed by processor 120 or other components of computer system 100 and operations performed by applications/programs executed by processor 120 are described as operations performed by computer system 100 for convenience of explanation. do.

The technical features described above with reference to FIG. 1 can be applied to FIG. 2 as they are, and redundant description will be omitted.

FIG. 3 shows an embodiment in which the three-dimensional positions (altitudes or three-dimensional positions including altitudes) of facilities included in a target area are determined based on aerial images of the same target area taken from different viewpoints. 4 is a flow chart illustrating a method of generating a map;

At step 310 , computer system 100 may acquire first aerial image 10 capturing area of interest 30 . The first aerial image 10 is an image captured by an unmanned flying object such as a drone or an aircraft (for target capturing) flying over a target area including the target area 30, or an image obtained by matching the captured images. you can For example, the first aerial image 10 may be an image whose pose has been established as a registered image. As an example, the first aerial image 10 may correspond to an image of the ground surface captured from a vertical direction. Alternatively, the first aerial image 10 may correspond to an image taken vertically or from any direction (ie, from a viewpoint) that makes it easier for the operator to identify or work with the target area.

The computer system 100 may acquire the first aerial image 10 by loading the saved first aerial image 10 . A drone or unmanned air vehicle flying over a target site may capture images of the target site with a predetermined degree of redundancy, and the first aerial image 10 may be determined by matching images with such a degree of redundancy. The first aerial image 10 may be an image of the target region 30 captured from a first viewpoint (as much as possible) from which the target region 30 is vertically viewed. In other words, the first aerial image 10 may be captured such that the target region 30 is most perpendicular to the target region 30 .

At step 320 , the computer system 100 may select a reference figure from the first aerial image 10 that captures the target area 30 from the first viewpoint viewing the target area 30 . For example, the user may use a tool for determining the three-dimensional position (altitude or three-dimensional position including altitude) of the reference figure from the target area 30 portion of the first aerial image 10 displayed on the screen. A reference graphic may be selected (using a provided user interface), and in accordance with such user input, computer system 100 may select the reference graphic. The reference graphic may include at least one of points, lines, and planes (closed curves or polygons) selected from within target area 30 . Within the target area 30, the reference graphic may be the target for determining altitude values.

The reference graphic may include the vertex of any one of the facilities included in the target area 30 of the first aerial image 10, or/and additionally, the vertex and the one facility. A reference line, which is a line connecting other vertices or vertices of other facilities among the facilities, may be included. Each of the facilities included within the target area 30 may include, for example, a road, a centerline included (drawn) in the road, and a lane included (drawn) in the road. Alternatively, the facility may be any feature contained within the target area 30 .

The selected reference figure may be visually distinguished from other parts of the first aerial image 10 .

At step 330 , computer system 100 may acquire a second aerial image 20 capturing region of interest 30 . As for the acquisition of the second aerial image 20 and the second aerial image 20, similar explanations regarding the first aerial image 10 can be applied as they are, so overlapping explanations will be omitted. However, the second aerial image 20 may be an image of the target region 30 captured from a second viewpoint different from that of the first aerial image 10 . In other words, the first aerial image 10 and the second aerial image 20 may be images from different viewpoints. For example, the first aerial image 10 captures the target region 30 vertically (or most vertically), while the second aerial image 20 captures the target region 30 in another non-vertical direction. It may be taken from an angle.

The computer system 100 may acquire the second aerial image 20 by loading the stored second aerial image 20 . At this time, when the first aerial image 10 is loaded or a reference figure is selected from the first aerial image 10, the computer system 100 may automatically search for and load the appropriate second aerial image 20. FIG. For example, the computer system 100 compares the coordinate values included in the first aerial image 10 and the second aerial image 20 that are matching images (or the coordinate values of the selected reference figure and the coordinates of the second aerial image 20). value comparison), a second aerial image 20 that captures the same region of interest 30 as the first aerial image 10 may be retrieved and loaded.

A plurality of second aerial images 20 may be provided. The second aerial image 20 may be displayed as a thumbnail when the first aerial image 10 is displayed as the main image on the screen, and when the corresponding thumbnail is selected, the second aerial image 20 is the main image. may be displayed on the screen. At this time, the first aerial image 10 may be displayed as a thumbnail. Alternatively, the second aerial image 20 may be arranged on the screen in the same size as the first aerial image 10 .

The first aerial image 10 may correspond to a master image, and the second aerial image 20 may correspond to a slave image.

