JP2006011544A - Method for generating earth surface polygon - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide earth surface polygon data capable of realizing natural three-dimensional display. <P>SOLUTION: A terminal 100, which is an earth surface polygon data generating device, inputs earth surface data 202 applying altitude at each point of time by a rectangular lattice with fixed intervals. The terminal 100 generates triangular earth surface polygon data by dividing each rectangular lattice of the earth surface data by polygonal lines. In this case, the polygon of two polygon lines of each lattice whose both ends has a smaller latitude difference is selected. Thus, it is possible to divide the lattice on a slope to a direction relatively following not a ridge line but a contour line. Therefore, it is possible to suppress any unnatural ridge line from being expressed during three-dimensional display. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a method for generating triangular polygon data for three-dimensional display of the ground surface.

  Three-dimensional display of the ground surface and buildings is realized using electronic map data. The altitude data on the ground surface is often given at the intersection of square meshes prepared at regular intervals, such as 50 m. Since these four points do not necessarily exist on the same plane, a triangular polygon obtained by dividing the ground surface into triangles is used for the three-dimensional display of the ground surface for the convenience of performing a three-dimensional display using a planar polygon. Is done. As a division method, for example, Patent Document 1 states that “for each square, one of the two diagonal lines having the higher elevation difference is selected and tied, and each square is divided into two triangles. Is disclosed.

Japanese Patent No. 3376868

  As described above, the conventional technique uses a diagonal line having a larger elevation difference. For example, when the ground surface is a slope, this is almost synonymous with dividing a quadrilateral polygon along its ridgeline. However, with this division method, it has been found that the three-dimensional display of the ground surface may become unnatural. FIG. 8 is an explanatory diagram showing a three-dimensional display example according to the prior art. The building and road Rp constructed on the slope of the mountain are shown. In the prior art, as a result of the polygon being divided along the ridgeline, the ridgeline Lp appears prominently on the road Rp running along the contour line as shown in the figure. The building installation part Bp may be displayed as if it is buried in the ground. An object of the present invention is to solve such problems and provide a technology for generating ground surface polygon data capable of realizing natural three-dimensional display.

  The polygon data generation apparatus of the present invention generates triangular polygon data for displaying the ground surface in three dimensions in the following procedure. The polygon data generation apparatus inputs ground surface data in which positions and altitudes are recorded for a plurality of grid points. Then, four points are sequentially specified from the plurality of grid points as processing targets. However, these four points are specified so that a quadrangle that does not include other lattice points can be formed. Next, the polygon data generating apparatus divides the quadrilateral by the diagonal line having the smaller elevation difference between the two ends of the two quadrilateral lines formed by the specified four points, thereby obtaining two triangular polygon data. Generate.

  According to the present invention, on the contrary to the prior art, the diagonal line on the side where the difference in elevation is small is adopted. By doing so, the quadrilateral polygon on the slope is also divided in a direction relatively along the contour line, not the ridge line. Therefore, an unnatural display such as a ridge line appearing across the road can be avoided.

  The present invention may target either the lattice point of the unstructured lattice or the lattice point of the structured lattice. As the latter example, for example, a rectangular structured grid can be used. An example of such a structured grid is a square mesh having a constant distance such as 50 m. For such a structured grid, if polygons are divided with respect to the four vertices of each grid, the processing can be simplified.

  When dividing the quadrangle, if the difference in elevation difference between the two diagonals is less than or equal to a predetermined value, for example, a case where the values are almost equal may occur. In such a case, any one may be selected at random, or may be selected according to a certain rule. As an example of the latter, for example, a diagonal line on the side having high continuity with the boundary line of the lattice adjacent to the quadrangle may be selected. Continuity can be evaluated in various ways. For example, as long as the diagonal line is on the boundary line of another polygon, it is possible to extend the outside of the rectangle to be processed and evaluate by the length. As another method, the score may be evaluated such that one end of the diagonal line is connected to a diagonal line obtained by dividing an adjacent quadrilateral polygon, and one point when the diagonal line is connected, and 0 point when the diagonal line is not connected. With these methods, it is possible to realize a more natural three-dimensional display by considering the continuity of the boundary lines dividing the lattice.

