KR101080985B1 - Method for checking height of tree and population using lidar - Google Patents

Abstract

PURPOSE: A method for checking the height of tree and population using LiDAR is provided to analyze the types of tree and height from the tree height data. CONSTITUTION: A point data is obtained through an aerial LiDAR. The point data is classified into a ground data, a building data, and a vegetation data. A ground model is formed. A true tree height data is calculated using the tree population data and the ground data. The types of tree and the height of tree are analyzed from the true tree height data.

Description

Method for Checking Height of Tree and Population Using LiDAR}

The present invention relates to a method for identifying the effort and the population by using the aviation lidar (LiDAR) results, and more particularly, to a method for identifying the effort and the population using the aviation LiDAR (LiDAR) for mixed forests. It is about.

Recently, with increasing interest in Geographic Information System (GIS) and active research and development in related fields, the application and utilization of GIS is rapidly expanding in many ways. GIS is a computer-based system for using and managing geographic data to solve space related problems.

Here, the most basic data in constructing the GIS is a digital map. Numerical maps, unlike classical paper maps, store and index various terrain data obtained by surveying into numerical files. The production of digital maps is generally based on topographical data obtained from aerial photographs and satellite images, and it is necessary to interpret and quantify these data. On the other hand, recently, techniques for acquiring digital terrain data using a global positioning system (GPS) have been actively studied.

GPS is a state-of-the-art navigation system using satellites, which is used to determine the exact position of measurement on the ground. Specially designed GPS receivers can measure the absolute position of a stationary or moving object by receiving information from satellites anywhere in the world without time constraints.

On the other hand, in order to manage trees, a manual method of measuring by a person directly, and a method of generating forest information on trees using remote sensing techniques such as aerial photographs or satellite images are widely used. However, the direct measurement method takes enormous time and cost, and the method using aerial photographs or satellite photographs only photographs the surface of the forest, thereby degrading the utility of information. The captured aerial and satellite images are provided as two-dimensional data, making it difficult to carry out three-dimensional analysis such as canopy analysis of trees and digital surface model (DSM) of terrain. In addition, there is a problem in that the aerial photograph shows a difference in spatial resolution according to the altitude taken, and the satellite image shows a difference in the spatial resolution of the image according to the type of the satellite photographed.

In order to solve this problem, Korean Patent No. 0948099 discloses a system for calculating the area of vegetation using aerial laser surveying and a method of calculating the area of vegetation using the same, and Korean Patent No. 0935857 describes a forest object. Analysis of the distribution in the vertical direction from the aeronautical lidar data and in the horizontal direction from the digital aerial photograph data to generate three-dimensional forest geographic information using the aeronautical lidar and digital aerial photographs. A system and method have been disclosed, and Korean Patent No. 084100 discloses a tree canopy extraction system and method using aerial laser surveying, which can easily obtain tree canopy data using aerial laser surveying.

However, although these patents are useful data for the area and vegetation distribution of forest areas, they lack information on individual trees, which are the main forest resources, and canopy information on individual trees can be easily obtained. In the case where trees are distributed at close intervals, trees are not detected.

Accordingly, the present inventors have made diligent efforts to solve the above problems, and as a result, obtains information about the shape and the effort from the independent tree objects at a certain distance from the surrounding trees, and forms a tree model based on this. Therefore, when the tree detection for the forest area is carried out, it is confirmed that the tree and the population can be accurately detected in the mixed forest where various species are mixed, thereby completing the present invention.

An object of the present invention is to provide a method for accurately identifying the effort and the population by using the aerial LiDAR performance in a mixed forest mixed with various species.

In order to achieve the above object, the present invention (a) the point acquired by the air LiDAR (LiDAR) by using the distribution characteristics and echo information of the point data acquired by the air LiDAR (LiDAR) (point) classifying the data into ground data, building data, and vegetation data; (b) forming a ground model, which is reference data for estimating effort using the classified ground data; (c) forming pure tree height data using the vegetation data and the ground model; (d) detecting the exclusion height and the highest height by analyzing the tree shape and the tree height from the formed pure tree height data; (e) performing sampling according to the shape of the water pipe on the individual trees independent of the surrounding tree objects in the pure tree height data; (f) detecting a variation range for the effort and the pipe width for the sampled individual trees, and establishing a tree model; And (g) matching the established tree model with the vegetation data to identify the tree's effort and population.

Effort and population identification method using the aviation LiDAR (LiDAR) performance according to the present invention is useful in the management of the individual trees, since it is possible to accurately identify the effort and population in the mixed forest mixed with various species. In addition, the analysis method for vertical structure of water pipes obtained from a series of processes for tree effort and population detection can measure the change of water pipes in the forest when the same analysis method is applied to time series data of forest resources. It is useful for measuring tree growth, which is an important parameter for estimating CO2 absorption by.