At step 340, the computer system 100 extracts a corresponding figure corresponding to the reference figure selected in the first aerial image 10 from the second aerial image 20 obtained by photographing the target area 30 from a second viewpoint different from the first aerial image 10. can be identified. The corresponding graphic may be a portion of the second aerial image 20 that matches the reference graphic of the first aerial image 10 .

Corresponding graphics may be automatically identified from the second aerial image 20 . For example, the computer system 100 compares the first aerial image 10 and the second aerial image 20 (as an example, the form (shape or color, etc.) and a comparison of the second aerial image 20) may identify a corresponding graphic corresponding to the reference graphic.

Alternatively, the corresponding graphic may be identified according to user selection (eg, using a user interface provided by a tool) from the region of interest 30 portion of the second aerial image. For example, the user compares the first aerial image 10 and the second aerial image 20 displayed on the screen, recognizes the portion of the second aerial image 20 corresponding to the reference figure of the first aerial image 10, and selects it as the corresponding figure. and, according to such user input, computer system 100 may identify corresponding graphics.

A corresponding graphic may include at least one of a line and a plane (a closed curve or a polygon) based on the corresponding reference graphic. The corresponding graphic may be compared to determine the three-dimensional position (altitude value) of the reference graphic.

The first viewpoint includes the three-dimensional position (for example, three-dimensional coordinates (x, y, z)) and shooting direction (shooting angle of the camera (yaw, roll, pitch)) of the camera that shot the first aerial image 10. OK. Similarly, the second viewpoint is the three-dimensional position (for example, three-dimensional coordinates (x, y, z)) of the camera that captured the second aerial image 20 and the shooting direction (camera shooting angle (yaw, roll, pitch )).

At step 350, the computer system 100 matches the reference figure of the first aerial image 10 and the corresponding figure of the second aerial image 20 with each other to determine the three-dimensional position (altitude or three-dimensional position including altitude) of the reference figure. You can decide.

A method of determining the three-dimensional position (altitude or three-dimensional position including altitude) of the reference figure by matching the reference figure of the first aerial image 10 and the corresponding figure of the second aerial image 20 will be described in more detail below. do.

The computer system 100 determines that the target area 30 of the first aerial image 10 and the target area 30 of the second aerial image 20 correspond to the same area as a reference figure (consisting of at least one of points, lines, and planes). and a corresponding figure corresponding thereto, and the accurate three-dimensional position of the corresponding area (that is, the reference figure) may be specified. For example, computer system 100 may determine the precise three-dimensional location of points of interest by triangulation. In addition, the computer system 100 calculates the altitude (height) of a specific point based on an aerial photograph by a plotter, or calculates the altitude of a point for floating contour lines. , the three-dimensional position (elevation or three-dimensional position including elevation) of the region of interest 30 relative to the reference figure may be determined.

As an example, the computer system 100 stores the three-dimensional position and shooting direction of the camera that captured the first aerial image 10 corresponding to the first viewpoint, which is the value of the base (through the image matching process), and the camera corresponding to the second viewpoint. Based on the three-dimensional position and shooting direction of the camera that captured the second aerial image 20, the three-dimensional position of the point of the target area common to the first aerial image 10 and the second aerial image 20 may be determined. For determination of the three-dimensional position, focal length information and principal point information of the camera may be further used as the information of the camera.

When the three-dimensional positions (x, y, z) and the shooting directions (yaw, roll, pitch) of the cameras that captured the first aerial image 10 and the second aerial image 20 are determined through the matching process, Two line segments looking at a point commonly included in the images of may be determined, and the three-dimensional position of the point may be determined by computing the intersection of these line segments.

The photographing altitude of the first aerial image 10 and the photographing altitude of the second aerial image 20 may be the same. According to the method described above, the computer system 100 selects a reference graphic from the first aerial image 10 and identifies the corresponding corresponding graphic from the second aerial image 20, thereby matching the reference graphic and the corresponding graphic with each other. The 3D position (elevation or 3D position including elevation) of the reference figure may be determined. "Matching" the reference graphic and the corresponding graphic includes comparing the first aerial image 10 and the second aerial image 20, and/or determining (calculating) the three-dimensional position (elevation or three-dimensional position including the elevation) as described above. It may mean the one that encompasses the operations for

Alternatively, the computer system 100 can calculate the parallax between the first viewpoint of the first aerial image 10 and the second viewpoint of the second aerial image 20 (for example, the three-dimensional position of the camera indicated by the first viewpoint and the three-dimensional position of the camera indicated by the second viewpoint). dimensional position difference), a first angle to the reference figure from the first viewpoint, and a second angle to the corresponding figure from the second viewpoint, the three-dimensional position of the reference figure (altitude or three-dimensional position including altitude ) may be determined. That is, the computer system 100 may match the reference graphic with the corresponding graphic and accurately identify the three-dimensional position of the reference graphic by triangulation.

