JP2022158951A - Method and system for updating aerial map utilizing drone image - Google Patents

Abstract

To provide a method and a system for updating an aerial map utilizing a drone image.SOLUTION: An update method according to an embodiment includes the steps of: generating a roof true-orthophoto including images of preset height or higher by using a Digital Elevation Model (DEM), a Digital Surface Model (DSM), and a true- orthophoto of an existing aerial map; matching a drone image and the roof true-orthophoto; and inputting a Ground Control Point (GCP) to another drone image based on the matched drone image.SELECTED DRAWING: Figure 2

Description

The following description relates to an aerial map update method and system utilizing drone footage.

An aerial map means a map generated using an aerial photograph. The acquisition cycle of an aeronautical map is long (for example, two years), and there is a problem that it is difficult to quickly reflect changes in areas such as roads.

In order to solve these problems, there is a method of using images generated by drones. (Ground Control Points) and have to be used to create and/or update maps.

Furthermore, there is also the problem that the matching work between the surveyed GCP and the drone image can only be done manually.

Korean Patent No. 10-0940118

When generating and/or updating an aeronautical map using drone images, it is possible to reduce the time and cost for generating and/or updating an aeronautical map by using an existing 3D aeronautical map as a GCP. , to provide an update method and system.

Another object of the present invention is to provide an update method and system that can save manpower and time by supporting automatic matching between a 3D aerial map used as GCP and drone images.

A method for updating a computing device comprising at least one processor, said at least one processor updating an existing aerial map-based DEM (Digital Elevation Model), DSM (Digital Surface Model) ), and using the true-orthophoto to generate a roof true-orthophoto containing an image above a preset height; matching roof true orthophotos; and inputting ground control points (GCPs) to another drone image based on the matched drone image by the at least one processor, Provide an update method.

According to one aspect, the step of generating the roof true orthophoto includes extracting a portion where the difference between the DSM and the DEM is equal to or greater than a preset height from the true orthophoto to generate the roof true orthophoto. It can be characterized as

According to another aspect, matching the drone image and the roof true orthophoto includes processing feature point matching of the drone image and the roof true orthophoto, matching at least one from a portion of the drone image. selecting a point; and selecting, as a virtual GCP for the selected matching point, the three-dimensional coordinates of the actual terrain for the location of the roof true orthophoto corresponding to the selected matching point. It can be a feature.

According to another aspect, the step of processing feature point matching may include matching feature points of the drone image and feature points of the roof true orthophoto using R2D2 matching.

According to another aspect, in matching the drone image and the roof true orthophoto, a seed is set based on a preset length of a range to be updated with the roof true orthophoto. loading a roof true orthophoto closest to the set seed; after extracting feature points from the loaded roof true orthophoto, the position of the set seed is added to the extracted feature points; The method may further include resetting based on the density, and loading a part of the drone image closest to the reset seed position.

According to yet another aspect, the step of setting the seed includes, for a range composed of a plurality of squares each having a side length of the preset length, a predetermined seed for each of the plurality of squares. It may be characterized by setting a number seed randomly.

According to another aspect, the step of resetting may be characterized by resetting the position of the seed to a position where the density of the extracted feature points is highest.

According to another aspect, the step of inputting GCPs to the other drone image includes inputting 3D coordinates of the matched points of the roof true orthophoto to the points of the matched drone image as virtual GCPs. , and inputting the virtual GCPs to the other drone image points that match the matched drone image points.

According to another aspect, the step of inputting the virtual GCP matches points of the matched drone image with points of the other drone image using dense matching. good.

According to yet another aspect, the updating method may further include filtering the GCPs by the at least one processor.

According to another aspect, the step of filtering the GCPs includes: based on the pixel location points of the matched drone image, an internal focal distance of the drone camera that has a minimum difference from the image coordinates corresponding to the GCPs. , distortion coefficients, rotation information of the drone camera, and a corrected GCP.

According to yet another aspect, the updating method includes, by the at least one processor, generating an updated DSM based on an input drone image of the filtered GCPs; and by the at least one processor, The method may further include generating an updated true orthophoto based on the updated DSM.

A computer program is provided for causing a computer device to perform the method.

