KR101355990B1 - Edge based method for levee mapping using lidar data and orthophotograph - Google Patents

Abstract

The present invention relates to a method for mapping edge-based levee lines using light detection and ranging (LiDAR) data and orthophotograph and, more specifically, to a method for mapping edge-based levee lines using LiDAR data and orthophotograph wherein the edge-based levee lines are extracted using the orthophotograph, and in an area where edge-based levee lines are not extracted, LiDAR-based levee lines extracted using the LiDAR data are connected to complete the levee lines. [Reference numerals] (AA) Start; (BB) End; (S10) Generate LiDAR-based levee lines; (S20) Emphasize images; (S30) Extract edges; (S40) Generate edge-based levee lines; (S50) Connect levee lines

Description

EDGE BASED METHOD FOR LEVEE MAPPING USING LIDAR DATA AND ORTHOPHOTOGRAPH}

The present invention relates to a method for mapping edge-based method lines using LiDAR data and ortho images, and more specifically, edge-based levee lines are extracted using ortho images, LiDAR-based levee lines (LiDAR-based levee lines) extracted using LiDAR (Lilight Detection and Range) data are used to complete the manufacturing line. Edge-based method line mapping method using data and orthoimage.

Water is an indispensable and indispensable resource throughout the entire process of birth, growth, reproduction, and extinction. As many countries in the world are expected to become water-scarce countries in the future, 20% of the world's population (about 1.1 billion people) can not drink clean water due to the failure of water management policies.

Especially, rainfall is biased year by year, seasonally and regionally, and seasonal river flow fluctuation is very characteristic due to the characteristics of mountainous terrain where slopes are urgent, and natural conditions are very unfavorable to water resource management. In the middle of the year, there is a problem of water supply due to shortage of river water. Unlike foreign countries, which have steady rainfall during the year, floods and droughts are frequent, and the necessity of river management will increase continuously.

Therefore, in order to properly perform the role of the river, the accurate status of the river and the management technology are increasing in importance, and the precise river surveying technology that provides the basic data of the river status is becoming important. Future river surveillance techniques will be more demanding for accuracy, speed and database, and the technology is expected to progress.

The embankment is a representative river system installed in the river basin, and is a workpiece installed along the river to maintain smooth communication and protect the flowing water. The structure of the embankment consists largely of the embankment site including the excavation, the front slope, the back slope, the dike floor, etc., and it can be used to adjust the shape if necessary. Means a line connected in the horizontal direction. Thus, the structure of the embankment is divided into several types, and it is necessary to accurately grasp the structure of the embankment so that it can be prepared in advance for disasters such as flood. However, in order to accurately measure and express the structure of the bank, there is a problem that it requires a lot of manpower, time and expenses, and it is difficult to acquire information related to the structure of the bank in real time.

Korean Unexamined Patent Publication [10-2001-0000443] discloses a building extraction system and method by aerial surveying image and radar data fusion, and a recording medium storing the program source.

Korean Published Patent [10-2001-0000443]

Accordingly, the present invention has been made to solve the above problems, an object of the present invention is to extract the edge-based levee lines using a lidar image and an ortho image, the edge In areas where the shape of the method is not extracted, the horizontal accuracy is high and connected by connecting LiDAR-based levee lines extracted using LiDAR data. It is to provide an edge-based method line mapping method using lidar data and orthoimage to map the method line consisting of.

Edge-based method line mapping method using a lidar data and an orthogonal image according to an embodiment of the present invention for achieving the above object is implemented in the form of a program executed by arithmetic processing means including a computer Generating a LiDAR-based levee lines by receiving light detection and ranging (LiDAR) point group data by a levee mapping program (S10); An image enhancement step (S20) of applying a filtering technique to the multispectral orthogonal image based on the lidar-based manufacturing method line to darken the brightness of the pixel as it moves away from the lidar-based manufacturing method line (S20); An edge detection step (S30) of detecting an edge from the image modified in the image emphasis step (S20) using an edge detection technique; Edge-based agent for generating edge-based levee lines based on the distance from the edges detected in the edge detection step (S30) to the lidar-based method line and the length of the detected edges Method line generation step (S40); And a method line connecting step (S50) of connecting the area where the edge-based method line is not generated based on the edge-based method line to the lidar-based method line.

In addition, the lidar-based method line generation step (S10) is an initial method line extraction step for extracting the initial method line located on the head and the bottom of the front and rear slopes using a vector-based method (Vector-based method) (S11); And a smoothing step (S12) of smoothing the initial manufacturing method line.