At this time, the first aerial image 10 and the second aerial image 20 are aligned images, and as described above, each image includes the three-dimensional position and shooting direction (expression) of the camera with which the image was taken. can be In other words, the first aerial image 10 and the second aerial image 20 contain three-dimensional coordinate information and rotation information (camera yaw, roll, pitch, etc.).

Specifically, the computer system 100 captures the three-dimensional position and shooting direction of the camera that captured the first aerial image 10 corresponding to the first viewpoint, and the three-dimensional image of the camera that captured the second aerial image 20 corresponding to the second viewpoint. The position and shooting direction, the first angle from the first viewpoint (that is, the three-dimensional position of the camera that captured the first aerial image 10) to the reference figure, and the second viewpoint (that is, the position of the camera that captured the second aerial image 20) The 3D position, including the elevation of the reference feature, may be determined by triangulation based on the second angle to the corresponding feature from the 3D position.

By the method described above, the computer system 100 selects a reference graphic from the first aerial image 10, identifies the corresponding corresponding graphic from the second aerial image 20, and matches the reference graphic and the corresponding graphic with each other. , may determine the 3D position (elevation or 3D position with elevation) of the reference figure by triangulation.

At step 350, computer system 100 may determine the 3D location of the facility within target area 30 including the reference graphic based on the 3D location (elevation or 3D location including elevation). The three-dimensional position of such a facility is obtained by interpolation based on the three-dimensional position of the determined reference figure. may be determined by calculating

On the other hand, if a preset reference altitude is used, the computer system 100 may use only the first aerial image 10 (ie, one aerial image) to determine the three-dimensional position of points within the region of interest. The preset reference altitude may be an altitude value (for example, altitude) of one point in the first aerial image 10 or the second aerial image 20 . Alternatively, such reference altitudes are set by a program (e.g., the tools described above) used to determine the 3D position of the reference figure (elevation or 3D position including elevation) to generate the HD map. It may be a reference altitude value. The reference altitude may be a peripheral altitude value associated with the first aerial image 10 or the second aerial image 20, and the altitude determined in step 350 may be based on such reference altitude.

For example, the computer system 100 captures the three-dimensional position and shooting direction of the camera that captured the first aerial image 10 corresponding to the first viewpoint, the first angle with respect to the reference graphic from the first viewpoint, and the preset reference altitude. Based on this, the altitude of the corresponding reference figure may be determined. Also, the computer system 100 may determine the three-dimensional position of the reference figure by determining the horizontal coordinates for the reference figure based on the determined altitude.

The 3D position of the fiducial figure determined (eg, by triangulation) as described above may include the horizontal coordinate as well as the altitude of the fiducial figure. In other words, in the embodiment, the three-dimensional position of the reference figure is determined using the aerial image in which the three-dimensional position of the camera that captured the image and the capturing direction have been determined. The horizontal coordinates can also be determined at the same time as the determination.

At step 360 , computer system 100 may generate HD map 50 of region of interest 30 using the three-dimensional positions of the reference figures determined at step 350 . That is, the computer system 100 may identify the correct altitude (height) for the facility included in the region of interest 30 at step 350 and use it as height information for the facility to map the HD map 50 to. may be generated. For example, in an embodiment, an HD map 50 may be generated that includes features such as lanes determined by concatenating the determined altitude values. Information contained in the aerial imagery described above may also be used to generate the HD map 50 .

Thus, according to embodiments, an HD map 50 of the region of interest 30 can be generated immediately, based on the (registered) aerial imagery, without performing the steps of generating a DSM or a DEM. Therefore, according to the embodiment, it is possible to solve the problem of terrain distortion that may occur when generating the HD map 50 by DSM and DEM, and the problem of height distortion of hidden facilities in aerial photographs.

According to the embodiment, since the height of facilities such as center lines or lanes in the target area 30 can be accurately specified, an HD map that accurately reflects the height information of such facilities can be obtained. 50 can be generated.

In addition, by utilizing a plurality of aerial images (that is, aerial images of the target area 30 taken from different viewpoints), the height of facilities that were hidden (by trees, vehicles, shadows, etc.) in one aerial image can be It can be accurately identified by utilizing other aerial images.