A computer-readable recording medium is provided in which a program for causing a computer device to execute the method is recorded.

at least one processor implemented to execute computer readable instructions, said at least one processor facilitating an existing aerial map-based Digital Elevation Model (DEM), Digital Surface Model (DSM), and True Ortho generating a roof true orthophoto including an image above a preset height using a true-ortho photo, matching the drone image with the roof true orthophoto, and matching the drone Provided is a computer device characterized by inputting a GCP (Ground Control Point) to another drone image based on an image.

When generating and/or updating an aeronautical map using drone images, it is possible to reduce the time and cost for generating and/or updating an aeronautical map by using an existing 3D aeronautical map as a GCP. .

Manpower and time can be saved by supporting automatic matching of 3D aerial maps used as GCP and drone images.

1 is a block diagram illustrating an example of a computing device, in accordance with one embodiment of the present invention; FIG. 4 is a flow chart showing an example of an update method in one embodiment of the present invention; 1 is a flowchart illustrating an example method for matching drone footage and roof true orthophotos in accordance with one embodiment of the present invention. FIG. 4 is a diagram showing an example of set seeds in one embodiment of the present invention; FIG. 10 is a diagram illustrating an example of repositioning seeds in one embodiment of the present invention; FIG. 4 is a diagram showing an example of matching drone images and roof true orthophotos in an embodiment of the present invention; FIG. 4 is a diagram showing an example of selecting 3D coordinates corresponding to coordinates of a drone image in an embodiment of the present invention; FIG. 10 is a diagram showing an example of setting a connection between a subdrone image and a virtual GCP in one embodiment of the present invention; FIG. 5 is a diagram showing an example of matching the same point by dense matching in one embodiment of the present invention; FIG. 5 is a diagram showing an example of matching the same point by dense matching in one embodiment of the present invention; FIG. 4 is a diagram showing an example of correcting weight values in one embodiment of the present invention; FIG. 4 is a diagram showing an example of correcting weight values in one embodiment of the present invention; FIG. 4 is a diagram showing the difference in resolution between drone images and aerial images in one embodiment of the present invention; FIG. 4 is a diagram showing an example of updating an aerial map using drone images in one embodiment of the present invention;

Embodiments will be described in detail below with reference to the accompanying drawings.

The update system according to embodiments of the present invention may be implemented by at least one computer device, and the update method according to embodiments of the present invention may be performed by at least one computer device implementing the update system. A computer program according to an embodiment of the present invention may be installed and executed in a computer device, and the computer device executes an update method according to an embodiment of the present invention under control of the executed computer program. good. The computer program described above may be recorded in a computer-readable recording medium in order to combine with a computer device and cause the computer device to execute the update method.

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

FIG. 1 is a block diagram illustrating an example of a computing device in one embodiment of the invention. The computer device 100 includes a memory 110, a processor 120, a communication interface 130, and an input/output (I/O) interface (I/O interface) 140. 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 computing device 100 based on computer programs 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 device 100 to communicate with other devices over network 160 . As an example, requests, commands, data, files, etc. generated by processor 120 of computer device 100 in accordance with program code recorded in a recording device such as memory 110 may be sent to others via network 160 under the control of communication interface 130 . device. Conversely, signals, instructions, data, files, etc. from other devices may be received by computing device 100 through communication interface 130 of computing device 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.

Input/output interface 140 may be a means for interfacing with input/output (I/O) devices 150 . For example, input devices may include devices such as a microphone, keyboard, or mouse, 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 computing device 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, computing device 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, databases, and the like.

FIG. 2 is a flowchart illustrating an example of an update method according to one embodiment of the invention. The update method according to this embodiment may be executed by the computer device 100 described with reference to FIG. At this time, the processor 120 of the computing device 100 may be implemented to execute control instructions according to the operating system code and at least one computer program code contained in the memory 110 . Here, processor 120 may control computing device 100 such that computing device 100 performs steps 210-260 included in the method of FIG. 2 according to control instructions provided by code recorded in computing device 100. .

At step 210, the computing device 100 utilizes existing aerial map-based DEMs (Digital Elevation Models), DSMs (Digital Surface Models), and true-orthophotos. to generate a roof true orthophoto containing images above a preset height. A DEM is a numerical model that expresses a bare-earth portion excluding buildings, trees, man-made structures, etc. from real-world terrain information. It is a model that expresses trees, buildings, man-made structures, etc. Therefore, the portion where the difference between DSM and DEM (DSM-DEM=CHM (Canopy Height Model)) is equal to or greater than a preset value corresponds to buildings, trees, man-made structures, etc. . At this time, the computer device 100 extracts the roof true orthophoto by extracting only the portion where the difference between the DSM and the DEM in the true orthophoto is equal to or greater than a preset height (for example, 17 m) as true. may be generated.