  In the present invention, it is not always necessary to apply the above dividing method to all the lattices. In the map, there are also areas such as mountainous areas where it is preferable to display the three-dimensional image with the ridgeline highlighted. Therefore, area attribute data used for determining whether or not each grid point exists in the area where the edge line should be emphasized in this way may be prepared, and a plurality of types of division methods may be used properly. The region attribute data may be set in association with each grid point, for example, or may be prepared as a database separate from the ground surface data, such as two-dimensional map data. If it is determined that the quadrangle to be processed exists in the region where the ridgeline should be emphasized based on the attribute data, the polygon data generation device divides the diagonal line by the diagonal line with the higher elevation difference at both ends. To emphasize the ridgeline. In other cases, the above-described division processing may be applied, or different types of division processing may be provided so as to be used properly depending on conditions.

  The present invention does not have to have all the features described above, and some of them may be omitted or combined as appropriate. The present invention may be configured as a polygon data generation method in which polygon data is generated by a computer in addition to an aspect as a polygon data generation apparatus. Moreover, you may comprise as a computer program for making a computer implement | achieve this process, and a computer-readable recording medium which recorded this computer program. When configured as a polygon data generation device, it may be a stand-alone device or may be configured by connecting a terminal and a server via a network. Recording media include flexible disks, CD-ROMs, magneto-optical disks, IC cards, ROM cartridges, punch cards, printed matter printed with codes such as bar codes, computer internal storage devices (memory such as RAM and ROM), and Various media that can be read by a computer, such as an external storage device, can be used.

Embodiments of the present invention will be described in the following order.
A. System configuration:
B. Ground surface data:
C. Ground surface polygon data generation processing:
C1. Split processing:
C2. Edge enhancement processing:
C3. Variations of split processing:
D. effect:

A. System configuration:
FIG. 1 is an explanatory diagram showing the overall configuration of a polygon data generation system as an embodiment. The route guidance system according to the embodiment is configured by connecting a terminal 100 functioning as a polygon data generation device and various servers for providing various databases via a network NET. The network NET may be limited such as a LAN or an intranet, or may be a wide area such as the Internet.

  Examples of the database provided by the server include the ground surface data 202, the 2D map DB 204, and the 3D map DB 210. The ground surface data 202 is data that gives the altitude of each point with a 50 m mesh. The two-dimensional map DB 204 is data for displaying a two-dimensional map, and stores terrain, polygon data representing features, and attribute data. The 3D map DB 210 is a database for performing 3D display of a map, and includes ground surface polygon data 212 and feature polygon data 214. The ground surface polygon data 212 is data for displaying the ground surface in three dimensions, and is generated by the terminal 100. The feature polygon data 214 is data for three-dimensional display of roads, buildings, and other features. In the present embodiment, the feature polygon data 214 will be described as being prepared in advance.

  The functional blocks of the terminal 100 are also shown in the figure. In this embodiment, these functional blocks are configured in software by the CPU executing a predetermined computer program installed in the terminal 100. You may implement | achieve at least one part of a functional block by hardware.

  The command input unit 102 inputs commands based on user operations on the keyboard, mouse, and the like of the terminal 100. The data input / output unit 104 exchanges data with various databases via the network NET. The map display unit 112 displays a map three-dimensionally on the terminal 100 based on the three-dimensional map DB 210.

  The generation of the ground surface polygon data 212 is realized by the area determination unit 106, the division processing unit 108, and the ridge line enhancement processing unit 110. The area determination unit 106 determines the topography of each point in the ground surface data 202 and determines which of the division processing unit 108 and the ridge line enhancement processing unit 110 is used. The division processing unit 108 and the ridge line enhancement processing unit 110 each divide the ground surface data 202 into triangular polygons. Both differ in the method of division processing. The details of the division processing of the division processing unit 108 and the ridge line enhancement processing unit 110 will be described later. The generated surface polygon data 212 is stored in the surface polygon data 212 of the three-dimensional map DB 210 by the data input / output unit 104.

  The terminal 100 may further include a functional block for generating the feature polygon data 214. In addition, in order to perform the division processing, functional blocks other than the division processing unit 108 and the ridge line enhancement processing unit 110 may be provided, and the area determination unit 106 may control the use of three or more persons. Conversely, the ridge line enhancement processing unit 110 may be omitted, and the division processing unit 108 may be applied to all points. In this configuration, the area determination unit 106 can be omitted.

