KR102393300B1 - Object map generation system and method therefor - Google Patents

Abstract

The present invention relates to an object map generation system and method therefor. Unlike conventional aerial maps, which take a relatively long time to produce, when creating an aerial map, ground objects and GPS coordinates are displayed through a drone and an object detector. Also, when an object map is needed, just by flying a drone, an aerial map of an area around an operator can be created and object information can be quickly grasped. The object map generation system comprises: an object detection module receiving ground-captured images and detailed information of capturing images from a drone and detecting an object from the received ground-captured images; an image processing module acquiring GPS coordinates of the detected object and mapping the obtained GPS coordinates of the object to the captured image; and an object map matching module creating an object map by matching the mapped captured images.

Description

Object map generation system and method therefor

The present invention relates to a system and method for generating an object map, and more particularly, to an object map generating system and method for displaying object information and GPS coordinates of objects on the ground when generating an aerial map through aerial photography will be.

The process of making a numerical map usually includes aerial photography, close-up photos and film production, ground reference point surveying, photo reference point surveying, numerical drawing, geographic survey and supplementary surveying, in-place editing, and structured editing processes.

First, aerial photography is the process of obtaining image information about the topographical features on the ground in the air using an aircraft equipped with a camera, which is the most important part of aerial photogrammetry. It is a factor that greatly affects the accuracy of the data.

Like general photos, the film goes through detailed processes such as develop, fixation, flushing, aviation film, plotting work, security inspection, film cycle insertion, adhesion and film production, etc. It is produced with positive film and close-up photos necessary for the painting process.

Ground reference point surveying is a ground survey conducted locally to obtain the results of reference points required for photo reference point surveying and interpretation drawing work.

Photo reference point surveying refers to the operation of converting the model (machine) coordinates of the sphere measured on the aerial photograph by means of a plotter or a coordinate measuring device with the ground reference point survey performance and converting it into ground coordinates.

Numerical drawing is the process of measuring the geographical features on the surveyed aerial photos as digital data using an analysis plotter and recording them with a computer. After installing the transparent positive film of the left and right photos constituting a model on the projector and shining light on it, the intersections of the light beams reproduce the same three-dimensional model as the real terrain, find the expression points and input their absolute coordinates on the drawing machine After tracing the desired feature with a Measuring Mark or Floating Mark, the principle of drawing on the topographic map is used accordingly.

Here, in principle, the drawing follows the standard code that is classified by layer in the order of buildings, roads, vegetation, rivers, and contours, and the standard diagram that defines the shape, size, structure and direction of figures.

Geographical survey refers to the on-site investigation of the topography and features to be shown on the drawing and related matters based on the geographic/geographical survey and aerial photos for exact location editing. This geographical survey work is accurately marked by using a drawing source map printed with digital data as a drawing and a double-magnified photograph.

Complementary surveying is a modification of the topographical features at the time of shooting and the topographical changes after the completion of the drawing. At this time, according to the "Field Survey Marking Guidelines", the contents of the field investigation and the data collected are arranged and edited in the order of topography, features, geographical names, etc.

And, as described above, the production of numerical maps is based on aerial photographed images. Since it is not possible to photograph the entire map production point with one photograph, several images are taken during aerial photographing, and the plurality of photographed images We will connect and proceed with the painting work.

At this time, since the aerial photographed image is usually obtained through an aerial photographing device installed in an aircraft, blind spots in which it is difficult to obtain an aerial photographed image are inevitably generated while being obscured by a relatively high building or topographical feature, and aerial photographing of these blind spots For image acquisition, it was not efficient to acquire aerial photographed images through the conventional aerial photographing method for each blind spot.

Therefore, for blind spots where it is difficult to obtain aerial photographic images through the aerial photographing device of the aircraft, supplementary photographing work is carried out through the ground photographing equipment installed in the vehicle or the ground photographing equipment carried and operated by the worker. A technology for acquiring aerial photographed images using drones is provided.

After all, aerial maps are generated by stitching ground images taken from the sky in a grid form. Recently, drones that can be operated at a relatively low cost are being used because they have several disadvantages.

Therefore, the prior patents related to map generation using drones to solve the above problems (Registration Patent 10-1692709, Registration Patent 10-1914164, Registration Patent 10-1925366) are mainly for the purpose of numerical map production. , it is somewhat difficult to display the recent appearance of objects located on the ground due to the nature of this numerical map to display terrain or roads.