The technical features described above with reference to FIGS. 1 and 2 can be applied to FIG. 3 as they are, and redundant description will be omitted.

4 and 5 show, in one example, the three-dimensional position (elevation or 3 is a flow chart illustrating a method for determining a three-dimensional position including altitude.

A method for determining the three-dimensional position (altitude or three-dimensional position including altitude) of the facility corresponding to the reference facility within the target area 30 will be described in more detail with reference to FIGS. 4 and 5. FIG.

At step 410 , computer system 100 may select a reference facility from those included within region of interest 30 of first aerial image 10 . For example, the user may select a reference facility from a portion of the target area 30 displayed on the screen (e.g., using a user interface provided by the tool), and according to such user input, the computer System 100 may select a reference facility.

At step 420, computer system 100 may select at least one guideline that intersects the reference facility. For example, the user may select at least one guideline that intersects the reference facility from the portion of the target area 30 displayed on the screen (e.g., using a user interface provided by the tool); Depending on the user input, computer system 100 may select guidelines.

Each of the selected "guidelines" may correspond to the reference graphic described above. In other words, each of the guidelines may be the above-described reference line corresponding to a line connecting a vertex of a certain facility and another vertex of the corresponding facility or a vertex of another facility among the facilities. .

The computer system 100 may determine the three-dimensional position (the altitude or the three-dimensional position including the altitude) of the guideline, which is the reference figure.

Then, at step 340 described above, the computer system 100 may identify corresponding lines corresponding to the guidelines from the second aerial image 20 as corresponding graphics.

At step 510, the computer system 100 determines the three-dimensional position (altitude or three-dimensional position including altitude) of the guideline by matching the guideline of the first aerial image 10 and the corresponding line of the second aerial image 20. good. As a method of determining the three-dimensional position of the guideline (altitude or three-dimensional position including altitude), the above-described method of determining the three-dimensional position of the reference figure (altitude or three-dimensional position including altitude) can be applied as it is. Therefore, redundant description is omitted.

At step 520, the computer system 100 may determine the 3D position (the altitude or the 3D position including the altitude) of the point where the guideline and the reference facility intersect based on the determined 3D position of the guideline. For example, computer system 100 may determine the same altitude value as the determined guideline altitude as the altitude of the point where the guideline and the reference facility intersect. Alternatively, the computer system 100 may determine the altitude of the point where the guideline and the reference facility intersect by interpolation when the altitudes of both end points of the guideline are different.

By using the guideline altitude determination method described in steps 510 and 520, the altitude and three-dimensional position of facilities that do not include vertices (i.e., difficult to match first aerial image 10 and second aerial image 20) can also be determined accurately.

As an example, the reference facility described above is the centerline of the road in the target area 30, and the guideline is a line connecting the vertices of two lanes centered on the centerline.

In this regard, FIG. 10 shows, in one example, selecting guidelines 1020-1 to 1020-3 from the first aerial photograph 10 and showing the three-dimensional positions of the guidelines 1020-1 to 1020-3 (elevation or three-dimensional positions including elevation). Fig. 3 shows how to determine the 3D position (elevation or 3D position with elevation) of a reference facility by determining .

Center line 1010 displayed in FIG. 10 may correspond to the reference facility described above, and guidelines 1020-1 through 1020-3 may correspond to the guidelines described above. As shown, each of the guidelines 1020-1 through 1020-3 may be selected by connecting the vertices of the lanes that intersect the centerline 1010 and are centered on the centerline 1010. FIG.

Steps 510, 520 may determine the altitudes of the points where the guide lines 1020-1 through 1020-3 and the centerline 1010 intersect. The computer system 100 determines the three-dimensional positions (altitudes or three-dimensional positions including altitudes) of the remaining points of the center line 1010 by interpolation using the altitude values of the intersecting points thus determined. good.

Using the tools described above, the user may select (ie, draw) a centerline 1010 from the first aerial image 1000 displayed on the screen, and then draw guidelines 1020-1 through 1020 that intersect the centerline 1010. -3 each may be selected. Each such guideline 1020-1 to 1020-3 may be selected as a reference figure and its three-dimensional position determined.

On the other hand, in the loaded second aerial image 20, corresponding lines corresponding to the guidelines 1020-1 to 1020-3 and objects corresponding to the center line 1010 may be identified, and the computer system 100 identifies the corresponding lines and the guideline 1020. By matching -1 to 1020-3, the three-dimensional position (elevation or three-dimensional position including elevation) of each of the guidelines 1020-1 to 1020-3 may be determined.

