KR20230076220A - Method and system for setting ground control point automatically for generating orthogonal map using drone images - Google Patents

Abstract

Disclosed is a method and system for automatically setting a ground control point (GCP) when generating an orthogonal map using drone images. The method for automatically setting a ground control point (GCP) when generating an orthogonal map using drone images according to the present invention comprises the steps of: collecting a plurality of images captured by a drone from the drone; applying the plurality of images to a deep learning engine to select a target image in which a GCP is captured from among the plurality of images; extracting coordinate values for a center point of the GCP from the target image; creating a mapping list by mapping latitude values and longitude values of the GCP identified in exchangeable image file format (EXIF) information of the target image to the coordinate values; and producing an orthogonal map with corrected GPS errors for the coordinate values using the mapping list. Accordingly, reliability of the orthogonal map can be improved.

Description

Method and system for automating the setting of ground reference points when generating orthographic maps using drone images

The present invention relates to a method for automating the setting of a Ground Control Point (GCP) when generating an orthographic map using a drone-captured image. It is about how to improve.

In general, when generating an orthographic map due to errors in the GPS mounted on the drone, it is not accurately mapped on the existing map and misalignment occurs.

To correct this, a ground control point (GCP) is installed on the ground and a survey is performed on the location to obtain accurate GPS coordinates.

For example, if a drone shoots including a ground control point (GCP), the x, y coordinates of the central point in the ground control point (GCP) appearing on the captured image, the latitude value of the ground control point (GCP) installation location, and GeoReferencing errors are minimized by correcting GPS errors by mapping longitude values.

In this process, it is inconvenient to repeat the process of finding a picture in which a ground control point (GCP) has been taken among numerous drone-captured pictures and extracting the location coordinate values of the ground control point (GCP) from the corresponding picture several times.

Accordingly, there is a need for a technology capable of automatically extracting the location coordinate values of the ground control point (GCP) by automatically selecting a picture in which the ground control point (GCP) is taken from among numerous drone photos.

An embodiment of the present invention selects an image in which a ground control point (GCP) is captured from among images taken by a drone, extracts coordinate values of the ground control point (GCP), and adds the coordinate values of the extracted ground control point (GCP) to the ground control point ( The purpose of this study is to increase the efficiency of orthographic map production by automating the process of correcting GPS errors by mapping latitude and longitude values of GCP).

According to the present invention, a method for automating the setting of ground control points (GCPs) when generating an orthographic map using a drone-captured image includes the steps of collecting a plurality of images taken by a drone from the drone, and sending the plurality of images to a deep learning engine. and selecting a target image in which a ground control point (GCP) is photographed from among the plurality of images, extracting coordinate values of the center point of the ground control point (GCP) from the target image, and generating a mapping list by mapping the latitude and longitude values of the ground control point (GCP) identified in the EXIF (exchangeable image file format) information of the target image; and using the mapping list, A step of producing an orthographic map correcting a GPS error may be included.

In addition, the system for automating the setting of a ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention includes a collection unit that collects a plurality of images taken by a drone from the drone, and a deep image of the plurality of images. Applied to a learning engine, a selection unit that selects a target image in which a ground control point (GCP) is photographed from among the plurality of images, and an extraction unit that extracts coordinate values of the center point of the ground control point (GCP) from the target image; , a generator for generating a mapping list by mapping the coordinate values with the latitude and longitude values of the ground control point (GCP) identified in the EXIF information of the target image, and using the mapping list, It may include a production unit for producing an orthographic map correcting GPS errors.

According to the present invention, an image including a ground control point (GCP) is selected from among numerous ground images taken by a drone, the location coordinate values of the ground control point (GCP) required for GPS error correction are extracted, and the ground control point (GCP) By automating the repetitive work of mapping to the latitude and longitude values of GeoReferencing, labor costs and time for producing orthographic maps with high accuracy are reduced, and the possibility of manual error is minimized, thereby increasing the reliability of orthographic maps. can improve

