WO2022249507A1

WO2022249507A1 - Individual tree modeling system and individual tree modeling method

Info

Publication number: WO2022249507A1
Application number: PCT/JP2021/037558
Authority: WO
Inventors: 圭司山口
Original assignee: 株式会社マプリィ
Priority date: 2021-05-24
Filing date: 2021-10-11
Publication date: 2022-12-01
Also published as: JP2022179840A; JP6931501B1

Abstract

An acquisition control unit 101 acquires three-dimensional point group data by irradiating an individual tree in a forest from a predetermined measurement point with a laser beam of a three-dimensional laser scanner. A generation control unit 102 generates individual tree mesh data representing the bark of the individual tree by converting the acquired three-dimensional point group data into mesh data composed of polygons. An estimation control unit 103 regards a plurality of points, located on the same plane in a horizontal direction with respect to the generated individual tree mesh data, as a portion of a circle formed by the bark of the individual tree, and extrapolates the circle from the plurality of points, and uses the radius of the extrapolated circle to estimate, as virtual data on the bark of the individual tree, data on the bark of the individual tree in which the individual tree mesh data is not present. A formation control unit 104 forms a three-dimensional cylinder model as a three-dimensional model of a stem of the individual tree on the basis of the individual tree mesh data, the estimated virtual data, the radius of the circle of the virtual data, and information on the height of the individual tree.

Description

Single tree modeling system and single tree modeling method

The present invention relates to a single tree modeling system and a single tree modeling method.

Forests with dense single trees (trees) have multifaceted functions such as water and soil conservation, global environment conservation, and biological protection. It is In the most primitive method, for example, a measurer enters a forest and obtains information on one single tree among a plurality of single trees, for example, tree information such as tree position, breast height diameter, and tree height. There is a method of measuring each. However, such a method has the problem that it takes a lot of work time, and accurate tree information cannot be obtained depending on the skill of the measurer.

Therefore, in recent years, methods for acquiring tree information using scanners have been developed. For example, Japanese Patent Laying-Open No. 2010-96752 (Patent Document 1) describes a tree information measurement method having distance data acquisition means, feature data extraction means, matching means, single tree extraction means, and tree information detection means. An apparatus and method for measuring tree information are disclosed. The distance data acquisition means measures distance data to an arbitrary portion of the object to be measured at a plurality of points, and the feature data extraction means extracts a set of feature data corresponding to the tree trunk from the distance data. The matching means matches the distance data of a plurality of points by scan matching and specifies a three-dimensional coordinate system, and the single tree extracting means extracts a single tree from the coordinate point data specified in the three-dimensional coordinate system. The tree information detection means detects tree information including one or more of tree height, trunk diameter, crown length, and crown diameter for each single tree. As a result, highly accurate tree information can be acquired with little effort.

In addition, in Japanese Patent Application Laid-Open No. 2012-98247 (Patent Document 2), a laser pulse is swept from the sky, and the three-dimensional point cloud data of the forest obtained by airborne laser measurement that measures the reflected signal waveform is used. A tree position detection device for detecting the positions of trees is disclosed. This tree position detection device includes normalization means, branch lower layer setting means, plane projection means, and position detection means. The normalization means converts the height represented by the point cloud data to the actual height from the ground surface to generate normalized point cloud data, and the branch lower layer setting means converts the tree crown area and the lower branch area of the forest. Based on the difference in the distribution of normalized point cloud data between and, a constant height range in the forest is set as the branch lower layer. The plane projection means extracts the normalized point cloud data belonging to the sublayer and projects it onto a plane along the ground surface to obtain a two-dimensional frequency distribution, and the position detection means calculates the two-dimensional frequency distribution based on a predetermined standard. , the location where the normalized point cloud data gather is detected and set as the tree location. As a result, it is possible to detect the position of trees with improved accuracy using data from aerial laser measurements without relying on field surveys.

Japanese Patent Application Laid-Open No. 2017-167092 (Patent Document 3) discloses a feature detection device that detects a target feature rising from the ground based on three-dimensional coordinate data of a point group extracted from the surface of the feature in a target space. is disclosed. Japanese Patent No. 6828211 (Patent Document 4) discloses a forest resource analysis device that evaluates forest resources based on point cloud data including coordinate values in the vertical, horizontal, and height directions of a forest. Chinese Patent Application Publication No. 110274549 (Patent Document 5) discloses a method and apparatus for measuring harvested and cultured targets. Chinese Patent Application Publication No. 109270544 (Patent Document 6) discloses a self-positioning system for mobile robots based on rod-like object recognition. Chinese Patent Application Publication No. 102834691 (Patent Document 7) discloses a mapping method for mapping an object and an object representative point corresponding to the type of object from a surveying instrument. Chinese Patent Application Publication No. 106408608 discloses a method for extracting stem diameter from terrestrial laser radar point cloud data.

