JPH09102029A - Topography data interpolation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a topography data interpolation device which can reflect the fine variation pattern of real topography on an interpolation value when topography is divided into mesh structures and given altitude data is interpolated. SOLUTION: An interpolation arithmetic processing part 108 which obtains an interpolation point coordinate dividing the size of the mesh of altitude data into two by a mesh division processing, selects a latest contact point in an orthogonal position relation for the respective interpolation point coordinates, divides the whole areas of interpolation objects into the areas of high correlation of altitude fluctuation, calculates random numbers having distribution equal to altitude variance in the divided areas, calculates the altitude of the interpolation points by adding the value of the random numbers to the altitude average value of the latest contact points of the interpolation points in the pertinent area and repeats an altitude calculation processing for re-interpolation points obtained by re-dividing the mesh size by the number of dividing times, which is previously set, is provided.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a terrain data interpolating device for dividing terrain into a mesh structure and interpolating given altitude data.

[0002]

2. Description of the Related Art In recent years, an increasing number of map display devices display numerically provided map data on a CRT or the like using an electronic medium. At this time, since the number of observation points of the map data that can be prepared (hereinafter, also referred to as actual data), that is, the data density is limited, the map displayed relatively becomes rough when the scale ratio on the display screen becomes small. Therefore, to draw a more realistic topographic map,
Interpolation of actual data is required. As a publicly known document of this interpolation method, there is, for example, a "terrain data creating device" disclosed in Japanese Patent Laid-Open No. 4-107687.

FIG. 19 is an explanatory view of the conventional interpolation method shown in the above patent publication, and only the drawings relevant to the present invention are extracted. The conventional interpolation method will be described below with reference to FIG. In the invention disclosed in the above-mentioned patent publication, as shown in (a) and (b) of FIG. 19, an orthogonal array set P (i, j) (that is, the actual data Horizontal component (hereinafter referred to as mesh data), the two-division point T (i, j) with the same coordinates is obtained, and this T (i, j)
The deviation corresponding to the distance from j) to the adjacent point is set by a fractal calculation method, and this is added to the average value (a + c) / 2 of the altitude data a and c at P (i, j) to interpolate points. The altitude of T (i, j) is obtained (this process is referred to as midpoint process in the same literature). This midpoint processing is shown in FIG.
To (d) are sequentially repeated until the final result is as shown in (d) of FIG. Compared to FIG. 19A, FIG. 19D shows that each mesh is divided into four equal parts.

[0004]

However, the above interpolation method has the following problems. That is, for example, T (i, j) is P (i, j), P (i, j + 1), P (i + 1,
j) and P (i + 1, j + 1) are obtained from four points, whereas T
(i, j-1) is two points of P (i, j) and P (i + 1, j), and T (i, j + 1) is P (i, j + 1) and P (i + 1). , j + 1), T (i-1, j) is P (i, j)
j) and P (i, j + 1), and T (i + 1, j) is P (i + 1, j) and P (i + 1, j).
Only two points (i + 1, j + 1) are used. As a result, the latter 4
Regarding points, only the undulations in one direction are taken into consideration when deriving the points, and it is difficult to reflect the actual fine undulation pattern in the interpolation value. Further, the deviation added to the altitude average value during the intermediate processing is calculated by the following equation (1).

[0005]

(Equation 1)

Given k, which determines the standard deviation of the Gaussian distribution,
Although the parameters h are introduced, no specific guideline for giving these parameters is mentioned.

[0007]

A terrain data interpolating apparatus according to the present invention is a terrain data interpolating apparatus for interpolating terrain into a mesh structure and interpolating given altitude data. The interpolating point coordinates that divide the mesh size into two are obtained by using, and the nearest contact point that is in a positional relationship orthogonal to each of these interpolating point coordinates is selected. Divide, calculate a random number having a distribution equal to the altitude variance in each of the divided regions, and add the value of this random number to the altitude average value of the closest points of the interpolation points in the region, The altitude is calculated, and the altitude calculation process is repeated for the re-interpolation points obtained by re-dividing the mesh size a preset number of times. Those having an arithmetic processing unit.

[0008]

1 is a diagram showing an example of the configuration of a terrain data drawing apparatus using the terrain data interpolation apparatus of the present invention. In FIG. 1, 101 is a topographic map display device, and 102
Is a display control unit, 103 is actual terrain data (actual data)
Is a terrain data base, 104 is a terrain data interpolating device of the present invention.
An input / output unit 106, an interpolation processing control unit 107, an interpolation calculation processing unit 108, and a terrain data memory 109 for temporarily storing data during the interpolation processing. Reference numeral 105 is a data bus that connects the above-mentioned devices 101 to 104. In FIG. 1, when the interpolation process is not performed, the display control unit 102 reads the data from the terrain database 103 and displays it on the display device 101. When performing the interpolation process, the control is transferred from the display control unit 102 to the interpolation process control unit 107 in the terrain data interpolation device 104.

The interpolation processing control unit 107 first secures a terrain data memory 109 of a sufficient size to store the actual data and the interpolation data. For example, when the actual data consists of m × m mesh, the data size is Dmm (the unit is arbitrary), and when the mesh size is halved by the interpolation process, the data amount is quadrupled. As a result, Dmm × 4 is required. Here, it is assumed that the number of divisions of the mesh size (or the mesh size after division) is designated in advance by an appropriate input means (not shown). Next, the actual data to be interpolated is read from the terrain database 103 via the input / output unit 106 and stored in the terrain data memory 109. The storage position of each data at this time is appropriately determined from the actual data size and the number of mesh divisions. Then, the interpolation calculation processing unit 107 follows the processing flow to obtain the terrain data memory 109.
While the necessary data is being read from, the calculation associated with the interpolation processing is performed, and the resulting interpolation data is written in an appropriate position on the terrain data memory 109.