At step 220, the computing device 100 may match the drone image with the roof true orthophoto. Here, the drone image may mean an image captured by a drone. The matched points can be used as GCPs (Ground Control Points), and the 3D coordinates of the matched points can be obtained using DSM true orthophotos. The method of matching drone images and roof true orthophotos will be described in more detail with reference to FIG.

In step 230, the computing device 100 may input GCPs to another drone image based on the matched drone image. Here, the GCP may be the 3D coordinates obtained from the roof true orthophoto matched to the drone image, instead of the actually surveyed GCO. In other words, the computer device 100 may set the connection between the peripheral sub-drone images and the virtual GCP while placing the matched drone image as the main image. The method of entering the GCP is described in more detail with reference to FIGS. 8-10.

At step 240, computing device 100 may filter the GCPs. The information inside the camera of the aerial image is accurate, but the information inside the camera is inaccurate because the drone image uses a low-cost camera, so it is necessary to correct the camera pose and the information inside the camera. For such correction, computing device 100 may utilize virtual GCPs to perform bundle adjustments. Bundle adjustment is described in more detail below.

At step 250, the computing device 100 may generate an updated DSM based on the filtered GCP input drone image. At this time, an issue caused by changing the source of the image from the aerial image to the drone image may be corrected.

At step 260, computing device 100 may generate an updated true orthophoto based on the updated DSM. Since this process is a well-known matter in the process for creating an aeronautical map, a detailed description thereof will be omitted.

FIG. 3 is a flowchart illustrating an example method for matching drone images and roof true orthophotos in accordance with one embodiment of the present invention. Steps 310 - 370 of FIG. 3 may be included in step 220 of FIG. 2 and may be performed by computing device 100 .

In step 310, the computing device 100 may set a seed based on a preset length of a region (Region Of Interest: ROI) to be updated with the roof true orthophoto. A seed is a point that will later become a GCP location, and the preset length may be 120 m as an example. Here, the preset length may be an interval between seeds, or may be a length indicating an area in which a preset number of seeds are set.

FIG. 4 is a diagram showing an example of set seeds in one embodiment of the present invention. FIG. 4 shows an example in which a square 420 whose side length is set to a preset length (eg, 120 m) is displayed in a specific range of the roof true orthophoto 410 . At this time, the computing device 100 may set m (m=5 in the embodiment of FIG. 4) seeds in such a square 420 . A plurality of squares may be randomly set in one roof true orthophoto 410, and m seeds may be randomly set for each square.

The embodiment of FIG. 4 is only one method for uniformly selecting GCPs to ensure accuracy, and the seed setting method is not limited to the embodiment of FIG. As an example, within the roof true orthophoto 410 to be updated, the seeds may be set uniformly at intervals of a preset length.

Referring again to FIG. 3, at step 320, computing device 100 may load the closest roof true orthophoto from each seed. At this time, the roof true orthophoto closest to each seed may be a part of the image including the seed among the roof true orthophotos.

At step 330, after extracting the feature points from the loaded roof true orthophoto, the computing device 100 may reset the positions of the set seeds based on the density of the extracted feature points. Since the seeds are set at regular intervals, the seed positions may not be on the roof. Accordingly, the computer device 100 may reset the seed position by moving the position of the seed to a position where feature points of the roof true orthophoto closest to the seed are concentrated.

FIG. 5 is a diagram showing an example of resetting seed positions in one embodiment of the present invention. FIG. 5 shows an example of repositioning the seed 520 to the feature point dense position on the nearest roof 530 when the seed 520 is not on the roof in the roof true orthophoto 510 . As an example, the computing device 100 may extract the feature points of the roof 530 that are closest to the position of the seed 520 and then reset the position of the seed 520 to the position with the highest density of feature points.

Referring again to FIG. 3, at step 340, the computing device 100 may load a portion of the drone image closest to the reset seed location. A portion of the drone image may mean a partial piece image cut from the corresponding drone image.