B. Ground surface data:
FIG. 2 is an explanatory diagram showing the structure of the ground surface data 202. For convenience of explanation, it is shown in a perspective view together with data examples of the two-dimensional map DB 204. The data located on the upper side is the data of the two-dimensional map DB 204, and the data located on the lower side is the ground surface data 202.

  The two-dimensional map DB 204 records terrain and feature drawing data and attribute data, as shown. Since various known data structures can be applied to the storage of these data, a detailed description regarding the data structure is omitted. Since the attribute data is recorded in the 2D map DB as described above, by identifying the latitude and longitude and referring to the 2D map DB 204, the land / sea distinction, the building The presence or absence can be determined.

  As shown in the figure, the ground surface data 202 is composed of rectangular meshes with a constant interval. In this embodiment, the interval is 50 m. Each mesh is defined by x and y coordinates defined on a plane. The lattice points of the mesh are expressed in a format of “P [number in x direction, number in y direction]”. Each lattice point is associated with three-dimensional position information (latitude, longitude, altitude). For example, the grid point P [1, 0] is “latitude = Lat, longitude = Lon, altitude = H”. In the present embodiment, the altitude means a probabilistic value in a 50 m square area centered on the lattice point. That is, the altitude value is not the exact altitude value of the grid point, but merely represents the altitude value at any point in the region. In the present embodiment, the ground surface data 202 is defined by a square mesh, but the mesh interval may be different in the x direction and the y direction.

C. Ground surface polygon data generation processing:
FIG. 3 is a flowchart of the ground surface polygon data generation process. This is a process executed by the CPU of the terminal 100 in accordance with a user command. When this processing is started, the CPU inputs the ground surface data 202 via the network NET (step S10), and sequentially executes the following processing for all grids (steps S12 to S52).

  First, the CPU specifies a grid to be processed (step S14). It may be specified randomly from an unprocessed grid, or may be specified according to a certain rule. Hereinafter, the specified grid is referred to as a processing target grid.

  The CPU determines in which area the processing target grid exists (step S16). In this embodiment, the area is divided into a city area where features such as houses, roads, and railways exist, and a mountain area where these features do not exist. This determination is performed according to the following procedure using the attribute data of the two-dimensional map DB 204. The judgment method is also shown in the figure.

  The CPU sets a circular area C1 having a constant diameter around the processing target grid P1. Although the radius of the circular area C1 can be arbitrarily set, in this embodiment, it is set to the same value as the mesh interval of the ground surface data 202. If there is a feature such as the house B1, road, railway, etc. within the circular area C1 set in this way, the CPU determines that the processing target grid exists in the city area, otherwise, it exists in the mountain area. Judge that. In the example in the figure, the grid P1 exists in the city area, and the grid P2 exists in the mountain area.

  Various other methods can be applied to the determination of the area. For example, the determination may be made based on whether or not a feature exists in the processing target grid without setting a circular region. In this embodiment, the determination is made with reference to the two-dimensional map DB 204, but attribute data such as a town area and a mountain area may be assigned to each grid point.

  The CPU generates ground surface polygon data by properly using two methods according to the determination result of the area. When the area is not a mountain area (step S18), that is, for the city area, the division process (step S20) is applied, and for the mountain area, the edge enhancement process is applied (step S40). Each processing content will be described later. In this embodiment, two types of processing are used properly, but more processing may be used properly. Conversely, steps S16, S18, and S40 may be omitted, and only the division process (step S20) may be used in a unified manner for all the processing target grids.

  The CPU stores the ground surface polygon data generated by the dividing process (step S20) or the ridge line emphasizing process (step S40) in the 3D map DB 210 via the network NET (step 50). The above processing is executed for all grids (step S52), and the ground surface polygon data generation processing is completed.

C1. Split processing:
FIG. 4 is a flowchart of the dividing process. This process corresponds to step S20 of the ground surface polygon data generation process (FIG. 3). In this process, a rectangular processing target grid is divided by diagonal lines to generate triangular polygon data.

  The CPU inputs the altitude of each grid point of the processing target grid (step S21), and calculates the altitude differences dh1 and dh2 at both ends for the two diagonal lines (step S22). The calculation method is shown in the figure. Assume that the altitudes of the grid points of the processing target grid are H1, H2, H3, and H4, respectively. In this case, the elevation difference between the two diagonal lines is given by “dh1 = | H1−H3 |, dh2 = | H2−H4 |”.