1 is a block diagram showing the configuration of an automated system for setting a ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention.
2 is a diagram illustrating a map in which ground control points (GCPs) are installed at a plurality of points on the ground in the automatic system for setting ground control points (GCPs) when generating an orthographic map using a drone-captured image according to the present invention.
3 is a diagram illustrating images captured by a drone including ground control points (GCPs) in the system for automatically setting ground control points (GCPs) when generating an orthographic map using images taken by a drone according to the present invention.
4 is a diagram showing an example of taking multiple shots at different locations for one ground control point (GCP) in the automatic system for setting the ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention.
FIG. 5 is a diagram illustrating multiple captured images at different locations shown in FIG. 4 in the automatic system for setting a ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention.
6 is a flowchart illustrating a sequence of an automated method for setting ground control points (GCPs) when generating an orthographic map using a drone-captured image according to the present invention.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes can be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents or substitutes to the embodiments are included within the scope of rights.

Terms used in the examples are used only for descriptive purposes and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description will be omitted.

1 is a block diagram showing the configuration of an automated system for setting a ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention.

Referring to FIG. 1 , the automatic system 100 for setting a ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention includes a collection unit 110, a sorting unit 120, and an extraction unit 130 , It can be configured to include a generation unit 140, a production unit 150 and a learning unit 160.

The collection unit 110 collects a plurality of images taken by the drone from the drone.

The selection unit 120 applies the plurality of images to the deep learning engine, and selects a target image in which a ground control point (GCP) is photographed from among the plurality of images.

Prior to the implementation of the invention, the collection unit 110 generates a control command for changing the shooting altitude and shooting angle of the drone that captures the ground control point (GCP), and the shooting altitude and shooting angle are changed according to the control command. Images of the ground control point (GCP) may be collected from the drone.

The learning unit 160 labels the area of the ground control point (GCP) in the collected image, deep-learns the collected image and the labeled value, and performs a deep learning operation on the subject where the ground control point (GCP) of the same shape is photographed. Build the deep learning engine that selects images.

Accordingly, the selection unit 120 can easily find a target image including a ground control point (GCP) of the same shape among numerous images captured by the drone by using the deep learning engine.

The extraction unit 130 extracts the coordinate values of the center point of the ground control point (GCP) from the target image.

The generating unit 140 identifies the latitude and longitude values of the ground control point (GCP) from the EXIF (exchangeable image file format) information of the target image, maps the latitude and longitude values to the coordinate values, and maps the latitude and longitude values to the coordinate values. create a list

At this time, when a plurality of target images are selected by the deep learning engine, the generation unit 140 selects the ground control point (GCP) among the latitude and longitude values of the ground control point (GCP) identified from each of the plurality of target images. ), the latitude and longitude values closest to the coordinates of the center point can be selected.

The generating unit 140 may generate the mapping list by mapping the selected latitude and longitude values to the coordinate values.

The production unit 150 uses the mapping list to create an orthographic map in which GPS errors for the coordinate values are corrected.

As such, according to the present invention, an image including a ground control point (GCP) is selected from among numerous terrestrial images taken by a drone, the location coordinate values of the ground control point (GCP) required for GPS error correction are extracted, and the ground control point By automating the repetitive work of mapping to the latitude and longitude values of (GCP), labor costs and time for producing orthographic maps with high GeoReferencing accuracy are reduced, and the possibility of manual errors is minimized, thereby improving the reliability of orthographic maps. can

Hereinafter, referring to FIGS. 2 to 5 , a ground control point (GCP) setting automation system when generating an orthographic map using a drone-captured image according to the present invention will be described in detail.

(Step 1 - Preparing for filming)

In this step, prior to drone flight, a ground control point (GCP) is installed in the area to be photographed, and a process of registering GPS latitude and longitude values is performed.

The ground control point (GCP) can be manufactured in a size that can be identified through a camera of a drone flying at a certain altitude, and an off-the-shelf product can be used.

2 is a diagram illustrating a map in which ground control points (GCPs) are installed at a plurality of points on the ground in the automatic system for setting ground control points (GCPs) when generating an orthographic map using a drone-captured image according to the present invention.

FIG. 2 shows a map 200 showing ground control points (GCPs) 210 installed at four identifiable points on the ground.