In addition, the smoothing of the smoothing step (S12) is characterized by using a polynomial equation technique.

In addition, the multispectral orthogonal image of the image emphasis step S20 may be a red-band image.

In addition, the filtering method of the image emphasis step S20 may be Gaussian filtering.

In addition, the Gaussian filtering is

Where X is the pixel, D is the nearest distance from the LiDAR-based method line, I (X) is the pixel's original brightness value, and G (X) is the Brightness value, sigma means standard deviation.), And the brightness value of each pixel is corrected.

In addition, the edge detection technique of the edge detection step (S30) is characterized in that the Canny (Canny) edge detection technique.

In addition, the edge-based method line generation step (S40) may include a noise deletion step (S41) of deleting edges corresponding to a predetermined range of lengths; And regions where the edges overlap among the edges remaining after the noise erasing step S41 are deleted, leaving edges closest to the lidar-based manufacturing method line to the edge-based levee lines. Edge-based method line determining step (S42) to determine the; characterized in that it comprises a.

In addition, the predetermined range of the noise canceling step (S41) is characterized in that more than 0m and less than 3m.

In addition, the edge-based method line determination step (S42) checks the edges remaining after the noise erasing step (S41) at intervals of 1m so that the average distance from the lidar-based method line is greater than 0m and less than 0.6m, and standard Selecting the edges with a deviation greater than 0m and less than 0.3m, and deletes all but the manufacturing method edges.

According to the edge-based method line mapping method using LiDAR data and orthoimage according to an embodiment of the present invention, the edge-based method line using LiDAR point group data and the edge-based method line using multispectral orthoimage Compared with the methods of extracting the method lines using only LiDAR data, the horizontal accuracy is relatively increased, and the method lines can be mapped to the connected lines.

In addition, more accurate edge-based method lines can be extracted using a vector-based method, a parinomial equation method, Gaussian filtering, and Canny edge detection method based on the red band image.

In addition, it is possible to remove noise by using the length of the edge.

In addition, there is an effect that can extract a more accurate edge-based method line using the average distance and standard deviation between the edge and the lidar-based method line.

In addition, it is possible to map the method line consisting of a line with a high horizontal accuracy, it is possible to evaluate the stability and efficient management of the embankment.

Figure 1 is a topographic view of a generally distinct stream.
FIG. 2 shows lines (edges) extracted from embankment area from high resolution orthoimages using Canny edge detection technique. FIG.
3 to 5 are flowcharts illustrating a method mapping method using LiDAR data and ortho images according to an embodiment of the present invention.
6 shows a lidar-based manufacturing method line.
7 is a view showing a rad band image of the multispectral orthoimage.
8 is a view showing an image generated by modifying the brightness value of each pixel using a Gaussian filtering technique on a multispectral orthoimage based on a lidar-based manufacturing method.
9 illustrates edges extracted using the Canny edge detector method from the image of FIG. 8. FIG.
10 shows edge candidates of a method line;
11 shows an edge based manufacturing method line.
FIG. 12 is a view illustrating a method line finally constructed by an edge-based method line mapping method using LiDAR data and ortho images according to an embodiment of the present invention. FIG.

Hereinafter, the present invention will be described in more detail with reference to the drawings. The following drawings are provided by way of example so that those skilled in the art can fully understand the spirit of the present invention. Therefore, the present invention is not limited to the following drawings, but may be embodied in other forms. In addition, like reference numerals designate like elements throughout the specification. It is to be noted that the same elements among the drawings are denoted by the same reference numerals whenever possible. Further, it is to be understood that, unless otherwise defined, technical terms and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Descriptions of known functions and configurations that may be unnecessarily blurred are omitted.

1 is a topographical map of a generally divided stream, FIG. 2 is a view showing lines (edges) extracted from a levee area from a high resolution orthogonal image using a Canny edge detection technique, and FIGS. 3 to 5 illustrate the present invention. FIG. 6 is a flowchart illustrating a method mapping method using LiDAR data and orthoimages according to an embodiment of the present disclosure. FIG. 6 is a view illustrating a method for mapping a LiDAR based method. FIG. 8 is a diagram illustrating an image generated by modifying a brightness value of each pixel by using a Gaussian filtering technique on a multispectral orthoimage based on a lidar-based method line, and FIG. 9 is a diagram of FIG. 8. Figure 10 shows the edges extracted using the Canny edge detector method from the image, Figure 10 shows the edge candidates of the method line, Figure 11 shows the edge A diagram showing how half the line, Figure 12 is a view showing the way the line finally built by the LIDAR data and the edge-based method using a line map orthoimage method according to one embodiment of the present invention.