  The CPU compares the absolute value | dh1−dh2 | of the difference between the two values dh1 and dh2 with a predetermined threshold Th (step S23). If the absolute value | dh1−dh2 | is greater than the threshold Th, the CPU determines that there is a significant difference between the elevation differences dh1 and dh2, and divides the grid along the diagonal line with the smaller elevation difference (step S24). ). For example, when dh1> dh2, the lattice is divided by a diagonal line connecting H1 and H3 in the drawing shown in step S22. Conversely, when dh1 <dh2, the lattice is divided by a diagonal line connecting H2 and H4. The threshold value Th used in step S23 is a determination reference value as to whether or not there is a significant difference between the elevation differences dh1 and dh2, and can be arbitrarily set. If Th is set to a value of 0, the process of step S24 is performed even if there is a slight difference of about a calculation error between the elevation differences dh1 and dh2.

  When the absolute value | dh1−dh2 | is equal to or smaller than the threshold Th (step S23), the CPU determines that the elevation differences dh1 and dh2 are substantially the same, and uses each of the indices as an index for selecting a diagonal to be used for dividing the grid. Diagonal continuity indexes V1 and V2 are set (step S25). The continuity index is an index representing the degree to which the diagonal line of the processing target grid is continuously stretched out of the grid. In this embodiment, the stretchable length is used as an index.

  In the figure, the setting method of the continuity indexes V1 and V2 is illustrated. Consider a case where two diagonal lines DL1 and DL2 of the processing target grid P are evaluated. A lattice around the processing target lattice P and a diagonal line used for dividing the lattice are indicated by broken lines. The diagonal line DL1 is continuous only with the diagonal line DL11 in the lower right grid. If the length of the diagonal line DL1 is “1”, the stretchable length is “2”, so the continuity index V1 is set to “2”. The diagonal line DL2 is continuous with the upper right diagonal line DL21 and the lower left diagonal line DL22. Therefore, the continuity index V2 is set to “3”. If the diagonal line is continuous even outside the illustrated range, the continuity index may take a larger value.

  Various settings can be made for the continuity index. For example, the continuity index may be evaluated only within a grid adjacent to the grid to be processed (eight grids shown by broken lines in the figure). In this evaluation method, the number of continuous diagonal lines can be scored for evaluation. For example, for the diagonal line DL1, only the diagonal line DL11 is continuous, so the continuity index V1 may be 1 point, and for the diagonal line DL2, the continuity index V2 may be 2 points. In addition, parallelism with a diagonal line in a lattice adjacent to the processing target lattice in the vertical and horizontal directions may be considered. For example, for the diagonal line DL1, the left and right diagonal lines and the lower diagonal line are parallel, so the diagonal line DL2 is evaluated as one point because only the upper diagonal line is parallel, and this evaluation value is taken into account. A continuity index may be set.

  Based on the continuity index set in this way, the CPU divides the grid by the diagonal line with the larger value (step S26). By doing so, since the boundary lines of the grid are continuous over a relatively wide area, more natural three-dimensional display can be realized on the ground surface.

  Various methods can be used for the division process. For example, when the elevation differences dh1 and dh2 can be determined to be substantially the same (step S23: YES), either one of the two diagonal lines is selected at random. May be. The continuity index does not necessarily evaluate the degree to which the diagonal line of the processing target grid can be extended linearly. If the ground surface data 202 is given by an unstructured grid, the continuity index has a polygonal line shape. You may set a continuity parameter | index based on the length which can be extended | stretched.

C2. Edge enhancement processing:
The edge line emphasis process (step S40) in the ground surface polygon data generation process (FIG. 3) is a process in which the diagonal line selected in step S24 of the above-described division process (FIG. 4) is reversed. In other words, in the ridge line emphasis process, when there is a significant difference between the elevation differences dh1 and dh2 between the two diagonal lines, the CPU divides the grid along the diagonal line with the higher elevation difference.

  When it is determined that the elevation differences dh1 and dh2 are substantially the same, the same processing as the division processing is executed in the edge enhancement processing (steps S25 and S26 in FIG. 4). However, these processes may be selected in the opposite manner to the division process, and the grid may be divided at the diagonal line on the smaller continuity index in step S26.