1 is a flow chart of the effort and individual identification method using the aviation LiDAR results according to the present invention.
FIG. 2 is a photograph (B) of classifying point data acquired by aerial photograph A of a test subject area and an air lidar LiDAR for the same test subject area according to an embodiment of the present invention. (Orange color: ground data, green: vegetation data, red: building data).
FIG. 3 illustrates surface data A and point data of an airborne lidar classified according to an embodiment of the present invention as surface data in the form of a triangulated irregular network (TIN). It is the result of the formed ground model (B).
4 is a conceptual diagram (A) and result photo (B) of pure tree height data according to an embodiment of the present invention.
5 is an example (A) and the analysis result (B) of the tree height according to an embodiment of the crown class of the present invention.
FIG. 6 is an explanatory diagram for forming tree modeling data by observing the height of an individual tree and the width of a pipe according to an embodiment of the present invention.
7 is a tree model result for independent trees established according to an embodiment of the present invention.
8 is a detection result of an individual tree according to an embodiment of the present invention.

According to the present invention, when a tree model for an independent tree individual is formed and applied to individual tree detection, it is confirmed that the tree and the number of trees can be accurately detected in a mixed forest where various species are mixed.

Accordingly, the present invention uses (a) the point data acquired by the aviation lidar (LiDAR) by using the distribution characteristics and echo information of the point data acquired by the aviation lidar (LiDAR). Classifying into ground data, building data, and vegetation data; (b) forming a ground model, which is reference data for estimating effort using the classified ground data; (c) forming pure tree height data using the vegetation data and the ground model; (d) detecting the exclusion height and the highest height by analyzing the tree shape and the tree height from the formed pure tree height data; (e) performing sampling according to the shape of the water pipe on the individual trees independent of the surrounding tree objects in the pure tree height data; (f) detecting a variation range for the effort and the pipe width for the sampled individual trees, and establishing a tree model; And (g) matching the established tree model with the vegetation data to identify the tree's effort and population. The method relates to the effort and population identification using LiDAR performance.

Hereinafter, with reference to the drawings to describe the labor and individual identification method using the air LiDAR (LiDAR) performance according to the present invention in detail.

1 is a flow chart of the effort and population identification method using the air LiDAR (LiDAR) performance according to the present invention. As shown in FIG. 1, first, when data is input, the data is acquired using the air LiDAR by using distribution characteristics and echo information of point data acquired by the air LiDAR. Point data is classified into ground data, building data, and vegetation data. After the point groups of the acquired laser points are configured with a constant lattice spacing, the lowest point is extracted as an initial ground point, and a Triangulated Irregular Network (TIN) data is created using the initial ground point as a vertex. The final ground data is classified by repeating the process of searching the ground data according to the iteration angle and the iteration distance with respect to the surrounding points based on the generated irregular triangle network data. The building data is extracted and classified using the Z tolerance and the Max. Building Size based on the irregular triangle network data of the created ground data. Vegetation data representing the tree is classified into vegetation according to the irregular time of the reflected pulse by analyzing echo information of the acquired point data.

FIG. 2 is a photograph (B) of classifying point data acquired by aerial photograph A of a test subject area and an air lidar LiDAR for the same test subject area according to an embodiment of the present invention. . In FIG. 2B, the ground data is orange, the vegetation data is green, and the building data is shown in red.

After the classification of the point data acquired by the air lidar (LiDAR) is completed, a ground model is formed. FIG. 3 illustrates surface data A and point data of an airborne lidar classified according to an embodiment of the present invention as surface data in the form of a triangulated irregular network (TIN). It is the result of the formed ground model (B).

As shown in FIG. 3, the ground model is reference data for calculating effort, and may be formed using the classified ground data.

The ground model is formed of Triangulated Irregular Network (TIN) data capable of expressing elevation differences due to complex terrain changes in consideration of characteristics of irregular ground data.

When the ground model is completed, pure tree height data is formed by using it.

4 is a conceptual explanatory diagram (A) and a result photograph (B) of tree height data according to an embodiment of the present invention. As shown in FIG. 4, the height performance of the point data acquired with the air LiDAR has a height from the mean sea level. That is, the height of the classified vegetation data is not the height of the tree from the ground but the height on the average sea level including the ground height. Thus, the pure tree height data is formed by subtracting the height of the ground model from the height values of the classified vegetation data.

When pure tree height data is formed, the tree height and the tree height are analyzed to detect the exclusion height and the highest height.

5 is an example (A) and the analysis result (B) of the tree height according to an embodiment of the crown class of the present invention.

The present invention focuses on the crown class of the National Forest Research Guidelines as a method for efficient population estimation in a mixed forest where several species of trees are randomly distributed. Analyze and apply the exclusion height, which is set based on the height of the crown and the height of the tree under the peak, as a criterion for forming the tree model. In other words, vegetation data having an extremely low height and vegetation data forming a lower layer were considered as a range of shrubs, vines, and trees under the analysis results, and were excluded from the extraction.

In the present invention, the crown class (Crown Class) refers to a criterion for judging the superiority of individual tree morphologies by dividing the crown development state of the tree, the degree of the crown occupying space, the position of the crown, the crown As the criteria for classifying timbers determined by blood pressure, damage level, shape of trunk, etc., it is also called tree trunk or tree type. It is used as a standard.