On the other hand, below, an example of determining the three-dimensional position (the altitude or the three-dimensional position including the altitude) of the reference facility without using the guideline will be described. If the reference facility includes vertices, the three-dimensional position of the reference facility can be accurately determined without using guidelines.

That is, in step 320 described above, the computer system 100 draws a line (corresponding to the reference figure described above) connecting the vertices of the reference facility among the facilities included in the target area 30 of the first aerial image 10. may be selected as a reference line. At step 340 described above, the computer system 100 may identify corresponding lines corresponding to the baseline from the second aerial image 20 as corresponding graphics. At step 350 described above, the computer system 100 may determine the three-dimensional position (elevation or three-dimensional position including elevation) of the reference line by matching the reference line and the correspondence line. A three-dimensional position of the reference facility may be determined based on the determined three-dimensional position of the reference line.

In this way, the 3D position of the facility (elevation or 3D position including elevation) can be determined with or without guidelines depending on the characteristics of the facility (e.g., whether vertices are included). Dimensional positions can be determined.

The technical features described above with reference to FIGS. 1 to 3 can also be applied to FIGS. 4, 5, and 10 as they are, and redundant description will be omitted.

FIG. 6 is a flow chart illustrating a method of determining the three-dimensional position (the altitude or the three-dimensional position including the altitude) of surrounding facilities based on the altitude of the reference facility in one example.

A method for determining the three-dimensional position (the altitude or the three-dimensional position including the altitude) of surrounding facilities will be described below with reference to steps 352 and 354 of FIG. 3 and steps 610 and 620 of FIG.

As in step 352 shown in FIG. 3, computer system 100 (by the method described with reference to FIGS. 4, 5, and 10) determines the three-dimensional position (altitude or three-dimensional position) may be determined.

As shown in step 354, the computer system 100 calculates the three-dimensional position (the altitude or the three-dimensional position including the altitude) of at least one peripheral facility around the reference facility based on the three-dimensional position of the reference facility. You can decide.

As an example, the reference facility may be the centerline of the road within the target area 30, and the peripheral facilities may be lanes centered on the centerline. Alternatively, the surrounding facilities may include facilities at least partially hidden in the aerial image as facilities around the reference facility.

If it is assumed that there is no (inclination) gradient between the reference facility and the surrounding facilities (e.g. if the relevant option from the tools described above is selected), then the elevation of the surrounding facilities is that of the reference facility It may be determined to be the same as altitude. For example, the center line and both lateral lanes may be assumed to be at the same elevation as each other. At this time, if the altitudes are the same, the three-dimensional position of the relevant peripheral facility can be determined simply by specifying the position of the relevant peripheral facility in the first aerial image 10, and the second aerial image 20 If the figure projected on the second aerial image 20 (the figure corresponding to the peripheral facility) matches the position of the corresponding peripheral facility displayed on the second aerial image 20, additional position confirmation on the second aerial image 20 is unnecessary. If the figure projected on the second aerial image 20 and the position of the corresponding surrounding facilities displayed on the second aerial image 20 do not match, the position of the projected figure on the second aerial image 20 is projected on the second aerial image 20 By adjusting the position of the facility, the altitude of the facility corresponding to the projected figure may be determined.

On the other hand, at step 610, the computer system 100 calculates an altitude value corresponding to the altitude of the reference facility (which is the altitude value of the surrounding facility) according to the gradient existing between the reference facility and the surrounding facility. can be adjusted.

At step 620, computer system 100 may determine the three-dimensional positions (elevations or three-dimensional positions including elevations) of surrounding facilities based on the adjusted elevation values. The altitude value corresponding to the slope may be adjusted according to user input using the user interface provided by the tools described above. In addition, the computer system 100 may automatically adjust the altitude value according to the terrain information of the base for the target area 30 .

In this regard, FIG. 11 illustrates, in one example, a method for determining the three-dimensional position (elevation or three-dimensional position including elevation) of surrounding facilities based on the three-dimensional position of a reference facility.

In the illustrated example, centerline 1010 may correspond to the reference facility described above, and guidelines 1020-1 through 1020-3 may correspond to the guidelines described above. On the other hand, the lane 1110 is a facility located around the center line 1010 and may correspond to the peripheral facilities described above.