The dike is an artificial structure that is installed along the stream to prevent flooding of the riverway into the mine. The dike means the line connecting the front slope head of the dike in the horizontal direction.

As shown in FIG. 1, since the inclination of the embankment is significantly changed with respect to the manufacturing method line, the previous manufacturing method study extracted the manufacturing method line from the LiDAR data using a breakline detection method ( Brgelgelmann, 2000; Briese, 2004; Brzank et al., 2008). Lidar data has the advantage of providing three-dimensional information of the terrain, but the horizontal accuracy is relatively lower than the image data, and has the disadvantage that it consists of point group data. Ortho image data has higher horizontal accuracy than LiDAR data and is composed of pixels that are connected without interruption. Thus, edge information extracted from ortho image data is more accurate than a line composed of LiDAR points. It has the advantage of being able to describe precisely.

The present invention generates edge-based levee lines using LiDAR data and high-resolution orthoimages, and LiDAR- in regions where edge-based levee lines are not extracted. based on the levee lines) to map the method lines.

The manufacturing method means a line connecting the front slope head of the bank, the head of the rear slope, the bottom of the front slope and the bottom of the rear slope in the horizontal direction.

As shown in FIG. 2, when a method line is extracted from a high resolution orthogonal image using an edge detection technique, the following problem occurs.

Problem 1. In general, the process line is formed at the point where the slope of the dike changes rapidly. However, as shown in FIG. 2, when it is desired to extract the manufacturing method line using only the image, it may not be known which edge is formed at the point where the slope of the dike is sharply changed.

Problem 2. If the pixels constituting the manufacturing method have a small brightness value, the corresponding edge cannot be extracted.

Problem 3. If the pixels of the edge constituting the manufacturing method are located at the wrong points, the edge may contain an error.

The present invention relates to an edge-based method mapping method (Edge-based method) for extracting the manufacturing method line from the fusion data of Lidar data and high-resolution orthoimage.

As shown in FIG. 3, the edge-based method line mapping method using lidar data and ortho images according to an embodiment of the present invention may include embankment mapping implemented in a program form executed by arithmetic processing means including a computer. By the program, the lidar-based method line generation step (S10), the image enhancement step (S20), the edge detection step (S30), the edge-based method line generation step (S40) and the method line connection step (S50) It is characterized by including.

In step S10 of generating a lidar-based method line, LiDAR-based levee lines are generated by receiving light detection and ranging (LiDAR) point group data.

As shown in FIG. 4, the lidar-based manufacturing method generating step S10 may include an initial manufacturing method extracting step S11 and a smoothing step S12.

The initial manufacturing method extraction step (S11) extracts the initial manufacturing method lines located on the head and the bottom of the front and rear slopes using a vector-based method. In other words, the initial method lines are extracted at the head of the front and rear slopes, the bottom of the front slope and the bottom of the rear slope using a vector-based method. Here, the vector-based method can be found in the Republic of Korea Patent Registration [10-1078238] and the like will be omitted a detailed description.

The smoothing step S12 smoothes the initial manufacturing method line. At this time, the smoothed manufacturing method line is called a lidar-based manufacturing method line.

At this time, the smoothing of the smoothing step S12 may be characterized by using a polynomial equation technique.

In other words, in order to eliminate the zigzag pattern of the extracted initial method lines, the initial method lines may be smoothed by using the Palonic equation method. Figure 6 shows the method lines smoothed using the PAEK method in ArcGIS software (ArcGIS, 2013).

The image enhancement step S20 applies a filtering technique to the multispectral orthogonal image based on the LiDAR based method line, thereby darkening the brightness of the pixel as it is farther from the LiDAR based method line. Multispectral orthoimages may be composed of red, green, blue and near infrared bands (NIR (Near-InfraRed) bands).

In this case, the multispectral orthogonal image of the image emphasis step S20 may be a red-band image.

In general, the manufacturing method line is formed along the boundary between the road and the concrete inclined block (upper manufacturing method) or the boundary between the concrete inclined block and the ground (lower manufacturing method line). Therefore, among the various bands of the multispectral image, a red band image (see FIG. 7), which is a visible band and has the longest wavelength, is most preferable for extracting a method line.