In addition, the prior patents have disadvantages in that the manufacturer has to directly display key points or objects, and it is difficult to use them as an electronic map because the GPS coordinates of the objects cannot be known.

Republic of Korea Patent Registration 10-16927090000 (2016. 12. 29) Republic of Korea Patent Registration 10-19141640000 (2018. 10. 26) Republic of Korea Patent Registration 10-19253660000 (2018. 11. 29)

Therefore, it is an object of the present invention to display an object and GPS coordinates on the ground through a drone and an object detector when generating an aerial map. An object map generation system and method are provided that allow the user to generate an aerial map of the area around the operator just by flying the drone and quickly grasp object information.

That is, an object of the present invention is to provide an object map generation system and method for producing an aerial map using a drone, displaying the main objects of the aerial map, acquiring GPS coordinates of each detected object, mapping them, and displaying them together there is a purpose to

The technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention belongs from the description below. will be able

An object map generating system according to a preferred embodiment of the present invention for achieving the above object includes an object detection module for detecting an object from a ground-captured image among ground-captured images and detailed image capturing information wirelessly transmitted from a drone; an image processing module that acquires the GPS coordinates of the detected object and maps the acquired GPS coordinates of the object to the captured image; and an object map matching module configured to generate an object map by matching the mapped captured images.

The object detection module may include: a receiver configured to receive a ground-captured image and image capturing detailed information transmitted from the drone, and provide the received image capturing detailed information to the image processing module; and an object detection unit configured to provide object detection result information to the image processing module after detecting at least one object from the ground-captured image received through the receiving unit.

The detailed image capturing information received from the drone may include at least one of altitude information at the time of capturing, focal length information of a capturing camera, and GPS coordinate information of the capturing point.

The GPS coordinate information of the photographing point may be GPS coordinates of a central region of the photographed image.

The object detection result information may include information on at least one of a probability of an object, a label, and coordinate information of four vertices of a rectangular box indicating an object area from the input image.

The object detection unit may detect an object using a Convolutional Neural Networks (CNNs)-based object detector such as a Single Shot Multibox Detector (SSD) or You Only Look Once (YOLO).

The image processing module may include: a scale calculator configured to calculate a scale of a captured image using height information of a photographing time point provided from the receiver of the object detection module and focal length information of a photographing camera; Using the object detection result information provided from the object detection unit of the object detection module, the GPS coordinate information of the photographing point provided from the receiver, and the scale value of the photographed image calculated from the scale calculation unit, GPS for the center point of each object an object GPS coordinate acquisition unit that acquires coordinates; and an image mapping unit that maps the detection result of the detected object and the GPS coordinates of each object to the captured image provided from the receiver of the object detection module, and then provides the mapped image to the object map matching module.

The scale of the captured image calculated by the scale calculator may use the following equation.

[ceremony]

은 축척의 분모수,

는 카메라 렌즈의 초점거리,

는 촬영 고도를 나타낸다. here,

is the denominator of the scale,

is the focal length of the camera lens,

represents the shooting altitude.

The object map matching module may include: a map matching unit configured to generate an object map by matching a plurality of mapped images provided from the image mapping unit; and a map output unit that provides the object map generated through the map matching unit to the operator.

The image registration in the map matching unit may be performed using at least one of a keypoint-based method, which is a regional feature extracted from an image, and a registration method based on geometrical positions of cameras and lenses.

Meanwhile, a method for generating an object map according to another embodiment of the present invention includes: detecting an object from a ground-captured image among ground-captured images and detailed image capturing information wirelessly transmitted from a drone; obtaining GPS coordinates of the detected object, and mapping the acquired GPS coordinates of the object to the captured image; and generating an object map by matching the mapped captured images.

The detecting of the object may include: receiving a ground-captured image and detailed image capturing information wirelessly transmitted from the drone; and detecting at least one object from the received ground-captured image.

The detailed image capturing information received from the drone may include at least one of altitude information at the time of capturing, focal length information of a capturing camera, and GPS coordinate information of the capturing point.

The GPS coordinate information of the photographing point may be GPS coordinates of a central region of the photographed image.

The object detection result information may include information on at least one of a probability of an object, a label, and coordinate information of four vertices of a rectangular box indicating an object area from the input image.

The detecting of the object may include detecting the object by using a Convolutional Neural Networks (CNNs)-based object detector such as a Single Shot Multibox Detector (SSD) or You Only Look Once (YOLO).