According to the present invention, when generating an orthographic map using a drone-captured image, the ground control point (GCP) setting automation system performs a survey on the center point of the ground control point (GCP) or identifies GPS latitude and longitude values on a satellite map. A ground control point (GCP) is installed at a location and the corresponding latitude and longitude values are registered in the automatic ground control point (GCP) setting system.

The ground control points (GCPs) may be evenly distributed around or in the center of the four vertexes of the area to be captured, and an appropriate number of at least 6 or more GCPs may be arranged according to the size or shape of the area to be captured.

When trying to produce a time-series orthographic map by repeatedly photographing the same shooting target area, the coordinate registration process for the ground control point (GCP) is initially performed only once.

(Step 2 - Deep learning training on ground control point (GCP) images)

In this step, a process of generating an image for learning is performed by controlling a drone to photograph a ground control point (GCP) in the air.

Several shots are taken while varying the altitude and shooting angle, and deep learning is performed after labeling the ground control point (GCP) area in the image.

The deep learning algorithm can be used with various image processing algorithms that have already been verified and widely used, and deep learning is performed only once for the first time on a ground control point (GCP) of the same shape.

(Step 3 - Shooting)

In this step, shooting is performed by controlling the drone in the shooting target area. The photographing altitude is set to an altitude range in which a ground control point (GCP) can be identified in the photographed image.

3 is a diagram illustrating images captured by a drone including ground control points (GCPs) in the system for automatically setting ground control points (GCPs) when generating an orthographic map using images taken by a drone according to the present invention.

Referring to FIG. 3 , when generating an orthographic map using a drone-captured image according to the present invention, a ground control point (GCP) setting automation system controls a drone to capture an image 300 including a ground control point (GCP) 310. can

In addition, the automatic system for setting the ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention captures the entire area of the shooting target with a uniform overlap, and takes several photos including each ground control point (GCP). can be filmed.

4 is a diagram showing an example of taking multiple shots at different locations for one ground control point (GCP) in the automatic system for setting the ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention.

FIG. 4 shows an example of taking multiple pictures of one of the four ground control points (GCPs) installed on the map 400 at different positions (viewpoints) 410 and 420 .

FIG. 5 is a diagram illustrating multiple captured images at different locations shown in FIG. 4 in the automatic system for setting a ground control point (GCP) when generating an orthographic map using a drone-captured image according to the present invention.

In (a) of FIG. 5, an image taken at the position 410 of FIG. 4 is shown, and in (b) of FIG. 5, an image taken at the position 420 in FIG. 4 is shown. 5 (a) and (b) are examples of displaying two pictures taken in consideration of the degree of overlap with respect to one installed ground control point (GCP).

The center point of the ground control point (GCP) appearing on each image shown in (a) and (b) of FIG. 5 has different X and Y coordinate values as the shooting position (view point) 410 and 420 changes, but the ground The GPS latitude and longitude values for the center point of the reference point (GCP) are the same in (a) and (b) of FIG. 5 .

(Step 4 - Image Upload and Automatic Orthographic Map Generation)

In this step, 1) collecting a plurality of images taken by the drone, 2) applying the collected plurality of images to the deep learning engine, selecting a target image in which a ground control point (GCP) is identified in the image, and selecting A process of extracting the X and Y coordinates of the center point of the ground control point (GCP) from one target image is performed.

Then, 3) EXIF (exchangeable image file format) information of the image including the ground control point (GCP) is extracted, and a process of identifying latitude and longitude values of the image is performed.

4) Latitude and longitude values are extracted from the list of ground control points (GCP) registered in advance (step 1 - shooting preparation step), and compared with the latitude and longitude values of the image extracted in step 3, the closest distance A process of mapping the ground control point (GCP) latitude and longitude values of X and Y coordinates on the image is performed.

If the above process is repeated as many times as the number of images taken by the drone, 6) A mapping list of X, Y coordinates for each image and GCP latitude and longitude values is completed, and based on this information, the ground control point A process of generating (GCP) information files and transmitting them to a production system for producing orthographic maps is performed.

For reference, well-known commercial and open-source orthographic mapping systems (eg, Pix4D, OpenDroneMap) are designed to receive such ground control point (GCP) information files as input.

6 is a flowchart illustrating a sequence of an automated method for setting ground control points (GCPs) when generating an orthographic map using a drone-captured image according to the present invention.