JP 2010-96752 A JP 2012-98247 A JP 2017-167092 A Japanese Patent No. 6828211 Chinese Patent Application Publication No. 110274549 Chinese Patent Application Publication No. 109270544 Chinese Patent Application Publication No. 102834691 Chinese Patent Application Publication No. 106408608

However, the technology described in Patent Document 1 has the problem that it requires distance data of multiple points in order to match the distance data of multiple points, and it takes time and effort to acquire the distance data. In addition, there is a problem that the distance data of a plurality of points has an extremely large amount of information and takes time to process. Furthermore, the technique described in Patent Literature 2 has a problem that normalization processing is required and data of airborne laser measurement is required. Although the techniques described in Patent Documents 3 to 8 use point cloud data, there is a problem that processing takes time.

On the other hand, it is convenient to have a 3D model of the single trees that make up the forest for calculating the timber volume (volume) of standing trees and the accumulation of forests. It is necessary to acquire 3D point cloud data from two directions of the front part and the back part, and there is a problem that it takes time and effort to acquire the 3D point cloud data. The techniques described in Patent Documents 1 to 8 above cannot solve such problems.

SUMMARY OF THE INVENTION Accordingly, the present invention has been made to solve the above-mentioned problems. It is an object of the present invention to provide a modeling system and a single tree modeling method.

A single tree modeling system according to the present invention includes an acquisition control unit, a generation control unit, an estimation control unit, and a formation control unit. The acquisition control unit acquires three-dimensional point cloud data of a single tree by irradiating a single tree in the forest from a predetermined measurement point with a laser of a three-dimensional laser scanner. The generation control unit converts the acquired three-dimensional point cloud data into mesh data composed of polygons, thereby generating single tree mesh data representing the bark of the single tree. The estimation control unit regards a plurality of points located on the same plane in the horizontal direction with respect to the generated single tree mesh data as part of a circle formed by the bark of the single tree, and calculates the points at the plurality of points. A circle is extrapolated, and the extrapolated circle radius is used to estimate bark data of a single tree for which the single tree mesh data does not exist as virtual data of the bark of the single tree. The formation control unit converts the three-dimensional cylindrical model into the single tree based on the single tree mesh data, the estimated virtual data, the circle radius of the virtual data, and the tree height information of the single tree. form a three-dimensional model of the trunk of

A single tree modeling method according to the present invention comprises an acquisition control process, a generation control process, an estimation control process, and a formation control process. Each step of the single tree modeling method corresponds to each controller of the single tree modeling system.

According to the present invention, it is possible to easily and quickly form a 3D model of a single tree by scanning a portion of the single tree with a 3D laser scanner.

1 is a schematic diagram illustrating an example of a single tree modeling system according to an embodiment of the present invention; FIG. 4 is a flow chart showing the execution procedure of the single tree modeling method according to the embodiment of the present invention; FIG. 3A shows an example of scanning by a measurer with a three-dimensional laser scanner, and FIG. 3B shows an example of three-dimensional point cloud data scanned by a measurer with a three-dimensional laser scanner. . It is a figure (FIG. 4A) which shows an example of three-dimensional point-group data, and a figure (FIG. 4B) which shows an example of three-dimensional point-group data actually obtained. FIG. 5A shows an example of mesh data, and FIG. 5B shows an example of actually obtained mesh data. FIG. 6A shows an example of extrapolating a circle to a plurality of points of single tree mesh data, and a diagram showing an example of stepwise extrapolation of a circle to single tree mesh data in the vertical direction. (FIG. 6B). FIG. 10 is a diagram showing an example of combining single tree mesh data and virtual data; FIG. 10 is a diagram showing an example of forming a three-dimensional model of a trunk of a single tree; FIG. 9A is a diagram showing an example of mesh data of a plurality of single trees, and FIG. 9B is a diagram showing an example of a three-dimensional model of trunks of a plurality of single trees. FIG. 10 is a diagram showing an example of a case in which an observer scans while moving around in a forest; FIG. 11A is a diagram showing an example of a three-dimensional data platform for forests, and FIG. 11B is a diagram showing an example of a tree-cutting simulation. A diagram (FIG. 12A) showing an example when tree crown data is combined with a three-dimensional model of the trunk of a single tree, and a diagram (FIG. 12B) showing an example of a database corresponding to the three-dimensional model of the trunk of a single tree. be. FIG. 4 is a schematic diagram showing an example of correction of position information by closing processing;

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings for understanding of the present invention. It should be noted that the following embodiment is an example that embodies the present invention, and is not intended to limit the technical scope of the present invention.

A single tree modeling system 1 according to the present invention is basically composed of a three-dimensional laser scanner 10 and a terminal device 11, as shown in FIG. The 3D laser scanner 10 can acquire 3D point cloud data of a single tree in the forest.