The mapping to the captured image may include: calculating a scale of the captured image using altitude information of a capturing time received from the drone and focal length information of a capturing camera; obtaining GPS coordinates for the center point of each object by using the object detection result information, GPS coordinate information of the photographing point received from the drone, and the scale value of the photographed image calculated from the scale calculator; and mapping the detection result of the detected object and the GPS coordinates of each object to the captured image.

In the step of calculating the scale, the scale of the captured image may use the following equation.

[ceremony]

은 축척의 분모수,

는 카메라 렌즈의 초점거리,

는 촬영 고도를 나타낸다. here,

is the denominator of the scale,

is the focal length of the camera lens,

represents the shooting altitude.

The generating of the object map may include: generating an object map by matching the plurality of mapped images; and outputting the generated object map so that an operator can check it, wherein the image registration is at least one of a keypoint-based method that is a regional feature extracted from an image, and a registration method based on geometrical positions of cameras and lenses. can be used to match.

According to the present invention, when an aerial map is created, ground objects and GPS coordinates are displayed through a drone and an object detector, that is, an aerial map is produced using the drone, main objects of the aerial map are displayed, and the GPS coordinates of each detected object are displayed. Unlike the existing aerial map, which takes a relatively long time to produce by acquiring, mapping and displaying the so that it can be easily understood.

Effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

The accompanying drawings below are provided to help understanding of the present embodiment, and provide embodiments together with detailed description. However, the technical features of the present embodiment are not limited to specific drawings, and features disclosed in the drawings may be combined with each other to constitute a new embodiment.
1 is a diagram showing a schematic block configuration of an object map generating system according to the present invention.
FIG. 2 is a diagram showing a detailed block configuration of the object map generating system shown in FIG. 1 .
3 is a diagram showing an operation flowchart of a method for generating an object map according to the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the present embodiments allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully understand the scope of the present invention to those skilled in the art, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. In this specification, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used herein will have the meaning commonly understood by those of ordinary skill in the art to which this invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless specifically defined explicitly.

Hereinafter, an object map generating system and method according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram showing a schematic block configuration of an object map generating system according to the present invention.

Referring to FIG. 1 , an object map generating system 200 according to the present invention may include an object detection module 210 , an image processing module 220 , and an object-object map matching module 230 . Here, the object map generating system 200 according to the present invention may receive a captured image and detailed image capturing information from the drone 100 by wirelessly interworking with the drone 100 .

First, the drone 100 shoots the ground image in the air according to the wireless transmission of flight interval, shooting interval, and altitude information wirelessly transmitted through the remote (ground) operator's control terminal, and then shoots the captured image and image. The detailed information on the information is wirelessly transmitted to the object map generating system 200 located on the ground. Here, the detailed information on image capturing transmitted from the drone 100 may include at least one of altitude information at the time of shooting, focal length information of the shooting camera, and GPS coordinate information of the shooting point. The GPS coordinate information of may be GPS coordinates of the central region of the captured image.

The object detection module 210 of the object map generating system 200 detects an object from the ground-captured image wirelessly transmitted from the drone 100, and detects object information, photographed image information, and altitude information at the time of photographing, and photographing. The focus distance information of the camera and the GPS coordinate information of the shooting point are provided to the image processing module 220 .

Meanwhile, the object detection module 210 may provide the GPS coordinate information of the photographing point to the object map matching module 230 .

The object detection in the object detection module 210 detects an object using a Convolutional Neural Networks (CNNs)-based object detector such as a Single Shot Multibox Detector (SSD) and You Only Look Once (YOLO), and from the input image. The probability of the object, the label, and the coordinates of the four vertices of the rectangular box indicating the object area are provided to the image processing module 220 .

The image processing module 220 acquires the GPS coordinates of the detection object provided from the object detection module 210 , maps the GPS information of the detection object obtained together with the detected object information to the captured image, and then maps the object and GPS coordinates The mapped image is provided to the object map matching module 230 .

That is, the image processing module 220 calculates the scale of the captured image using the photographing altitude information and focal length information provided from the object detection module 210 , and calculates the scale value and the object detection result (probability of detected objects). , the label, and the four vertex coordinate information of the rectangular box indicating the object area) and the GPS coordinates of the shooting point to obtain the GPS coordinates of the detection object, and the obtained GPS information of the detection object together with the detected object information. After mapping to , the image to which the object and GPS coordinates are mapped is provided to the object map matching module 230 . Here, the scale calculation of the captured image will be described in detail with reference to FIG. 2 .