Computer system 100 may determine that the elevation of lane 1110 and the elevation of centerline 1010 are the same, assuming no grade for roads. Alternatively, computer system 100 may determine the elevation of lane 1110 by adjusting the elevation of centerline 1010 based on the grade set by the user or terrain information.

Therefore, according to the embodiment, it is possible to easily determine the three-dimensional positions of the peripheral facilities simply by determining the three-dimensional positions of the reference facilities.

The technical features described above with reference to FIGS. 1 to 5 and 10 can also be applied to FIGS. 6 and 11 as they are, and redundant description will be omitted.

FIG. 7 shows, in one example, the three-dimensional position (altitude or three-dimensional position including altitude) of the reference figure by matching the reference figure selected from the first aerial image with the corresponding figure identified from the second aerial image. It is a flow chart showing a method of determination.

In FIG. 7, by matching the reference figure of the first aerial image 10 and the corresponding figure of the second aerial image 20 using the tools described above, the three-dimensional position of the reference figure (altitude or three-dimensional position including altitude) A method for determining is described.

Steps 710-740 described below may be included in step 350 described above.

At step 710 , the computer system 100 may project the first aerial image 10 containing the reference graphic and the reference graphic onto the second aerial image 20 . For example, computer system 100 may project a layer of first aerial image 10 or a layer of reference figures onto second aerial image 20 . The reference graphic layer may be generated by selecting a reference graphic from the first aerial image 10 .

At step 720, computer system 100 may move the reference figure so that the projected reference figure and the corresponding figure identified from second aerial image 20 match. The reference graphic to be moved may be a layer corresponding to the reference graphic. For example, the user may move the layer of the reference figure projected on the second aerial image 20 (eg, using a user interface provided by a tool), and according to such user input, the computer System 100 may move the reference figure onto second aerial image 20 .

In step 740, the computer system 100 may determine the altitude value of the corresponding reference figure as the altitude of the reference figure when the corresponding figure on the second aerial image 20 and the moving reference figure match. For example, at step 730, computer system 100 may move the reference figure (at step 720) and output an altitude value corresponding to the moved position, the altitude value being output when the corresponding figure and the reference figure match. A value may be determined as the elevation of the reference figure. The altitude value that is output may be a height value relative to a reference altitude. The horizontal coordinates of the reference figure may also be determined in a similar manner. Accordingly, the 3D position of the reference figure, including altitude and horizontal coordinates, may be determined.

In this regard, FIG. 9 illustrates, in one example, the three-dimensional position (altitude or altitude) of the reference feature by matching the reference feature 905 selected from the first aerial image 900 and the corresponding feature identified from the second aerial image 910. FIG. 3 illustrates a method for determining a three-dimensional position including

As shown, in the first aerial image 900 , reference lines corresponding to lanes may be identified as reference graphics 905 . Such a reference figure 905 (a layer of reference figures 905 ) may be projected onto the second aerial image 910 . The user may use the user interface provided by the tools described above to move the layer of the projected reference figure 905 between positions 920-1 to 4 (corresponding to corresponding figure 915).

Computer system 100 may determine the three-dimensional position (elevation or three-dimensional position including elevation) of each of positions 920-1-4 (915) of projected reference figure 905. FIG. In other words, the computer system 100 assumes that each of the positions 920-1 to 920-4 (915) is a position where the projected reference figure 905 and the corresponding figure match, and calculates the three-dimensional position of the reference figure 905 at each position. (elevation or 3D position with elevation) may be determined. As the method for determining the three-dimensional position (the altitude or the three-dimensional position including the altitude) of the reference graphic 905, the above-described method can be applied as it is, so redundant description will be omitted.

Thereby, as shown in the figure, the computer system 100 may output the altitude value of the reference figure 905 at each position as the projected reference figure 905 moves through the positions 920-1 to 920-4 (915). . That is, by moving the position of the reference figure 905 along the epipolar of the corresponding figure in the second aerial image 910, the computer system 100 may change the height of the reference figure 905 and output it.

By moving the projected reference figure 905 between positions 920-1 to 920-4 (915), the user sets the altitude value (ie, height 22.7 m) at position 915 that exactly matches the corresponding figure as the reference. It is possible to know that it corresponds to the figure 905 (in other words, the user can easily identify at what height the reference figure 905 and the corresponding figure match (match)). Accordingly, computer system 100 can determine the altitude value of location 915 as the altitude value of reference feature 905 .

As described above, since the user interface of the tool can be used to determine the altitude value of the figure 905, it is possible to intuitively grasp the position 915 where the reference figure 905 and the corresponding figure match. The three-dimensional position, including elevation, of fiducial feature 905 can also be conveniently determined.