A method for automating setting of ground control points (GCPs) when generating an orthographic map using a drone-captured image according to the present embodiment may be performed by the above-described system 100 for automating setting of ground control points (GCPs).

Referring to FIG. 6 , in step 610, the ground control point (GCP) setting automation system 100 collects a plurality of images taken by the drone from the drone.

In step 620, the ground control point (GCP) setting automation system 100 applies the plurality of images to the deep learning engine, and selects a target image in which the ground control point (GCP) is captured from among the plurality of images.

The deep learning engine may be built by performing deep learning on a plurality of images for learning obtained by taking ground control points (GCPs) through drones.

Specifically, the ground control point (GCP) setting automation system 100 generates a control command to change the shooting altitude and shooting angle of the drone that shoots the ground control point (GCP), and the shooting altitude and shooting according to the control command. From the drone whose angle is changed, a plurality of learning images of the ground control point (GCP) are collected, and after labeling the area of the ground control point (GCP) in the collected training images, deep learning is performed, and the ground control point (GCP) of the same shape ( GCP) can build the deep learning engine that selects the captured target image.

In step 630, the ground control point (GCP) setting automation system 100 extracts the coordinate values of the center point of the ground control point (GCP) from the target image.

In step 640, the ground control point (GCP) setting automation system 100 identifies the latitude and longitude values of the ground control point (GCP) from the EXIF information of the target image, and the coordinates extracted in step 630. Map to a value to create a list of mappings.

According to an embodiment, when a plurality of target images are selected by the deep learning engine, the ground control point (GCP) setting automation system 100 determines the latitude value of the ground control point (GCP) identified from each of the plurality of target images. The mapping list may be generated by selecting a latitude value and a longitude value closest to the coordinate value among the longitude values and mapping the selected latitude and longitude values to the coordinate values.

In step 650, the ground control point (GCP) setting automation system 100 uses the mapping list to produce an orthographic map in which GPS errors for the coordinate values are corrected.

Therefore, according to the present invention, the coordinate values of the ground control points (GCPs) are extracted by selecting the images in which the ground control points (GCPs) are photographed from among the images taken by the drone, and the coordinate values of the extracted ground control points (GCPs) are added to the ground control points. By automating the process of correcting GPS errors by mapping latitude and longitude values of (GCP), the efficiency of orthographic map production can be increased.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. The device can be commanded. Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

As described above, although the embodiments have been described with limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

100: Automatic system for setting ground control points (GCP)
110: collection unit
120: selection unit
130: extraction unit
140: generating unit
150: production department
160: learning unit

Claims (3)

Collecting a plurality of images taken by the drone from the drone;
applying the plurality of images to a deep learning engine and selecting a target image in which a ground control point (GCP) is photographed from among the plurality of images;
extracting coordinate values of the center point of the ground control point (GCP) from the target image;
generating a mapping list by mapping latitude values and longitude values of the ground control point (GCP) identified in EXIF (exchangeable image file format) information of the target image to the coordinate values; and
Creating an orthographic map correcting GPS errors for the coordinate values using the mapping list
A method for automating the setting of a ground control point (GCP) when generating an orthographic map using a drone-captured image. According to claim 1,
generating a control command for changing a photographing altitude and a photographing angle of a drone that photographs the ground control point (GCP);
collecting an image of the ground control point (GCP) from the drone whose photographing altitude and photographing angle are changed according to the control command;
labeling an area of the ground control point (GCP) in the collected image; and
Constructing the deep learning engine for selecting a target image in which the ground control point (GCP) of the same shape is captured by deep learning the collected image and the labeled value
A method for automating the setting of a ground control point (GCP) when generating an orthographic map using a drone photographed image further comprising a. According to claim 1,
When a plurality of target images are selected by the deep learning engine,
Selecting a latitude value and a longitude value closest to the coordinate value among the latitude values and longitude values of the ground control point (GCP) identified from each of the plurality of target images
Including more,
The step of generating the mapping list,
generating the mapping list by mapping the selected latitude and longitude values to the coordinate values;
A method for automating the setting of a ground control point (GCP) when generating an orthographic map using a drone-captured image.

Images