Here, although there is no particular problem with the configuration of the three-dimensional laser scanner 10, for example, the three-dimensional laser scanner 10 includes a laser irradiation section, a scattered light detection section, and a point cloud calculation section. The laser irradiation unit emits a pulsed laser to the object. The scattered light detection unit detects scattered light from the laser irradiated to the object. Based on the detected scattered light, the point cloud calculator calculates three-dimensional data of the point cloud of the distance from the three-dimensional laser scanner 10 to the object and the position where the scattered light is scattered on the surface of the object. The type of the three-dimensional laser scanner 10 is not particularly limited, but examples thereof include Lidar (Light Detection and Ranging, Laser Imaging Detection and Ranging) sensors.

The terminal device 11 also includes a display unit, a reception unit (input unit), a storage unit, a control unit, an output unit, and a communication unit. The display unit displays a screen. The reception unit receives input of a predetermined instruction by a user's operation. The storage unit stores data. The control section controls each section. The output unit outputs data. The communication unit acquires position information of the terminal device 11 (for example, GPS position information, position information obtained by a dual-frequency multi-GNSS receiver).

Examples of the terminal device 11 include a desktop terminal device, a tablet terminal device, a portable notebook computer, a mobile terminal device (smartphone) with a touch panel, and the like. Note that the terminal device 11 is not particularly limited as long as it is a device capable of analyzing the three-dimensional point group data of a single tree acquired by a three-dimensional laser scanner.

Here, since the three-dimensional laser scanner 10 may be built in the terminal device 11, in that case, the terminal device 11 may also serve as the three-dimensional laser scanner 10.

The terminal device 11 incorporates a CPU, ROM, RAM, HDD, SSD, etc. (not shown). to run. Further, each control unit, which will be described later, is implemented by the CPU executing a program.

Next, the configuration and execution procedure according to the embodiment of the present invention will be described with reference to FIGS. 1 and 2. FIG. First, a measurer who collects tree information of single trees in the forest visits the forest carrying the three-dimensional laser scanner 10 and the terminal device 11 of the single tree modeling system 1, and 11 is activated. Then, the measurer goes to a predetermined measurement point near the single tree to be measured, points the three-dimensional laser scanner 10 at the single tree, and inputs a measurement key into the terminal device 11 . Then, the acquisition control unit 101 of the terminal device 11 acquires the three-dimensional point cloud data of the single tree by irradiating the laser of the three-dimensional laser scanner 10 ( FIG. 2 : S101).

Here, the acquisition method of the acquisition control unit 101 is not particularly limited. For example, as shown in FIG. 3A, the measurer directs the laser irradiation unit of the three-dimensional laser scanner 10 toward a single tree and inputs a measurement key into the terminal device 11 . Then, the acquisition control unit 101 receives the input of the measurement key and starts laser irradiation of the three-dimensional laser scanner 10 . The laser irradiated to the single tree in the forest is reflected by the front portion of the single tree facing the three-dimensional laser scanner 10 (for example, the front portion of the surface of the bark) and scattered as scattered light. By detecting the scattered light, the three-dimensional laser scanner 10 acquires three-dimensional point cloud data of the positions where the scattered light is scattered.

Here, since the three-dimensional laser scanner 10 also detects scattered light scattered from other than the front portion of the single tree, the three-dimensional point cloud data includes data indicating the front portion of the single tree as shown in FIG. 3B. Other data may also be included that indicate the ground near the base of a single tree.

Now, the acquisition control unit 101 may acquire position information at a measurement point when acquiring three-dimensional point cloud data. For example, when acquiring the three-dimensional point cloud data, the acquisition control unit 101 acquires the position information of the terminal device 11 using the communication unit of the terminal device 11, and converts the position information of the terminal device 11 into the three-dimensional point cloud data. be stored in association with This makes it possible to associate the three-dimensional point cloud data with the positional information of the measurement points. This position information may be used, for example, when obtaining tree height information of a single tree (described later). Further, this position information may be used when obtaining latitude, longitude, altitude, etc. indicating the installation position of the three-dimensional model of the trunk of a single tree formed later.

Now, when the acquisition by the acquisition control unit 101 is completed, next, the generation control unit 102 of the terminal device 11 converts the acquired three-dimensional point cloud data into mesh data made up of polygons, so that simple A single tree mesh data representing the bark of a tree is generated (Fig. 2: S102).

Here, the generation method of the generation control unit 102 is not particularly limited. For example, as shown in FIG. 4A, the obtained three-dimensional point cloud data expresses a plurality of single trees in the forest and the ground at the base of the single trees. FIG. 4B shows three-dimensional point cloud data obtained by actually scanning a single tree in the forest with the three-dimensional laser scanner 10 . As described above, the three-dimensional point cloud data is a collection of points, and the shape of a single tree cannot be recognized.