The technical features described above with reference to FIGS. 1 to 6, 10, and 11 can also be applied to FIGS. 7 and 9 as they are, and redundant description will be omitted.

FIG. 8 illustrates an example method for generating a DSM and DEM of a region of interest based on the determined altitude and generating a true ortho image of the region of interest.

As described above, according to embodiments, based on the (registered) aerial imagery, the HD map 50 of the region of interest 30 can be generated immediately without performing the steps of generating a DSM or a DEM. .

At step 810, computer system 100 generates a Digital Surface Model (DSM) and a Digital Elevation Model (DEM) of target area 30 using the determined elevation associated with target area 30. at least one of which may be generated. In other words, the computer system 100 does not determine the three-dimensional position (the altitude or the three-dimensional position including the altitude) of the target area 30 based on the DSM or DEM, but determines the three-dimensional position of the target area 30 in advance. , may generate a DSM and/or a DEM based on the determined altitude.

At step 820, computer system 100 may generate a True Ortho image of region of interest 30 based on the generated DSM and DEM. Accordingly, computer system 100 is able to construct a three-dimensional map based on a true orthoimage of target area 30 based on aerial imagery.

The information contained in the aerial images described above may further be used to generate DSM, DEM and true ortho images of the region of interest 30 .

At least one of the DSM, DEM, and true orthoimages of the region of interest 30 may be generated by the computer system 100 in accordance with user selections after the altitude of the region of interest 30 is determined.

The technical features described above with reference to FIGS. 1 to 7 and FIGS. 9 to 11 can also be applied to FIG. 8 as they are, so redundant description will be omitted.

The apparatus described above may be realized by hardware components, software components, and/or a combination of hardware and software components. For example, the devices and components described in the embodiments include processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field programmable gate arrays (FPGAs), programmable logic units (PLUs), microprocessors, Or may be implemented using one or more general purpose or special purpose computers, such as various devices capable of executing and responding to instructions. The processing unit may run an operating system (OS) and one or more software applications that run on the OS. The processor may also access, record, manipulate, process, and generate data in response to executing software. For convenience of understanding, one processing device may be described as being used, but those skilled in the art will appreciate that the processing device may include multiple processing elements and/or multiple types of processing elements. You will understand. For example, a processing unit may include multiple processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

Software may include computer programs, code, instructions, or a combination of one or more of these, to configure a processor to operate at its discretion or to independently or collectively instruct a processor. You can Software and/or data may be embodied in any kind of machine, component, physical device, computer storage medium, or device for interpretation by, or for providing instructions or data to, a processing device. good. The software may be stored and executed in a distributed fashion over computer systems linked by a network. Software and data may be recorded on one or more computer-readable recording media.

The method according to the embodiments may be embodied in the form of program instructions executable by various computer means and recorded on a computer-readable medium. Here, the medium may record the computer-executable program continuously or temporarily record it for execution or download. In addition, the medium may be various recording means or storage means in the form of a combination of single or multiple hardware, and is not limited to a medium that is directly connected to a computer system, but is distributed over a network. It may exist in Examples of media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and ROM, RAM, flash memory, etc., and may be configured to store program instructions. Other examples of media include recording media or storage media managed by application stores that distribute applications, sites that supply or distribute various software, and servers.

As described above, the embodiments have been described based on the limited embodiments and drawings, but those skilled in the art will be able to make various modifications and variations based on the above description. For example, the techniques described may be performed in a different order than in the manner described and/or components such as systems, structures, devices, circuits, etc. described may be performed in a manner different from the manner described. Appropriate results may be achieved when combined or combined, opposed or substituted by other elements or equivalents.

Accordingly, different embodiments that are equivalent to the claims should still fall within the scope of the appended claims.