C3. Variations of split processing:
As a method for selecting a diagonal line for dividing the grid, a method using the normal length of a triangular polygon generated when the diagonal line is temporarily divided, instead of the elevation differences dh1 and dh2 at both ends of the diagonal line, can be mentioned. Such a selection method will be described as a modification. This selection method is a process replacing steps S22 and S23 in FIG.

  FIG. 5 is an explanatory diagram showing a method for selecting a diagonal line using a normal line of a triangular polygon. The lattice points P0, P1, P2, and P3 of the processing target lattice are shown in a perspective view according to the position coordinates x and y on the plane and the altitude h. If the processing target grid is divided by the diagonal line P1P3, two triangular polygons, triangles P0P1P3 and P2P1P3, are generated. For each of these triangular polygons, normalized normal vectors N1, N2 can be obtained. Further, the components on the xy plane of the normal vectors N1 and N2 are N1xy and N2xy. The component value increases as the inclination of the triangular polygon from the ground surface increases. In the modification, the sum of the component values, that is, “N1xy + N2xy” is used as an evaluation value instead of the altitude difference.

  FIG. 5 illustrates the state in which the lattice is divided by the diagonal line P1P3, but the normal line and components on the xy plane can be obtained in the same manner for the state divided by the diagonal line P0P2. The evaluation value obtained based on this normal is defined as “N3xy + N4xy”.

  In the modified example, a diagonal line to be used for dividing the grid is selected based on the evaluation value thus obtained. In the dividing process (FIG. 4), in step S24, a diagonal line with a smaller evaluation value is selected, and in a ridge line emphasis process, a diagonal line with a larger evaluation value is selected.

  In the modification, the evaluation value for the state where the grid is divided by the diagonal line P1P3 may use the larger one of the components N1xy and N2xy on the xy plane of the normal vectors N1 and N2. The evaluation value for the state divided by the diagonal line P0P2 uses the larger of the components N3xy and N4xy on the normal line and the xy plane. In the dividing process, a diagonal line corresponding to the smaller evaluation value thus obtained is used. As described above, in the dividing process, various methods can be applied which are equivalent to selection criteria for selecting a diagonal line having a smaller elevation difference between both ends of the diagonal line.

D. effect:
FIG. 6 is an explanatory view showing a three-dimensional display example of the ground surface based on this embodiment. The building and road Rp constructed on the slope of the mountain are shown. Unlike the display according to the prior art (FIG. 8), when the ground surface polygon data according to this embodiment is used, a ridge line crossing the road Rd does not appear remarkably. Thus, according to the ground surface polygon data generation system of the present embodiment, unnaturalness in the three-dimensional display of the ground surface can be suppressed.

  In the embodiment, the division process and the ridge line enhancement process are separately used in the city area and the mountain area. Such proper use is not essential in the present embodiment, but by using such proper use, an unnatural ridge line is suppressed from appearing in a city area where features exist (see FIG. 6), while in a mountain area. It becomes possible to perform three-dimensional display with the ridgeline emphasized.

E. Second embodiment:
In the first embodiment, an example is shown in which ground surface polygon data is generated using the ground surface data 202 given by a square structured grid. This embodiment can also be applied to ground surface data given by an unstructured grid. Such an example will be described in the second embodiment.

  FIG. 7 is a flowchart of the ground surface polygon data generation process as the second embodiment. Similar to the first embodiment, this corresponds to the processing executed by the CPU of the terminal 100. When the process is started, the CPU inputs ground surface data (step S100). In this example, it is assumed that the ground surface data is given by an unstructured grid.

  The CPU executes processing for generating ground surface polygon data for all input grid points in the following procedure (steps S102 to S112). First, the CPU attempts to divide the ground surface into triangular polygons by Deloni division processing (step S104). Deloni division is a process known in numerical geometry for defining an unstructured grid.

  In the figure, the Delaunay division processing method is shown. Black circles represent lattice points. In Delaunay division, a circle passing through three or more of these lattice points is found. However, as shown by a circle Cd2 in the figure, a circle including another lattice point inside is excluded. Find circles (hereinafter referred to as “empty circles”) that do not include other lattice points, such as circles Cd1 and Cd3. A state passing through four or more lattice points like the circle Cd3 is called degeneracy.