Therefore, the generation control unit 102 converts the three-dimensional point cloud data into mesh data in which all faces are composed of triangles with three vertices, for example, as shown in FIG. 5A. Although there is no particular limitation on the meshing algorithm, for example, a triangulation method can be used. The mesh data is data expressing three-dimensional point cloud data in a geographical coordinate system as a mesh (network), and is composed of items such as a mesh ID, a mesh vertex, and a normal vector. The mesh vertex stores a predetermined number (1 or more) of vertices of the three-dimensional position information in the geographic coordinate system. The normal vector is a vector perpendicular to the mesh surface, meaning a vector perpendicular to all straight lines on the mesh surface, and stores a three-dimensional representation of the normal vector of the mesh surface. Here, since the mesh data consists of the vertices of the polygons forming the mesh and the outward normal vectors of the polygons, the amount of data is compressed compared to individual geographical coordinate system three-dimensional point cloud data. By meshing the three-dimensional point cloud data in this way, it is possible to convert a large amount of point cloud data into simple mesh data, which can be easily handled.

In the above description, triangles were used as the mesh shape, but there is no particular limitation on the type of mesh shape, and other than triangles, for example, quadrilaterals, pentagons, etc. can be mentioned.

Here, before meshing the 3D point cloud data, the generation control unit 102 performs noise processing to remove variations in points, thereby removing points unrelated to single trees or the ground from the 3D point cloud data. , and the shape of a single tree or the shape of the ground may be converted into highly accurate mesh data. As noise processing, for example, a processing of calculating a boxplot for three-dimensional point cloud data and removing points with large variations can be mentioned.

The mesh data includes single tree mesh data that indicates the bark of the front part of a single tree and ground mesh data that indicates the shape of the ground. Therefore, the generation control unit 102 selects, among the mesh data, mesh data having a shape similar to a cylindrical shape corresponding to the bark of the front part of a single tree, or having a height equal to or greater than a predetermined height (for example, several meters). Mesh data is extracted as single tree mesh data. Examples of the shape similar to the cylindrical shape include a shape having a circular arc, a semi-cylindrical shape, a semicylindrical shape, and the like. The predetermined height is set, for example, to a height higher than the ground, so that mesh data with a height corresponding to a single tree can be extracted. For the extraction method of single tree mesh data, an extraction method for extracting mesh data having a shape similar to a cylinder may be adopted, or an extraction method for extracting mesh data having a height equal to or higher than a predetermined height may be adopted. may be used, any extraction method may be used, or both may be combined. As a result, only single tree mesh data corresponding to single trees can be extracted from the mesh data.

Now, when the generation control unit 102 completes the generation, next, the estimation control unit 103 of the terminal device 11 selects a plurality of points located on the same plane in the horizontal direction with respect to the generated single tree mesh data. Assuming that it is part of the circle composed of the bark of the tree, extrapolate the circle to the plurality of points, and use the radius of the extrapolated circle to convert the bark data of a single tree that does not have single tree mesh data into a single tree It is estimated as virtual data of tree bark (Fig. 2: S103).

Here, the estimation method of the estimation control unit 103 is not particularly limited. For example, the obtained single tree mesh data basically corresponds only to the bark of the front part of the single tree irradiated with the laser of the three-dimensional laser scanner 10. There are no bark data on the dorsal part of In addition, depending on the scan range and scan mode of the measurer, the single tree mesh data corresponds to the portion where the three-dimensional point cloud data could be acquired by the three-dimensional laser scanner 10, and the single tree mesh data does not exist. corresponds to the portion where the 3D point group data could not be acquired by the 3D laser scanner 10 .

Therefore, in principle, if the measurer holds the three-dimensional laser scanner 10 and goes around the back side of the single tree and irradiates the back side of the single tree with the three-dimensional laser scanner 10, the single tree Single tree mesh data corresponding to the bark on the back of the tree can be obtained, but in that case, the operator must go around the back side of the single tree, which takes time and effort. Moreover, it is necessary to match the single tree mesh data of the bark of the front part of the single tree with the single tree mesh data of the bark of the back part of the single tree, which takes a long processing time.

Therefore, in the present invention, basically, the three-dimensional point cloud data obtained by one scan of the three-dimensional laser scanner 10 is used to estimate the bark data of a single tree for which there is no single tree mesh data. . Specifically, as shown in FIG. 6A, first, the estimation control unit 103 acquires a plurality of points located on the same horizontal plane at a predetermined height h from the ground from the single tree mesh data. Next, the estimation control unit 103 regards the obtained points as part of a circle formed by the bark of a single tree, and extrapolates the circle C to the points. Here, the extrapolation method of the circle is not particularly limited. For example, after assuming the formula of a circle with a predetermined radius, by applying the least squares method, the formula of the circle C extrapolated to a plurality of points can be obtained.

Then, as shown in FIG. 6A, the estimation control unit 103 uses the radius R obtained by the extrapolated circle C formula to form data of a circle in which a plurality of points do not exist as virtual data. Estimate data. Here, the virtual data is a part of the circle C using the radius R and does not contain the single tree mesh data.