Meanwhile, each operation of the object detection module 210 and the image processing module 220 repeatedly operates in the same manner with respect to the image transmitted from the drone 100 .

The object map matching module 230 receives the images to which the object information provided from the image processing module 220 and the GPS coordinates of each object are mapped, and determines the timing of the shooting of the captured image provided from the object detection module 220 . After receiving GPS coordinate information, matching the respective mapping images in order, and outputting the matched object map.

A detailed configuration and detailed operation of the object map generating system 200 according to the present invention will be described with reference to FIG. 2 .

2 is a view showing the detailed block configuration of the object map generating system 200 shown in FIG. 1, that is, the object detection module 210, the image processing module 220 and the object map matching module shown in FIG. 230) is a diagram showing a detailed block configuration.

Referring to FIG. 2 , the object detection module 210 may include a receiver 211 and an object detector 212 .

The image processing module 220 may include an image mapping unit 221 , an object GPS coordinate obtaining unit 222 , and a scale calculating unit 223 .

The object map matching module 230 may include a map matching unit 231 and a map output unit 232 .

First, the receiver 211 of the object detection module 210 receives captured image information and image capturing detailed information transmitted from the drone 100 . Here, the image capturing detailed information may include at least one of altitude information at the time of capturing, focal length information of the capturing camera, and GPS coordinate information of the image capturing point, and the GPS coordinate information of the capturing point is the captured image. may be the GPS coordinates of the central region of .

The receiver 211 may provide the image information received from the drone 100 to the object detector 212 and the image mapping unit 221 of the image processing module 220 .

And, the receiving unit 211 provides the altitude information of the shooting time and focal length information of the shooting camera among the detailed image shooting information received from the drone 100 to the shooting image scale calculation unit 223 of the image processing module 220 and , the GPS coordinate information of the image capturing point among the detailed image capturing information may be provided to the object GPS coordinate obtaining unit 222 of the image processing module 220 and the map matching unit 231 of the object map matching module 230, respectively. .

The object detection unit 212 of the object detection module 210 detects object information from the captured image information provided from the receiver 211 using a CNNs (Convolutional Neural Networks)-based object detector such as SSD and YOLO, and then detects The obtained object information may be provided to the object GPS coordinate obtaining unit 222 of the image processing module 220 . Here, the object information may be at least one piece of information among four vertex coordinates of a rectangular box indicating a probability of objects, a label, and an object area.

On the other hand, the photographed image scale calculation unit 223 of the image processing module 220 uses the altitude information of the photographing point provided from the receiving unit 211 of the object detection module 210 and the focal length information of the photographing camera of the photographed image. After calculating the scale value, the calculated scale value of the captured image may be provided to the object GPS coordinate obtaining unit 222 of the image processing module 220 . Here, the scale value of the captured image may be calculated using the following equation.

[ceremony]

은 축척의 분모수,

는 카메라 렌즈의 초점거리,

는 촬영 고도를 나타낸다. here,

is the denominator of the scale,

is the focal length of the camera lens,

represents the shooting altitude.

For example, the scale of a photo taken at 350 m using a 35 mm focal length lens may be 1/1000.

Meanwhile, the object GPS coordinate obtaining unit 222 of the image processing module 220 includes the scale value of the captured image calculated by the captured image scale calculating unit 223 and the GPS of the capturing point provided from the object detection module 210 . After acquiring the GPS coordinates for the central points of the objects provided from the object detection unit 212 of the object detection module 210 by using the coordinate values, the obtained GPS coordinate values for the respective objects are combined with the image mapping unit 221 . can be provided as

In addition, the image mapping unit 221 of the image processing module 220 uses the coordinate values for each object provided from the object GPS coordinate obtaining unit 222 and the object detection unit 211 of the object detection module 210 . After mapping each detected object information to the captured image provided from the receiver 211 of the object detection module 210 , the map matching unit 231 of the object map matching module 230 maps the image to which the object and GPS coordinates are mapped. ) can be provided.

Such an operation may be repeatedly performed in the same way while a captured image is provided from the drone 100 .

The map matching unit 231 of the object map matching module 230 generates an object map by matching the object provided from the image processing module 22 and a plurality of images to which the GPS coordinates of the objects are mapped, and the generated object The map data is provided to the map output unit 232 . Here, for the image registration, a keypoint-based method, which is a regional feature extracted from a captured image, or a registration method based on geometrical positions of cameras and lenses, etc. may be used. This method is a known technique, and detailed description thereof will be omitted.