In addition, the filtering method of the image enhancement step S20 may be characterized by Gaussian filtering.

Here, the Gaussian filtering is

Da food

Where X is the pixel, D is the nearest distance from the LiDAR-based method line, I (X) is the pixel's original brightness value, and G (X) is the The brightness value, σ, is the standard deviation.)

It may be characterized in that by modifying the brightness value of each pixel.

The closest distances between each pixel of the image and the lidar-based manufacturing method lines are respectively calculated, and this is defined as D of each pixel, and it is assumed as follows.

  1) The method line edges extracted from the image are located near the lidar based method line.

  2) Pixels located near the manufacturing line have relatively high intensity values.

3) Among the edges extracted using the Canny edge detector method, the edges far from the method line are likely to be noise edges, not edges forming the method line.

The edge-based method line mapping method using the LiDAR data and the orthogonal image according to the embodiment of the present invention is based on the above three prerequisites. Increasing the brightness value of the pixel, pixels far away from the lidar-based method line can be reduced by using a Gaussian filtering technique, the image generated using the above equation is shown in FIG.

When comparing FIG. 7 and FIG. 8, pixels close to the lidar-based manufacturing method have preserved or further increased the brightness value. On the other hand, it can be seen that pixels far from the lidar based method line have their brightness values reduced or converged to zero.

The edge detection step S30 detects an edge from an image modified in the image enhancement step S20 by using an edge detection technique.

In this case, the edge detection technique of the edge detection step S30 may be a canny edge detection technique.

In other words, a Gaussian filtering is used to generate the highlighted image (see FIG. 8) based on the distance from the lidar-based method lines, and then edges from the highlighted image (see FIG. 8) using the Canny edge detector method. Extraction (see FIG. 9).

An edge-based method line generation step S40 is based on the distance from the edges detected in the edge detection step S30 to a lidar-based method line and the length of the detected edges. -based levee lines)

From the extracted edges (see FIG. 9), the edges from the extracted edges to the lidar based method line and the length of the extracted edges should be selected to be suitable for use as the method line.

Edge-based method line mapping method using a Lidar data and an orthogonal image according to an embodiment of the present invention as shown in Figure 5, the step of generating the edge-based method line (S40) is the noise erasing step (S41) And it may be characterized in that it comprises an edge-based method line determination step (S42).

The noise erasing step S41 deletes edges corresponding to a certain range of lengths. At this time, the predetermined range of the noise erase step (S41) may be characterized in that more than 0m and less than 3m.

The dike is an artificial structure constructed along the stream, so the edges of the dike must have a certain length or more. Therefore, short edges are likely noise edges. In consideration of the average point density (about 70 cm) of the lidar point groups, edges of less than 3 m in length can be deleted.

In other words, the edge length can be used to remove noise.

Edge-based method line determination step (S42) is the edge-based region of the edges remaining after the noise erasing step (S41), the edge overlaps, deleting all the edges closest to the lidar-based method line Determine edge-based levee lines.

As shown in FIG. 10, there are regions where the manufacturing method edge candidates overlap. In these areas, it is desirable to select and connect edges closer to the lidar-based method line. 11 shows the results of the near edges remaining in the region where the edges overlap. The edges of the overlapped area are removed and the remaining edges are defined as edge-based levee lines.

At this time, the edge-based method line determination step (S42) by inspecting the edges remaining after the noise erasing step (S41) at intervals of 1m, the average distance from the lidar-based method line is greater than 0m 0.6m or less, and standard The edges having a deviation greater than 0m and less than or equal to 0.3m may be selected to delete all of them, leaving the manufacturing method edges.

The method line is formed at the point where the slope of the dike changes. Thus, the edges constituting the method line are located near the lidar based method line. Among edges (Condition 1) longer than 3m, inspected at intervals of 1m to select edges (Condition 2) with average distances greater than 0m and less than 0.6m and standard deviations greater than 0m and less than 0.3m Can be determined as candidate method edges.

10 shows method line edge candidates satisfying the above two conditions.

In other words, edges with higher horizontal accuracy are obtained by using vector-based methods, Palinomial equations, Gaussian filtering and Canny edge detection techniques, and average distances between edges and lidar-based method lines based on rad band images. Base manufacturing method line can be extracted.

In the manufacturing method connecting step (S50), the area where the edge-based manufacturing method line is not generated based on the edge-based manufacturing method line is connected to the lidar based manufacturing method line.