Thus, in the present invention, the bark of a single tree is substantially circular on the same plane in the horizontal direction, and a plurality of points in the horizontal direction of the single tree mesh data form part of the circle. By using it, it is possible to obtain bark data of a single tree for which there is no single tree mesh data, from a single set of three-dimensional point cloud data. Therefore, after scanning the bark of the front part of the single tree, the measurer does not need to go around the back part of the single tree and scan the bark of the back part of the single tree again. and time can be reduced.

Here, the predetermined height h for extrapolating the circle C can be arbitrarily set by the measurer, administrator, or the like. Further, for example, as shown in FIG. 6B, by stepwise repeating the extrapolation of a circle C for a plurality of points of the single tree mesh data at a predetermined interval i along the vertical direction from the ground, A stepped circle C corresponding to the stepped height and its radius R can be calculated. By calculating the stepwise circle C and its radius R in this way, it is possible to obtain virtual data according to height, and also to calculate the difference in the diameter of the single tree according to the height, the bending of the single tree, etc. It is possible to confirm.

FIG. 7 shows the case where the bark data of a single tree for which there is no single tree mesh data is estimated from the single tree mesh data shown in FIG. 5A. In this way, even if there is no single tree mesh data, by extrapolating the circle C and calculating the radius R of the circle C, virtual data is supplemented in addition to the single tree mesh data, It is possible to create single trees of books. In addition, in FIG. 7, the portion where the single tree mesh data does not exist is also estimated from the radius R of the circle C. As shown in FIG.

Now, when the estimation by the estimation control unit 103 is completed, next, the formation control unit 104 of the terminal device 11 generates the single tree mesh data, the estimated virtual data, the radius R of the circle C of the virtual data, and the unit Based on the height information of the tree, a three-dimensional cylinder model is formed as a three-dimensional model of the trunk of a single tree (FIG. 2: S104).

Here, since the scanning range in which the measurer directs the three-dimensional laser scanner 10 to the single tree in the forest is usually limited, the three-dimensional point cloud data can be obtained from all single trees taller than the measurer's height. Acquiring 3D point cloud data is difficult.

On the other hand, if there is information on the trunk of a single tree, it is possible to make an approximate calculation of the volume of standing trees and the accumulation of forests, which are related to the multifaceted functions of forests. That is, these approximations do not require single tree canopy information, and single tree trunk information is sufficient.

Therefore, in the present invention, a three-dimensional model of the trunk of a single tree is created using the minimum necessary information, which is used to calculate the volume of standing trees and the accumulation of forests.

Here, the formation method of the formation control unit 104 is not particularly limited. For example, as shown in FIG. 8, the single tree mesh data corresponding to the trunk near the base of the single tree, the virtual data, and the radius R of the circle C have already been obtained. , it is possible to form a three-dimensional cylinder model consisting of height and radius, and this three-dimensional cylinder model can be formed as a three-dimensional model of the trunk of a single tree.

Here, there is no particular limitation on the method by which the formation control unit 104 acquires the tree height information of a single tree. For example, when the measurer acquires the tree height information of the single tree by visually confirming the height H of the single tree, by directly inputting the tree height information into the terminal device 11, the formation control unit 104 , the height information of a single tree can be obtained.

In addition, if the measurer has a laser rangefinder in addition to the three-dimensional laser scanner 10, the single tree can be measured by measuring the height H of the single tree using the laser rangefinder. and directly inputting the tree height information to the terminal device 11 or inputting it to the terminal device 11 via wireless communication such as Bluetooth, the formation control unit 104 acquires the tree height information of the single tree. can be done.

Furthermore, when the position information of the measurement point is acquired when acquiring the three-dimensional point cloud data, a DEM (Digital Elevation Model) and a DSM (Digital Elevation Model) corresponding to the position information of the measurement point Surface Model, Numerical Surface Model) exists, the tree height information of a single tree may be obtained using DEM and DSM. The DEM indicates the elevation of the ground surface at a given point, and the DSM indicates the height of the earth's surface including buildings and trees at the given point. DEM and DSM are obtained by, for example, airborne laser surveying, drone photogrammetry, drone laser surveying, surveying using artificial satellites, and the like. Therefore, the formation control unit 104 acquires the DEM and DSM of the point corresponding to the position information of the three-dimensional point cloud data of the single tree, and subtracts the DEM from the DSM to obtain the single tree at the measurement point of the position information. By calculating as the tree height information of , the tree height information of a single tree can be obtained. The tree height information of this single tree can be considered as the average value of the tree height information of a plurality of single trees spreading around the position information.

Then, when the formation control unit 104 acquires the tree height information of the single tree, as shown in FIG. is calculated, and the first height H1 is subtracted from the tree height information H of the single tree to calculate the second height H2 of the portion where no data exists. Next, the formation control unit 104 forms an upper three-dimensional cylindrical model using the radius R and the second height H2 from the upper end of the single tree mesh data and virtual data to the tree height information of the single tree ( A three-dimensional cylinder with a radius R and a second height H2 is extended from the upper end of the single tree mesh data and the virtual data), and the upper three-dimensional cylinder model, the single tree mesh data, and the virtual data are all combined into a single tree form a three-dimensional model of the trunk of

Here, the radius R to be used may be, for example, the radius R of the circle C at the upper end of the virtual data. can be This makes it possible to easily form a three-dimensional model of the trunk of a single tree.