  The CPU determines whether or not degeneration has occurred based on the empty circle setting result (step S106). If no degeneration has occurred, triangular ground surface polygon data is generated and stored by three points on the sky circle (step S110). When degeneration occurs, the triangle cannot be uniquely defined, so division processing is performed (step S108), and ground surface polygon data is generated and stored (step S110).

  The same processing as in the first embodiment (FIG. 4) can be applied to this division processing. Between step S106 and step S108, the process of steps S16 and S18 of the ground surface polygon data generation process (FIG. 3) in the first embodiment may be applied, and the division process and the edge line enhancement process may be used separately.

  In the second embodiment, when degeneration occurs in which five or more points are present on the empty circle, for example, the division process (step S108) may be executed after dividing the rectangle into squares by any diagonal line. At this time, the diagonal line may be selected at random, or the diagonal line that minimizes the elevation difference may be selected.

  According to the second embodiment described above, similarly to the first embodiment, ground surface polygon data capable of natural three-dimensional display can be generated for the unstructured grid.

It is explanatory drawing which shows the whole structure of the polygon data generation system as an Example. It is explanatory drawing which shows the structure of the ground surface data. It is a flowchart of a ground surface polygon data generation process. It is a flowchart of a division | segmentation process. It is explanatory drawing which shows the selection method of the diagonal using the normal line of a triangular polygon. It is explanatory drawing which shows the example of a three-dimensional display of the ground surface based on a present Example. It is a flowchart of the ground surface polygon data generation process as 2nd Example. It is explanatory drawing which shows the example of a three-dimensional display by a prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Terminal 102 ... Command input part 104 ... Data input / output part 106 ... Area determination part 108 ... Division processing part 110 ... Edge line emphasis processing part 112 ... Map display part 202 ... Ground surface data 204 ... Two-dimensional map DB
210 ... 3D map DB
212 ... ground surface polygon data 214 ... feature polygon data

Claims (6)

A polygon data generation device for generating triangular polygon data for three-dimensional display of the ground surface,
For a plurality of grid points, a ground surface data input unit for inputting ground surface data in which positions and altitudes are recorded,
A target grid point specifying unit that specifies, from the plurality of grid points, four points that can form a quadrilateral that does not include the grid point as a processing target;
A polygon data generation apparatus comprising: a division processing unit that generates two triangular polygon data by dividing the quadrilateral with a diagonal line having a smaller elevation difference between both ends of the two diagonal lines of the quadrilateral. The polygon data generation device according to claim 1,
The ground surface data is given by a rectangular structured grid,
The processing target is a polygon data generation device that is four vertices of each grid in the structured grid. The polygon data generation device according to claim 1,
When the difference in elevation difference with respect to the two diagonal lines is equal to or less than a predetermined value, the division processing unit performs polygon data division using the diagonal line having higher continuity with the boundary line of the grid adjacent to the quadrangle. Generator. The polygon data generation device according to claim 1,
The ground surface data input unit inputs region attribute data used to determine whether or not each grid point exists in a region where a ridge line should be emphasized,
When it is determined that the quadrangle specified by the target grid point specifying unit is present in the region where the ridgeline should be emphasized based on the region attribute data, the elevations at both ends of the two diagonal lines of the quadrangle A polygon data generation apparatus including a ridge line emphasis processing unit that generates two triangular polygon data by dividing the quadrangle by a diagonal line having a larger difference. A polygon data generation method for generating triangular polygon data for three-dimensional display of the ground surface,
As a process executed by the computer,
For a plurality of grid points, a ground surface data input process for inputting ground surface data in which positions and altitudes are recorded,
A target grid point specifying step of specifying, from the plurality of grid points, four points that can form a quadrilateral that does not include the grid point as a processing target;
A polygon data generation method comprising: a division processing step of generating two triangular polygon data by dividing the quadrilateral by a diagonal line having a smaller elevation difference between both ends of the two diagonal lines of the quadrilateral. A computer program for generating triangular polygon data for three-dimensional display of the ground surface,
For a plurality of grid points, ground surface data input program code for inputting ground surface data in which positions and altitudes are recorded, and
A target grid point specifying program code for specifying, from the plurality of grid points, four points that can form a quadrilateral that does not include the grid point as a processing target;
A computer program comprising: a division processing program code for generating two triangular polygon data by dividing the quadrilateral with a diagonal line having a smaller elevation difference between both ends of the two diagonal lines of the quadrilateral.