At step 350, the computing device 100 may process feature point matching of the loaded portion of the drone image and the roof true orthophoto. As an example, the computing device 100 may match a portion of the loaded drone image and points (feature points) of the roof true orthophoto by R2D2 (Repeatable and Reliable Detector and Descriptor) matching. In addition to R2D2, point matching may utilize one of well-known image matching techniques such as SIFT (Scale-Invariant Feature Transform), HardNet, LogPolarDesc, and the like. However, since R2D2 is a matching technique trained to improve the performance of matching two images with different visuospatial resolutions, it considers the difference in acquisition time and spatial resolution between true orthophotos and drone images. Sometimes it is preferable to handle point matching using R2D2. At this time, the computer device 100 may match n (n is 100, for example) points.

At step 360, computing device 100 may select at least one matching point from the portion of the drone image. As an example, the computer device 100 may select the most appropriately matched point among the plurality of matched points. As an example, when performing feature point matching, the computing device 100 may perform matching of feature points having the most similar pixel values around the feature points in the two images. The similarity of pixel values around such a feature point is called a distance, and it may be determined that the larger the distance value, the lower the similarity, and the smaller the distance value, the higher the similarity. Therefore, the computer device 100 may select the matching point with the smallest distance value among the matched points as the most appropriately matched point.

FIG. 6 is a diagram showing an example of matching drone images and roof true orthophotos in one embodiment of the present invention. FIG. 6 shows an example of connecting corresponding points in a drone image 610 and a roof true orthophoto 620 . At this time, in FIG. 6, the line 630 connecting the most appropriately matched points among the plurality of points is shown thicker than the other lines. At this point, points 640, 650 connected by line 630 may be selected. FIG. 6 simplifies the description of the points to be matched, but as described above, after matching the n points, the computer apparatus 100 selects the most appropriate matching point from among the n matched points. You may choose the point made. Here, matching n points may mean matching n points based on the drone image 610 with n points of the roof true orthophoto 620 .

Referring again to FIG. 3, at step 370, computing device 100 selects the three-dimensional coordinates of the actual terrain for the position of the roof true orthophoto corresponding to the selected matching point as the virtual GCP for the selected matching point. You can In this way, in the roof true orthophoto obtained from the existing aerial map, the computer device 100 can determine the three-dimensional coordinates (X , Y, Z) can be obtained.

FIG. 7 is a diagram illustrating an example of selecting 3D coordinates corresponding to coordinates of a drone image in an embodiment of the present invention. FIG. 7 selects the image position (x, y) of the point of the drone image 610 and the three-dimensional coordinates (X, Y, Z) of the real terrain corresponding to the matched point of the roof true orthophoto 620 as the virtual GCP. You can Here, the three-dimensional coordinates (X, Y, Z) of the real terrain corresponding to the matched points of the roof true orthophoto 620 may be obtained by the DSM. More specifically, X and Y may be obtained from the true orthophoto, and since the DSM contains height values for each pill cell, we can extract the height values from the DSM locations corresponding to the true orthophoto. Z may be obtained.

Referring back to FIG. 2 , in step 230 , the computer device 100 inputs GCPs to other drone images based on the matched drone image, so that the computer device 100 inputs peripheral sub-drone images and virtual images based on the matched drone image. Concatenation of GCPs may be set.

FIG. 8 is a diagram showing an example of setting a connection between a subdrone image and a virtual GCP in one embodiment of the present invention. FIG. 8 shows the image positions of the sub-drone images ( x, y) and three-dimensional coordinates (X, Y, Z) of the virtual GCP are shown. As an example, the computer device 100 may match the points of the sub-drone images 830 that are the same as the points where the main drone image 810 is connected to the virtual GCP of the roof true orthophoto 820, thereby matching the points of the sub-drone images 830 with virtual GCPs may be linked. In other words, the computing device 100 may input the three-dimensional coordinates of the matched points of the roof true orthophoto 820 to the points of the main drone image 810 that are matched to the roof true orthophoto 820 as virtual GCPs. A virtual GCP may be input to each of the other sub-drone image 830 points that match the main drone image 810 point. At this time, the same points of the main drone image 810 and the sub drone image 830 may be matched by dense matching.