If there is no data from the ground to the lower end of the single tree mesh data and virtual data, the formation control unit 104 determines the number of points from the ground to the lower end of the single tree mesh data and virtual data where there is no data. A third height H3 is calculated, and a lower three-dimensional cylinder model is formed using the radius R and the third height H3 from the lower end of the single tree mesh data and virtual data to the ground (radius R and third height H3 A three-dimensional cylinder with a height H3 is extended from the lower end of the single tree mesh data and the virtual data), and all of the lower three-dimensional cylinder model, the upper three-dimensional cylinder model, the single tree mesh data, and the virtual data can be used to form a three-dimensional model of a single tree trunk.

Since such a 3D model of the trunk of a single tree includes information such as the diameter and height of a single tree, distortion within the scan range, etc., the 3D model of the trunk of a single tree can be divided into , it becomes possible to use lumber from single trees before felling, construction materials, etc. In addition, since the three-dimensional model of the trunk of a single tree can be output as three-dimensional CAD data (DXF, OBJ, STL file, etc.), it can be utilized by sawmills, construction shops, architects, and the like. The installation position of the three-dimensional model of the trunk of the formed single tree can be calculated, for example, by adding the center position information of the circle (diameter) of the single tree to the position information of the measurement point obtained in advance. I can.

Now, when formation of the formation control unit 104 is completed, next, the repetition control unit 105 of the terminal device 11 processes other single tree mesh data that does not form a three-dimensional model of the trunk of a single tree among the mesh data. , estimation of virtual data, and formation of a three-dimensional model of the trunk of a single tree are repeated (FIG. 2: S105).

Here, the repetition method of the repetition control unit 105 is not particularly limited. For example, the repetition control unit 105 determines whether or not other single tree mesh data that does not form a three-dimensional model of the trunk of a single tree exists among the mesh data ( FIG. 2 : S105).

Here, as shown in FIG. 9A, three single tree mesh data exist in the previous mesh data, and for one single tree mesh data, estimation of virtual data and three-dimensional model of the trunk of the single tree are performed. Forming and doing. Therefore, there are two single tree mesh data that do not form a three-dimensional model of the single tree trunk.

Therefore, the repetition control unit 105 determines that other single tree mesh data exists (FIG. 2: S105 YES), returns to S103, and the repetition control unit 105 controls the other single tree mesh data via the estimation control unit 103. Virtual data is estimated from the data (Fig. 2: S103). Next, in S104, the repetition control unit 105 forms a three-dimensional model of the trunk of a single tree from other single tree mesh data and virtual data via the formation control unit 104 (FIG. 2: S104). For the remaining single tree mesh data, similarly, the estimation of virtual data and the formation of the three-dimensional model of the trunk of the single tree are repeated. In this way, by forming all of the single tree mesh data existing in the mesh data into a three-dimensional model of the trunk of a single tree, as shown in FIG. 9B, three-dimensional models of multiple single tree trunks are accumulated. can do In addition, the measurer can easily create a three-dimensional model of the trunk of a single tree in the scanning range obtained by one scanning.

Now, when all the single tree mesh data have been processed, in S105 the repetition control unit 105 determines that there is no other single tree mesh data (Fig. 2: S105 NO). This completes the repetition of the repetition control unit 105 .

Here, the measurer determines whether or not to scan with the three-dimensional laser scanner 10 at another measurement point (Fig. 2: S106). For example, as shown in FIG. 10, if there are a plurality of other single trees spreading and it is necessary to scan at other measurement points, the measurer determines to scan (FIG. 2: S106 YES), and the three-dimensional Carrying the laser scanner 10 and the terminal device 11, the user moves to another measurement point. Then, when returning to S101 and scanning at another measurement point, the acquisition control unit 101 acquires three-dimensional point cloud data of a single tree ( FIG. 2 : S101). Subsequent processing is repeated from S102 to S105.

Here, in the present invention, as shown in FIG. 10, the measurer only needs to scan the bark of the front part (one side part) of a single tree in the forest, thus reducing the time and effort required for scanning and enabling the measurement. People can scan easily.

On the other hand, there is no need to scan at other measurement points, and the measurer determines not to scan (Fig. 2: S106 NO), and completes all processing. Here, the three-dimensional model of the trunk of a single tree is formed entirely in the range scanned by the measurer. Such a three-dimensional model of the trunk of a single tree can be utilized as a three-dimensional data platform for forests, for example, as shown in FIG. 11A. The 3D data platform for forests makes it possible to create a 3D map showing the position of a single tree in a forest, and a 2D map obtained by converting a 3D model into 2D data. In addition, the measurer periodically scans the single trees in the forest to update the three-dimensional model of the trunk of the single tree and accumulate changes over time in the three-dimensional model of the trunk of the single tree. can be done.