In the present invention, "trouble" means the height of the tree, "water pipe" means the upper part of the stem of the tree and the leaf a lot.

In the case of independent trees, it is possible to model toils and crowns, but in forest areas where several species of trees are mixed, there are many height variations, so one crown is an independent tree or a few small trees. Judgment as to whether or not it is known is difficult.

Therefore, in the present invention, sampling is performed according to the shape of the pipe on individual trees that are separated from the surrounding tree objects in the pure tree height data, and the variation range for the effort and the width of the pipe for each sampled tree is detected. A tree model was established.

6 is an explanatory view of forming a tree model for individual trees by observing the hardwood and pipe widths for individual trees according to an embodiment of the present invention.

In Figure 6 (A) is a process of observing the effort and the width of the pipe for the individual trees, for the tree model according to the various types of pipes to observe the location of the widest pipe and the location where the inflection occurs in various forms (B) is a tree modeling formed by applying the height of the pipe obtained by analyzing the vertical structure of the pipe using the tree height data formed above and the observed tree model of the individual tree. Data picture.

The tree model includes (a) tree name data for species identification, (b) exclusion height and maximum height, which are individual detection conditions according to the effort, and (c) vertex data of the water pipe model, which is detection conditions according to the water pipe type, and It can be established based on the fluctuation range of the pipe width.

7 is a tree model result for independent trees established according to an embodiment of the present invention. As shown in Fig. 7, coniferous trees such as pine and fir have a lot of spire-like water pipes, such as arrowheads, hardwoods have a variety of forms, such as a wide triangle or inverted triangle, round or cone-shaped water pipes In addition, various types of tree models can be established by forming a model according to the shape of the water pipe.

When a tree model for various types of individual trees is established, the tree model and the number of individuals are identified by matching and comparing the established tree model with the vegetation data.

The process of identifying the effort and the number of individuals is a series of search algorithms, which detects the point data of the vegetation data with the following two conditions between the ground data and the vegetation data and the tree model.

The first condition for the effort and the number of individuals detected is the effort condition. The height result obtained by calculating the height difference between the points of the ground data and the vegetation data is compared with the exclusion height and the maximum height of the tree model, Vegetation data is regarded as vegetation of shrubs, vines and subcutaneous trees, and is not applied to tree objects. Only vegetation data existing within the range above the exclusion height and below the maximum height are considered as trees. The second condition is the shape condition. In addition, the modeling result of the point group of the vegetation data and the pipe shape of the tree model is a condition to be matched within the fluctuation range of the pipe width.

8 is a detection result of an individual tree according to an embodiment of the present invention.

As shown in Figure 8, by using the aviation lidar (LiDAR) performance according to the present invention using the effort and population identification method it is possible to accurately identify the effort and population even in mixed forests mixed with various species.

While the present invention has been particularly shown and described with reference to specific embodiments thereof, those skilled in the art will readily appreciate that many modifications are possible in the exemplary embodiments without materially departing from the novel teachings and advantages of this invention. something to do. Thus, the substantial scope of the present invention will be defined by the appended claims and their equivalents.

Claims (5)

(a) The ground data and the building data of the point data acquired by the aviation lidar (LiDAR) by using the distribution characteristics and echo information of the point data acquired by the aviation lidar (LiDAR) And classifying the vegetation data;
(b) forming a ground model, which is reference data for estimating effort using the classified ground data;
(c) forming pure tree height data using the vegetation data and the ground model;
(d) detecting the exclusion height and the highest height by analyzing the tree shape and the tree height from the formed pure tree height data;
(e) performing sampling according to the shape of the water pipe on the individual trees independent of the surrounding tree objects in the pure tree height data;
(f) detecting a variation range for the effort and the pipe width for the sampled individual trees, and establishing a tree model; And
(g) a method of identifying effort and population using aerial LiDAR results comprising the step of matching and comparing the established tree model with the vegetation data.
The ground model of claim 1, wherein the ground model is formed of Triangulated Irregular Network (TIN) data capable of expressing an altitude difference according to a complicated terrain change in consideration of characteristics of irregularly existing ground data. How to identify the effort and population using aviation LiDAR performance.
According to claim 1, The pure tree height data is the height of the classified vegetation data considering that the height of the classified vegetation data is the height on the average sea level including the ground height, not the height of the actual tree of the ground model Effort and population identification method using the aviation LiDAR (LiDAR) performance, characterized in that formed by excluding the height.
According to claim 1, The height of the exclusion and the number of individuals using the aerial LiDAR (LiDAR) performance, characterized in that the set based on the height of the tree under the vertical structure analysis through the vertical structural analysis of the water pipe of the pure tree height data checking way.
The tree model according to claim 1, wherein the tree model includes (a) tree name data for identifying species, (b) exclusion height and maximum height, which are individual detection conditions according to effort, and (c) detection conditions according to water pipe type. A method for identifying labor and population using LiDAR performance, which is established based on vertex data and fluctuation ranges of pipe widths.

Images