9 and 10 show an example of matching the same point by dense matching in one embodiment of the present invention. FIG. 9 shows an example of a process of matching the same point of a main drone image 910 and a sub-drone image 920 using dense matching, and FIG. An example of matching the same point of sub images 1020 and 1030 is shown. Here, the main image 1010 may correspond to the main drone image, and the sub-images 1020, 1030 may correspond to the sub-drone images.

Referring again to FIG. 2, at step 240, computing device 200 may process the bundle adjustment to filter the GCPs. At this time, the GCPs may be automatically entered into the drone image by performing steps 210-230. Therefore, there is no need for a human to actually go out to survey the GCP, and there is no need to manually match the surveyed GCP with the drone image, thereby reducing costs and time.

However, since the automatically input virtual GCP (accuracy 8-15 cm) is less accurate than the surveyed GCP (accuracy 2-3 cm), multiple filtering may be applied. Since drone images have less accurate initial values than aerial images, the computing device 100 may process the 7-parameters to correct the GCP. 7-parameters is a technique called Rigid Transformation, which adjusts the scale, rotation, and translation of two identical models to fit them in the same way. In the present embodiment, three-dimensional coordinates (X, Y, Z) are obtained by triangulating GCPs (X, Y, Z) extracted from an aerial map and GCPs matched with drone images. and three-dimensional coordinates must be at the same position. However, because drone footage uses low-cost GPS, the 3D coordinates may indicate a different location than the GCP. At this time, scale, rotation, and movement are calculated so that the 3D coordinates match the GCP position by 7-parameters processing. It is possible to estimate the position of the drone photographed at the time similar to the terrain. While processing the 7-parameters, the computing device 100 may consider the evaluation for each of X, Y, and Z in the three-dimensional coordinates of the virtual GCP with weighted values. As an example, computing device 100 may correct the weight values to perform self-calibration. At this time, the computing device 100 may place a stronger weight on the side of the image, and give a stronger weight on the farther from the density of the feature points.

11 and 12 are diagrams showing an example of correcting weight values in one embodiment of the present invention. FIG. 11 shows that covariance decreases as points are located on the side of the image. When self-calibrating, the key is to correct for image distortion, which usually occurs at the end of the image. Therefore, the distortion can be corrected more accurately by placing less covariance on the points at the end of the image so that they are more affected when correcting the distortion.

FIG. 12 shows that the weight increases as the density of feature points decreases.

Equation (1) below shows an example of a basic equation for bundle adjustment.

ここで、ｘｇｃｐはドローン映像のピクセルの位置点を示してよく、Ｋはドローンカメラの内部焦点距離および歪曲係数を含む情報を含んでよく、Ｒはドローンカメラの回転情報を、Ｇは３次元座標（ＧＣＰ）に対応する映像座標を示してよい。ρは損失関数（ｌｏｓｓ ｆｕｎｃｔｉｏｎ）を、ｅは費用関数（ｃｏｓｔ ｆｕｎｃｔｉｏｎ）を、Ｐはプロジェクション関数（ｐｒｏｊｅｃｔｉｏｎ ｆｕｎｃｔｉｏｎ）を、Ｗｇｃｐは加重値をそれぞれ示してよい。Ｇは外部に出向いて直接測量したデータではないため、正確度で８～１５ｃｍの差が生じることがある。これにより、ＧＣＰを完全なトゥルー（ｔｒｕｅ）として置かず、多少の調整が可能な未知数として置いて方程式を解くようになる。言い換えれば、Ｇは初めに得られたＧＣＰであってよく、

Here, xgcp may indicate the pixel position point of the drone image, K may contain information including the internal focal length and distortion coefficient of the drone camera, R the rotation information of the drone camera, and G the three-dimensional coordinates. The image coordinates corresponding to (GCP) may be indicated. ρ may denote a loss function, e a cost function, P a projection function, and Wgcp a weighted value. Since G is not directly surveyed by going to the outside, there may be a difference of 8 to 15 cm in accuracy. This leads to solving the equations without putting the GCP as a perfect true, but as an unknown that can be adjusted somewhat. In other words, G may be the originally obtained GCP,