Also, if there is a three-dimensional model of the trunk of a single tree, as shown in FIG. 11B, it is possible to simulate cutting the tree when cutting the single tree. In the tree trimming simulation, by dividing the three-dimensional model of the trunk of the formed single tree into a predetermined setting range, it is possible to calculate the lumber volume (volume) of the single tree that can be obtained, and calculate the value of the tree trimming. In addition, since it is possible to simulate changes in the forest after tree removal, it is also possible to predict the deterioration of water and soil conservation function, global environment conservation function, and biological protection function after tree removal.

If tree crown data corresponding to the crown of a single tree exists for the three-dimensional model of the trunk of a single tree, crown data M2 is added to the three-dimensional model M1 of the trunk of a single tree as shown in FIG. 12A. It is also possible to synthesize by The three-dimensional model M1 of the trunk of a single tree can be obtained by scanning by an operator, and the crown data M2 can be obtained from external data such as DEM and DSM. When the crown data M2 is obtained from external data, for example, the formation control unit 104 corresponds to the position information associated with the three-dimensional point cloud data and the position information of the three-dimensional model M1 of the single tree trunk from the external data. The crown data M2 of the positional information is acquired, and the acquired tree crown data M2 is combined with the three-dimensional model M1 of the trunk of the single tree. By synthesizing the crown data M2 with the three-dimensional model M1 of the trunk of the single tree in this manner, a more realistic three-dimensional model of the single tree can be formed.

Also, if there is a three-dimensional model of the trunk of a single tree, it is also possible to output a database of the numerical values of the elements that make up the three-dimensional model, in addition to the three-dimensional model that is easy to understand visually. Become. For example, as shown in FIG. 12B, the database 1200 contains a number (No) 1201, a latitude 1202, a longitude 1203, an altitude 1204, a single tree species 1205, a diameter 1206, and a tree height (information) 1207. , and volume 1208 are stored in association with each other. A number 1201 is assigned to the trunk of a single tree. A latitude 1202, a longitude 1203, and an altitude 1204 indicate the installation position of the trunk of a single tree. Diameter 1206 indicates the diameter of the circle of the three-dimensional model of the single tree trunk. A wood volume 1208 is calculated from a trunk diameter 1206 and a tree height 1207 of a single tree. Note that the calculation method of the volume 1208 is not particularly limited. By outputting this database 1200 from the single tree modeling system 1, the measurer or the like can confirm the tree information of the single tree in the forest at a glance.

In addition, the position information of the terminal device 11 may be position information obtained by an IMU (inertial measurement unit) built into the terminal device, in addition to GPS position information and position information obtained by a two-frequency multi-GNSS receiver. As for the location information of the terminal device, any one of GPS location information, location information of the dual-frequency multi-GNSS receiver, and location information of the IMU may be selected by the measurement person. Here, when the measurement person selects the position information of the IMU and obtains the position information of the terminal device 11 at a plurality of measurement points with the position information of the IMU, the position information of the IMU is characterized by the accumulation of errors. . Therefore, in order to prevent the accumulation of errors, the operator sets a predetermined survey point as the starting survey point, goes around the forest, returns to the starting survey point, and uses the starting survey point as the final survey point. If set, the measurement person may correct the position information of the terminal device 11 as shown in FIG. 13 by performing closing processing from the start measurement point to the end measurement point. By correcting the position information of the terminal device 11, it is possible to correct the 3D point cloud data associated with the position information of the terminal device 11 and the 3D coordinate values of the 3D model.

Thus, in the present invention, it is possible to easily and quickly form a three-dimensional model of a single tree by scanning a portion of the single tree with a three-dimensional laser scanner.

In the embodiment of the present invention, the terminal device 11 is configured to include each control unit, but the program for realizing each control unit may be stored in a storage medium and the storage medium may be provided. do not have. In this configuration, the program is read by the device, and the device implements each control unit. In that case, the program itself read from the recording medium exhibits the effects of the present invention. Furthermore, it is also possible to provide a method of storing the steps executed by each control unit in a hard disk.

INDUSTRIAL APPLICABILITY As described above, the single tree modeling system and single tree modeling method according to the present invention are useful in fields ranging from upstream to downstream in the forestry industry in which three-dimensional models of entire single trees are acquired, shared, and utilized. It is effective as a single tree modeling system and a single tree modeling method that can easily and quickly form a single tree three-dimensional model by scanning a part of the tree with a three-dimensional laser scanner.