The map output unit 232 may convert the object map data provided from the map matching unit 231 into data that the operator can check, that is, display it on the display screen, and output the converted data.

An object map generating method according to the present invention corresponding to the operation of the object map generating system according to the present invention will be described in stages with reference to FIG. 3 .

3 is a diagram illustrating an operation flowchart of a method for generating an object map according to the present invention.

Referring to FIG. 3 , first, the drone 100 shoots an image on the ground while flying in a specific area according to the operator's flight control, and then transmits the captured image and detailed image capturing information to the object map generating system 200 . do.

The object map generating system 200 receives captured image information and image capturing detailed information transmitted from the drone 100 (S301). Here, the image capturing detailed information may include at least one of altitude information at the time of capturing, focal length information of the capturing camera, and GPS coordinate information of the image capturing point, and the GPS coordinate information of the capturing point is the captured image. may be the GPS coordinates of the central region of .

Next, the object map generating system 200 detects object information from the captured image information received from the drone 100 using a CNNs (Convolutional Neural Networks)-based object detector such as SSD and YOLO (S302). Here, the detected object information may be information on at least one of a probability of objects, a label, and information on coordinates of four vertices of a rectangular box indicating an object area.

Next, the object map generating system 200 calculates a scale value of the captured image by using the altitude information of the photographing time and focal length information of the photographing camera provided from the drone 100 ( S303 ). Here, the scale value of the captured image may be calculated using the above equation.

Then, the object map generating system 200 detects through step S302 using the scale value of the captured image calculated through step S303 and the GPS coordinate value of the photographing point received from the drone 100 through step S301. The GPS coordinates of the central points of the objects are acquired (S304).

Next, each of the GPS coordinate values for each object obtained in step S304 and information on each object detected in step S302 are added to the captured image received in step S301, that is, transmitted from the drone 100, respectively. Mapping (S305).

Next, it is determined whether a mapping operation has been performed on all the captured images transmitted from the drone 100 (S306).

As a result of the determination, if the mapping operation is not performed on all images, steps S301 to S305 may be repeatedly performed until the mapping operation is performed on all images.

As a result of the determination, when the mapping operation is performed on all images transmitted from the drone 100, the object detected in step S302 and a plurality of images to which GPS coordinates for all objects are mapped through step S305 They are matched to create an object map (S307). Here, for the image registration, a keypoint-based method, which is a regional feature extracted from a captured image, or a registration method based on geometrical positions of cameras and lenses, etc. may be used.

Next, the object map generated through the step S307 is output (S308). That is, the object map data matched through the step S307 can be converted into data that the operator can check, that is, can be displayed on the display screen, and output.

The method for generating an object map according to an embodiment of the present invention described above may be implemented as a program (or application) and stored in a medium to be executed in combination with a computer, which is hardware.

The above-mentioned program, in order for the computer to read the program and execute the methods implemented as a program, C, C++, JAVA, Ruby, which the processor (CPU) of the computer can read through the device interface of the computer; It may include code coded in a computer language such as machine language. Such code may include functional code related to a function defining functions necessary for executing the methods, etc., and includes an execution procedure related control code necessary for the processor of the computer to execute the functions according to a predetermined procedure. can do. In addition, the code may further include additional information necessary for the processor of the computer to execute the functions or code related to memory reference for which location (address address) in the internal or external memory of the computer to be referenced. there is. In addition, when the processor of the computer needs to communicate with any other computer or server located remotely in order to execute the above functions, the code uses the communication module of the computer to determine how to communicate with any other computer or server remotely. It may further include a communication-related code for whether to communicate and what information or media to transmit and receive during communication.

The storage medium is not a medium that stores data for a short moment, such as a register, a cache, a memory, etc., but a medium that stores data semi-permanently and can be read by a device. Specifically, examples of the storage medium include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. That is, the program may be stored in various recording media on various servers accessible by the computer or in various recording media on the computer of the user. In addition, the medium may be distributed in a computer system connected to a network, and a computer-readable code may be stored in a distributed manner.