100: Computer System 110: Memory 120: Processor 130: Communication Interface 140: Input/Output Interface 150: Input/Output Device 160: Network

Claims (18)

1. A computer system implemented method for generating HD maps based on aerial imagery comprising:
selecting a reference figure from a first aerial image of the target area taken from a first viewpoint viewing the target area;
identifying a corresponding figure corresponding to the reference figure from a second aerial image obtained by photographing the target area from a second viewpoint different from the first image viewpoint;
determining a three-dimensional position of the reference figure by matching the reference figure and the corresponding figure to each other; and generating an HD map of the region of interest using the determined three-dimensional location. How to generate maps. The determining step includes:
The three-dimensional position and shooting direction of the camera that captured the first aerial image corresponding to the first viewpoint, the three-dimensional position and shooting direction of the camera that captured the second aerial image corresponding to the second viewpoint, determining a three-dimensional position, including altitude, of the reference figure by triangulation based on a first angle to the reference figure from the first viewpoint and a second angle to the corresponding figure from the second viewpoint;
A method for generating an HD map according to claim 1. the reference graphic includes a vertex of any one of the facilities included in the target area of the first aerial image;
A reference line that is a line that connects the vertex and another vertex of the one facility or a vertex of another facility among the facilities,
3. A method for generating an HD map according to claim 1 or 2. The selecting step includes:
selecting a reference facility from facilities included within the region of interest of the first aerial image; and selecting the reference line as at least one guideline that intersects with the reference facility;
The step of identifying identifies a corresponding line corresponding to the guideline from the second aerial image as the corresponding graphic,
The determining step includes:
determining a three-dimensional position of the guideline by matching the guideline and the corresponding line; and determining a three-dimensional position of a point where the guideline and the reference facility intersect based on the three-dimensional position of the guideline. 4. A method of generating an HD map as claimed in claim 3, comprising the step of determining. The reference facility is the centerline of the road in the target area, and the guideline is a line connecting the vertices of each of two lanes centered on the centerline.
5. A method of generating an HD map according to claim 4. The step of selecting selects, as the reference line, a line connecting vertices of reference facilities among facilities included in the target area of the first aerial image,
The step of identifying identifies a corresponding line corresponding to the reference line from the second aerial image as the corresponding graphic,
the determining step determines the three-dimensional position of the reference line by matching the reference line and the corresponding line;
A method of generating an HD map according to claim 3. determining a three-dimensional position of the reference facility; and determining a three-dimensional position of at least one peripheral facility around the reference facility relative to the three-dimensional position of the reference facility; A method for generating an HD map according to any one of claims 4-6. Determining the three-dimensional position of the surrounding facility includes:
adjusting an altitude value corresponding to the three-dimensional position of the reference facility according to a gradient existing between the reference facility and the peripheral facility; and based on the adjusted altitude value of the peripheral facility. 8. A method of generating an HD map as claimed in claim 7, comprising: determining a three-dimensional position. The reference facility is the center line of the road in the target area, and the peripheral facilities are lanes centered on the center line.
A method of generating an HD map according to claim 7. determining a three-dimensional position of a facility within the target area including the reference figure based on the determined three-dimensional position;
The three-dimensional position of the facility is determined by calculating the three-dimensional position of the point constituting the facility by an interpolation method based on the determined three-dimensional position of the reference figure.
A method for generating an HD map according to any one of claims 1-9. The determining step includes:
Projecting the first aerial image containing the selected reference graphic or the selected reference graphic onto the second aerial image;
moving the reference figure so that the projected reference figure and the identified corresponding figure from the second aerial image match; and corresponding to when the identified corresponding figure and the reference figure match. 2. The method of claim 1, comprising determining an altitude value of said reference figure as a three-dimensional position of said reference figure. The determining step includes:
outputting an altitude value corresponding to the moved position by moving the reference figure;
determining an altitude value output when the identified corresponding graphic and the reference graphic match as the three-dimensional position of the reference graphic;
A method for generating an HD map according to claim 11. The moving reference figure is a layer corresponding to the reference figure,
A method for generating an HD map according to claim 11. 2. The HD map of claim 1, further comprising generating at least one of a Digital Surface Model (DSM) and a Digital Elevation Model (DEM) of the area of interest using the determined three-dimensional locations. How to generate . 15. The method of generating an HD map of claim 14, further comprising: generating a true orthoimage of the region of interest based on the generated DSM and DEM. A computer program for causing the computer system to perform the method according to any one of claims 1-15. A computer-readable recording medium on which a program for causing the computer system to execute the method according to any one of claims 1 to 15 is recorded. a computer system,
at least one processor configured to execute computer readable instructions contained in memory;
The at least one processor
A reference figure is selected from a first aerial image obtained by photographing the target area from a first viewpoint viewing the target area, and the reference figure is selected from a second aerial image obtained by photographing the target area from a second viewpoint different from the first image viewpoint. identifying a corresponding figure corresponding to the figure; determining a three-dimensional position of the reference figure by matching the reference figure and the corresponding figure with each other; and using the determined three-dimensional position to create an HD map of the region of interest. A computer system that generates

Images