1 single tree modeling system 10 three-dimensional laser scanner 11 terminal device 101 acquisition control unit 102 generation control unit 103 estimation control unit 104 formation control unit 105 repetition control unit

Claims

an acquisition control unit that acquires three-dimensional point cloud data of a single tree by irradiating a single tree in the forest from a predetermined measurement point with a laser from a three-dimensional laser scanner;
a generation control unit that generates single tree mesh data representing the bark of the single tree by converting the acquired three-dimensional point cloud data into mesh data composed of polygons;
Considering a plurality of points located on the same plane in the horizontal direction with respect to the generated single tree mesh data as part of a circle formed by the bark of the single tree, extrapolating a circle to the plurality of points, an estimation control unit for estimating bark data of a single tree for which the single tree mesh data does not exist as virtual data of the bark of a single tree, using the extrapolated circle radius;
Based on the single tree mesh data, the estimated virtual data, the radius of the circle of the virtual data, and the tree height information of the single tree, a three-dimensional cylindrical model is generated from the three-dimensional tree trunk of the single tree. a formation control unit formed as a model;
A single tree modeling system with
The generation control unit performs noise processing to remove variations in points before meshing the three-dimensional point cloud data.
A single tree modeling system according to claim 1.
The generation control unit extracts, from the mesh data, mesh data having a shape similar to a cylindrical shape corresponding to the bark of the front part of a single tree, or mesh data having a predetermined height, as the single tree mesh data. to get,
A single tree modeling system according to claim 1 or 2.
The estimation control unit obtains, from the single tree mesh data, a plurality of points located on the same plane in the horizontal direction at a predetermined height from the ground, and the bark of the single tree constitutes the obtained plurality of points. extrapolate the circle to the points, and form the data of the circle where the points do not exist as virtual data using the radius obtained by the formula of the extrapolated circle to estimate the hypothetical data,
A single tree modeling system according to any one of claims 1-3.
The acquisition control unit acquires the position information of the terminal device using the communication unit of the terminal device constituting the single tree modeling system, and stores the position information of the terminal device in association with the three-dimensional point cloud data. ,
The formation control unit acquires the DEM and DSM of the point corresponding to the position information of the three-dimensional point cloud data, and subtracts the DEM from the DSM to obtain the tree height information of the single tree at the measurement point of the position information. Obtain the tree height information of the single tree by calculating as
A single tree modeling system according to any one of claims 1-4.
The formation control unit calculates a first height of a portion where data from the ground to the upper end of the single tree mesh data and the virtual data exists, and calculates the first height from the tree height information of the single tree. is subtracted to calculate the second height of the part where data does not exist, and from the upper end of the single tree mesh data and the virtual data to the single tree height information, the radius and the second height forming an upper three-dimensional cylindrical model using the above, and forming all of the upper three-dimensional cylindrical model, the single tree mesh data, and the virtual data as a three-dimensional model of the single tree trunk;
A single tree modeling system according to any one of claims 1-5.
If there is no data from the ground to the lower end of the single tree mesh data and the virtual data, the formation control unit determines that there is no data from the ground to the lower end of the single tree mesh data and the virtual data. calculating a third height of the portion and forming a lower three-dimensional cylindrical model from the lower end of the single tree mesh data and the virtual data to the ground using the radius and the third height; Forming all of the upper three-dimensional cylinder model, the lower three-dimensional cylinder model, the single tree mesh data, and the virtual data as a three-dimensional model of the trunk of the single tree;
A single tree modeling system according to claim 6.
The repetition control unit performs the estimation of the virtual data and the three-dimensional model of the trunk of the single tree for other single tree mesh data that does not form the three-dimensional model of the trunk of the single tree among the mesh data. repeating the formation of
A single tree modeling system according to any one of claims 1-7.
The acquisition control unit acquires the position information of the terminal device using the communication unit of the terminal device constituting the single tree modeling system, and stores the position information of the terminal device in association with the three-dimensional point cloud data. ,
The formation control unit acquires crown data of position information corresponding to position information associated with the three-dimensional point cloud data from external data, and converts the acquired crown data into a three-dimensional model of the trunk of the single tree. to synthesize,
A single tree modeling system according to any one of claims 1-8.
an acquisition control step of acquiring three-dimensional point cloud data of a single tree by irradiating a single tree in the forest from a predetermined measurement point with a laser of a three-dimensional laser scanner;
a generation control step of generating single tree mesh data representing the bark of the single tree by converting the acquired three-dimensional point cloud data into mesh data composed of polygons;
Considering a plurality of points located on the same plane in the horizontal direction with respect to the generated single tree mesh data as part of a circle formed by the bark of the single tree, extrapolating a circle to the plurality of points, an estimation control step of estimating the bark data of the single tree for which the single tree mesh data does not exist as virtual data of the bark of the single tree, using the extrapolated circle radius;
Based on the single tree mesh data, the estimated virtual data, the radius of the circle of the virtual data, and the tree height information of the single tree, a three-dimensional cylindrical model is generated from the three-dimensional tree trunk of the single tree. A formation control process to form as a model;
A monotree modeling method comprising