The description of the present invention described above is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a dispersed form, and likewise components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

100: drone
200: object map generation system
210: object detection module
211: receiver
212: object detection unit
220: image processing module
221: image mapping unit
222: object GPS coordinate acquisition unit
223: shot image scale calculation unit
230: object map matching module
231: map matching unit
232: map output unit

Claims (20)

An object map generation system comprising:
an object detection module configured to detect an object from a ground-captured image among ground-captured images and detailed image capturing information wirelessly transmitted from a drone;
an image processing module that acquires the GPS coordinates of the detected object and maps the acquired GPS coordinates of the object to the captured image; and
and an object map matching module for generating an object map by matching the mapped captured images;
The object detection module,
a receiving unit for receiving the ground-captured image and image capturing detailed information transmitted from the drone, and providing the received image capturing detailed information to the image processing module;
and an object detection unit configured to provide object detection result information to the image processing module after detecting at least one object from the ground-captured image received through the receiving unit;
The detailed image capturing information received from the drone is altitude information at the time of shooting, focal length information of the shooting camera, and GPS coordinate information of the shooting point, and the object detection result information includes:
It includes at least one or more of information about the probability, label, and coordinates of the four vertices of a rectangular box indicating the object area from the input image,
The image processing module,
a scale calculator configured to calculate a scale of a captured image using altitude information at a time of capturing and focal length information of a capturing camera provided from the receiving unit of the object detection module;
Using the object detection result information provided from the object detection unit of the object detection module, the GPS coordinate information of the photographing point provided from the receiver, and the scale value of the photographed image calculated from the scale calculation unit, GPS for the center point of each object an object GPS coordinate acquisition unit that acquires coordinates; and
Object map comprising an image mapping unit that maps the detection result of the detected object and the GPS coordinates of each object to the captured image provided from the receiving unit of the object detection module, and then provides the mapped image to the object map matching module generation system.
delete delete According to claim 1,
The GPS coordinate information of the shooting point is an object map generating system that is the GPS coordinates of the central region of the captured image.
delete According to claim 1,
The object detection unit,
An object map generation system that detects an object using a CNNs (Convolutional Neural Networks)-based object detector such as SSD (Single Shot Multibox Detector) and YOLO (You Only Look Once).
delete According to claim 1,
The object map generation system in which the scale of the captured image calculated by the scale calculator is calculated using Equation 1 below.
[Equation 1]

here,

is the denominator of the scale,

is the focal length of the camera lens,

represents the shooting altitude.
9. The method of claim 8,
The object map matching module,
a map matching unit generating an object map by matching a plurality of mapped images provided from the image mapping unit; and
and a map output unit providing an operator with the object map generated through the map matching unit.
10. The method of claim 9,
The image registration in the map matching unit is performed using at least one of a keypoint-based method, which is a regional feature extracted from an image, and a registration method based on geometrical positions of cameras and lenses.
A method for generating an object map, comprising:
detecting an object from a ground-captured image among ground-captured images and detailed image capturing information wirelessly transmitted from a drone;
obtaining GPS coordinates of the detected object, and mapping the acquired GPS coordinates of the object to the captured image; and
generating an object map by matching the mapped captured images,
Detecting the object comprises:
receiving a ground-captured image and detailed image capturing information wirelessly transmitted from the drone; and
detecting at least one object from the received ground-captured image,
The detailed image shooting information received from the drone is,
It includes at least one of altitude information at the time of shooting, focal length information of the shooting camera, and GPS coordinate information of the shooting point, wherein the object detection result information includes:
It includes at least one or more of information about the probability, label, and coordinates of the four vertices of a rectangular box indicating the object area from the input image,
The step of mapping to the captured image,
calculating the scale of the photographed image using the altitude information of the photographing time received from the drone and the focal length information of the photographing camera;
obtaining GPS coordinates for the center point of each object using the object detection result information, the GPS coordinate information of the photographing point received from the drone, and the calculated scale value of the photographed image; and
and mapping the detection result of the detected object and the GPS coordinates of each object to the captured image.
delete delete 12. The method of claim 11,
The method of generating an object map, wherein the GPS coordinate information of the photographing point is the GPS coordinate of the central region of the photographed image.
delete 12. The method of claim 11,
Detecting the object comprises:
An object map generation method that detects an object using a Convolutional Neural Networks (CNNs)-based object detector such as SSD (Single Shot Multibox Detector) and YOLO (You Only Look Once).
delete 12. The method of claim 11,
In the step of calculating the scale, the scale of the captured image is calculated using Equation 1 below.
[Equation 1]

here,

is the denominator of the scale,

is the focal length of the camera lens,

represents the shooting altitude.
19. The method of claim 18,
The step of generating the object map comprises:
generating an object map by matching the plurality of mapped images; and
and outputting the created object map so that an operator can check it.
20. The method of claim 19,
The image registration is performed using at least one of a keypoint-based method, which is a regional feature extracted from an image, and a registration method based on a geometrical position of a camera and a lens.

Images