は推定されたＧＣＰであってよい。推定されたＧＣＰと初めに得られたＧＣＰの差が極端に開かないようにするために加重値（Ｗｇｃｐ）を置いて推定してよい。このように、ピクセルの位置点ｘｇｃｐは事前マッチングによって既に決定されているため、コンピュータ装置１００は、ピクセルの位置点ｘｇｃｐを利用してＧＣＰに対応する映像座標との差が最小となるＲ、Ｋを決定してよい。追加的に、コンピュータ装置１００は、推定された（または、補正されたＧＣＰである）

may be the estimated GCP. A weighted value (Wgcp) may be placed in the estimation so that the difference between the estimated GCP and the initially obtained GCP does not become too large. In this way, since the pixel location point xgcp has already been determined by pre-matching, the computer device 100 uses the pixel location point xgcp to determine the R, K coordinates that minimize the difference from the image coordinates corresponding to the GCPs. may be determined. Additionally, the computing device 100 is the estimated (or corrected GCP)

を決定してよい。言い換えれば、コンピュータ装置１００は、マッチングされたドローン映像のピクセルの位置点に基づいて、ＧＣＰに対応する映像座標との差が最小となるドローンカメラの内部焦点距離、歪曲係数、ドローンカメラの回転情報、および補正されたＧＣＰを決定してよい。

may be determined. In other words, the computer device 100 obtains the internal focal length of the drone camera, the distortion coefficient, and the rotation information of the drone camera that minimizes the difference from the image coordinates corresponding to the GCP based on the pixel position points of the matched drone image. , and a corrected GCP may be determined.

On the other hand, the loss function may be expressed as Equation (2) below.

ここで、ｒはエラーを、δは予め設定された閾値をそれぞれ示してよい。また、ｃ’は数式（３）のように、ｓ’は数式（４）のように表現されてよい。

Here, r may indicate an error, and δ may indicate a preset threshold. Also, c' may be expressed as in Equation (3), and s' may be expressed as in Equation (4).

Equation (2) may be used to fix the weighted value to a constant value when the error is less than a preset threshold.

By filtering the virtual GCPs in the computing device 100 in this way, even if there are some outliers, the drone image can be properly mapped on the existing aerial map.

FIG. 13 is a diagram showing the difference in resolution between the drone image and the aerial image in one embodiment of the present invention, and FIG. 14 is a diagram showing the update of the aerial map using the drone image in one embodiment of the present invention. It is the figure which showed the example which carried out.

FIG. 13 shows that the drone image (Drone) 1310 has a higher resolution than the aerial image (Aerial) 1320, and FIG. 14 shows that the drone image with relatively higher resolution is used to update the aerial map. shows an example of increased resolution of an area of an aerial map updated using drone footage.

Thus, according to the embodiment of the present invention, when generating and/or updating an aeronautical map using drone images, an existing three-dimensional aeronautical map is used as a GCP to generate and/or update an aeronautical map. Time and cost for updates can be reduced. In addition, manpower and time can be saved by supporting automatic matching of 3D aerial maps used as GCP and drone images.

The systems or devices described above may be realized by hardware components or a combination of hardware and software components. For example, the devices and components described in the embodiments may include, for example, processors, controllers, ALUs (arithmetic logic units), digital signal processors, microcomputers, FPGAs (field programmable gate arrays), PLUs (programmable logic units), microcontrollers, It may be implemented using one or more general purpose or special purpose computers, such as a processor or various devices capable of executing instructions 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, virtual device, computer storage medium or device for interpretation on or for providing instructions or data to a processing device. may be changed. 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. The computer-readable media may include program instructions, data files, data structures, etc. singly or in combination. The medium may be a continuous recording of the computer-executable program or a temporary recording 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. Examples of program instructions include high-level language code that is executed by a computer, such as using an interpreter, as well as machine language code, such as that generated by a compiler.