As shown in Fig. 11, due to the erosion of the dike occurring in the river area or the change of the material constituting the dike, the boundary between the ground of the dike and the slope of the dike or the slope of the embankment of the dike and the slope of the dike is an image. It may appear unclear at. Thus, as shown in Figure 11, the area where the edge-based method lines do not appear in the bank is connected by LiDAR-based levee lines. In other words, as shown in FIG. 12, the method line is finally constructed by connecting the edge-based method lines with the lidar-based method lines.

According to the edge-based method line mapping method using LiDAR data and orthoimage according to an embodiment of the present invention, the edge-based method line using LiDAR point group data and the edge-based method line using multispectral orthoimage Compared with the method of extracting the manufacturing method line using only LiDAR data, the horizontal accuracy can be increased relatively and the manufacturing method line can be mapped with the connected line. In addition, the method line finally constructed by the edge-based method line mapping method using the lidar data and the orthogonal image according to an embodiment of the present invention is the edge of 50 ~ 70% or more lines extracted from the image information Since it is possible to map the manufacturing method line composed of high horizontal accuracy and connected lines, it is possible to perform stability evaluation and efficient management of the dike.

It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

S10: step of generating lidar based manufacturing method
S11: Initial Method Line Extraction Step
S12: Smoothing Step
S20: Video highlight step
S30: edge detection step
S40: Generation of edge-based method lines
S41: Noise Erase Step
S42: edge based manufacturing method determination step
S50: manufacturing method connection step

Claims (10)

By the embankment mapping program embodied in the form of a program executed by an arithmetic processing means including a computer,
A lidar-based method line generation step (S10) of receiving light detection and ranging (LiDAR) point group data to generate LiDAR-based levee lines;
An image enhancement step (S20) of applying a filtering technique to the multispectral orthogonal image based on the lidar-based manufacturing method line to darken the brightness of the pixel as it moves away from the lidar-based manufacturing method line (S20);
An edge detection step (S30) of detecting an edge from the image modified in the image emphasis step (S20) using an edge detection technique;
Edge-based agent for generating edge-based levee lines based on the distance from the edges detected in the edge detection step (S30) to the lidar-based method line and the length of the detected edges Method line generation step (S40); And
A method line connecting step (S50) of connecting the area where the edge-based method line is not generated based on the edge-based method line to the lidar-based method line;
Edge-based method line mapping method using a lidar data and ortho images comprising a.
The method of claim 1,
Lida-based method line generation step (S10)
An initial method line extraction step (S11) of extracting an initial method line located at the head and the bottom of the front and rear slopes using a vector-based method; And
A smoothing step of smoothing the initial manufacturing method line (S12);
Edge-based method line mapping method using a lidar data and ortho images comprising a.
3. The method of claim 2,
Smoothing of the smoothing step (S12) is
Edge-based method line mapping method using LiDAR data and orthoimages using the polynomial equation technique.
The method of claim 1,
The multispectral orthogonal image of the image emphasis step S20 is
Edge-based method line mapping method using a lidar data and ortho images characterized in that the image is a red-band (Red-band).
The method of claim 1,
The filtering technique of the image emphasis step S20 is
Edge-based method line mapping method using Lidar data and ortho images characterized by Gaussian filtering.
delete The method of claim 1,
The edge detection technique of the edge detection step (S30)
Edge-based method line mapping method using Lidar data and ortho images characterized by Canny edge detection technique.
The method of claim 1,
The edge-based manufacturing method generation step (S40)
A noise erasing step (S41) of deleting edges corresponding to a predetermined range of lengths; And
Areas where the edges overlap among the edges remaining after the noise erasing step S41 are deleted, leaving edges closest to the LiDAR based method line to remove edge-based levee lines. Determining an edge-based manufacturing method line (S42);
Edge-based method line mapping method using a lidar data and ortho images comprising a.
9. The method of claim 8,
The constant range of the noise erasing step (S41) is
Edge-based method line mapping method using lidar data and ortho images, characterized in that more than 0m less than 3m.
9. The method of claim 8,
Edge-based method line determination step (S42) is
The edges remaining after the noise erasing step S41 are inspected at intervals of 1 m to select edges having an average distance of more than 0 m and less than 0.6 m, and a standard deviation of more than 0 m and less than 0.3 m from the lidar-based method line. Edge-based method line mapping method using lidar data and ortho images characterized by deleting all of the line edges.

Images