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

Claims (20)

A method of updating a computing device comprising at least one processor, comprising:
generating, by the at least one processor, a roof true orthophoto containing images above a preset height using existing aerial map-based DEMs, DSMs, and true orthophotos;
matching the drone image with the roof true orthophoto by the at least one processor; and inputting GCPs to other drone images based on the matched drone image by the at least one processor. An update method characterized by: Generating the roof true orthophoto comprises:
2. The update method according to claim 1, wherein the roof true orthophoto is generated by extracting from the true orthophoto a portion where the difference between the DSM and the DEM is equal to or greater than a preset height. The step of matching the drone image and the roof true orthophoto includes:
processing feature point matching between the drone image and the roof true orthophoto;
selecting at least one matching point from a portion of said drone image; and 3D coordinates of actual terrain relative to a position of said roof true orthophoto corresponding to said selected matching point. 3. The update method according to claim 1 or 2, comprising the step of selecting as a virtual GCP for . Processing the feature point matching includes:
4. The update method of claim 3, wherein the feature points of the drone image and the feature points of the roof true orthophoto are matched using R2D2 matching. The step of matching the drone image and the roof true orthophoto includes:
setting a seed based on a preset length of the range to be updated with the roof true orthophoto;
loading the closest roof true orthophoto from the set seed;
After extracting feature points from the loaded roof true orthophoto, resetting the set seed positions based on the density of the extracted feature points; and resetting the seed positions. 4. The update method of claim 3, further comprising: loading a portion of the drone image closest to the . Setting the seed comprises:
A preset number of seeds are randomly set for each of the plurality of squares in a range composed of a plurality of squares having a side length of the preset length. The update method according to Item 5. The resetting step includes:
6. The update method according to claim 5, wherein the position of the seed is reset to a position where the density of the extracted feature points is highest. The step of inputting GCPs to the another drone image includes:
inputting the 3D coordinates of the matched point of the roof true orthophoto to the matched drone image point as a virtual GCP; and the other drone image point matching the matched drone image point. The update method according to any one of claims 1 to 7, characterized in that it comprises the step of inputting the virtual GCPs into the . The step of inputting the virtual GCPs includes:
The update method of claim 8, wherein the matched drone image points and the other drone image points are matched using dense matching. The updating method according to any one of claims 1 to 9, further comprising: filtering said GCPs by said at least one processor. Filtering the GCPs comprises:
Based on the pixel position point of the matched drone image, the internal focal length of the drone camera that minimizes the difference from the image coordinates corresponding to the GCP, the distortion coefficient, the rotation information of the drone camera, and the corrected 11. The updating method according to claim 10, characterized in that GCPs are determined. generating, by the at least one processor, an updated DSM based on an input drone image of the filtered GCPs; and updating true, by the at least one processor, based on the updated DSM. 11. The updating method of claim 10, further comprising: generating an orthophoto. A computer program for causing a computer device to perform the method according to any one of claims 1-12. A computer-readable recording medium recording a computer program for causing a computer device to execute the method according to any one of claims 1 to 12. at least one processor implemented to execute computer readable instructions;
by the at least one processor;
Utilizing existing aerial map-based DEM, DSM, and true orthophotos to generate roof true orthophotos containing footage above a preset height,
Matching the drone image and the roof true orthophoto,
A computer device, characterized in that, based on the matched drone image, a GCP is input to another drone image. by the at least one processor to generate the roof true orthophoto;
16. The computer apparatus according to claim 15, wherein the roof true orthophoto is generated by extracting a portion where the difference between the DSM and the DEM is equal to or greater than a preset height from the true orthophoto. for matching the drone imagery and the roof true orthophoto, by the at least one processor;
processing feature point matching between the drone image and the roof true orthophoto;
selecting at least one matching point from the portion of the drone image;
17. The method according to claim 15 or 16, characterized in that the three-dimensional coordinates of the real terrain for the position of the roof true orthophoto corresponding to the selected matching point are selected as the virtual GCP for the selected matching point. The computer device described. for matching the drone imagery and the roof true orthophoto, by the at least one processor;
setting a seed based on a preset length of the range to be updated with the roof true orthophoto;
load the closest roof true orthophoto from the set seed,
After extracting feature points from the loaded roof true orthophoto, resetting the set seed positions based on the density of the extracted feature points;
18. The computer device of claim 17, wherein the portion of the drone image closest to the reset seed location is loaded. By the at least one processor for resetting the set seed positions based on the density of the extracted feature points,
19. The computer device according to claim 18, wherein the position of the seed is reset to a position where the density of the extracted feature points is highest. by the at least one processor to input GCPs into the other drone imagery;
inputting the three-dimensional coordinates of the matched points of the roof true orthophoto to the matched points of the drone image as a virtual GCP;
The computer device according to any one of claims 15 to 19, characterized in that it inputs the virtual GCPs to the other drone image points that match the matched drone image points.

Images