When the processing flow ends, the interpolation processing control unit 1
Reference numeral 07 returns the control right to the display control unit 102 to execute drawing. The actual data at the time of interpolation is also the topographical data memory 109.
The terrain data memory 109 stores a semiconductor memory, and the terrain database 103 stores a relatively slow CD-.
This is because it is assumed that a ROM, a magnetic memory, or the like is used. Therefore, when it is not necessary to consider the speed or when the terrain data memory 109 having a sufficient size cannot be secured, the actual data is stored in the terrain database 103. Processing will be performed with the data stored.

The method of interpolating the terrain data (that is, three-dimensional data) will be described below. FIG. 12 is a diagram showing a topographic map composed of a plurality of rectangular areas. Here, it is assumed that the topographical data is given within a certain rectangular area as shown in FIG. 12, for example. Further, generally, the horizontal component of this topographical data is given as a difference value when an arbitrary point inside or outside the area is set as an origin, but in the following processing, there is no part that exists at this origin position.

FIG. 2 is a diagram for explaining the mesh division method according to the present invention, and the mesh (3 × 3) division process will be described with reference to FIGS. Figure 2 (a)
What is indicated as [0] is the actual data. First,
Interpolated coordinates [1] (see FIG. 2) in the center of the rectangle forming each mesh.
(See (b) of)). As the temporary altitude of each [1], the average value of the altitude data of the four closest [0] points is used. Next, interpolation coordinates [2] are set at the bisectors of the rectangular sides that form each mesh on the sides of the rectangular area formed by all the meshes (see (c) in FIG. 2). As the temporary altitude of each [2], the average value of the altitude data of the two nearest [0] points and one [1] point is used. Finally, interpolation coordinates [3] are set at the bisectors of each side of the rectangle forming each mesh except for the side of the rectangular area formed of all meshes (see (d) of FIG. 2). As the provisional altitude of each [3], the average value of the altitude data of the two nearest [0] points and two [1] points is used. The interpolation points [1] to [3] thus obtained are [0]
When rearranged by adding to the actual data of, the mesh size becomes half of the original nut. After that, the above process may be repeated until the required number of divisions (the number of divisions) is reached.

Embodiment 1 FIG. 3 is a flowchart of all interpolation processing including mesh division processing according to the first embodiment of the present invention. The numbers following S in the figure indicate step numbers. When the interpolation process is started in S301 of FIG. 3, S30
The actual data {0} is read in 2 and then, in S303, a numerical value called a group number is set for each coordinate point. This group number is used to classify a region in the real data having a high correlation of altitude fluctuations (that is, a similar undulation state), and is used when obtaining altitude data after mesh division. Details of the group number setting process will be described with reference to FIG. In S304 of FIG. 3, an initial value of the number of divisions is set to 0, and in S305, interpolation coordinates are set by division and an average altitude of the corresponding coordinates is set. In addition,
In the figure, the symbol {} is used to represent a set of all data of coordinates [0] to [3]. The mesh division processing reaches the preset number of divisions set in advance (S304 → S).
317) or until the mesh size after division reaches a specified size.

FIG. 4 is a diagram for explaining the grouping processing method according to the present invention, and FIG. 5 is a diagram showing the structure of the terrain data according to the present invention. FIG. 4A shows an example of setting the group number. In this example, the entire area is divided into three areas 0 to 2 having similar undulations. The point A is shared by four areas 0, 1, 2, 2. A data structure as shown in FIG. 5 is prepared in order to assign a group number to each coordinate. The data structure includes mesh coordinates (x, y) (that is, horizontal components of coordinates), altitude data (h), and group numbers (n0 (,
n1 (, n2 (, n3)))). Up to four group numbers can be specified, because in the case of a rectangular mesh, one vertex may be shared by up to four meshes. As a result, the group number of point A is written (0,1,2) (excluding duplicate numbers), and points C and E are written (0,2).

The area division based on the group number is not limited to the one following the original mesh shape, but may be based on the triangular shape as seen in FGI and FIH in FIG. Area can be set. (The difference between BCDE and FGHI arises from the fact that G is (0) while C is (0,2))
Since it is necessary to calculate the altitude variance value in each group area in the altitude calculation process described later, one group must include at least three points. Also, the areas do not necessarily have to be continuous.

An example of how to give such a group number in S303 of FIG. 3 is shown below. First, after dividing the whole area into small areas of appropriate size that are easy to handle as data, (1) directly looking at the topography,
Roughly group into mountainous areas, etc. (2) Obtain information about the average and variance values of the altitudes of all small areas and their distribution status, and then automatically group into groups based on that information. Further, a method of inheriting and setting the group number of each small area as the group number of the mesh vertices included in the small area is conceivable.

Next, the procedure of grouping processing after mesh division in S306 will be described with reference to FIG. FIG. 4C shows the region BCDE of FIG. 4A.
The white circles are the original coordinate points and the black circles are the coordinate points obtained by the division processing. The group numbers of B, C, and E are B (02), C (02), and E (20), respectively, and in accordance with this, D is also written as D (22) instead of D (2) (punctuation is omitted. ). First, the center interpolation point c in FIG. 4C is B, C,
Since it is calculated from D and E, the group numbers of these four points are listed (02022220). When counting the numbers of 0 and 2 in this sequence and making them correspond one-to-one, two 2 are left. Therefore, let c (22). This is the group number of c. Since b of FIG. 4 (c) is calculated from B, D, and c, it becomes (022222) to b (22). Similarly,
The a (02), d (02) and e (22) are also clear. By performing the above processing for all points in FIG. 4A, FIG. 4D is obtained as grouping after mesh division.

When the grouping is completed, S308 in FIG.
~ Move to the altitude calculation process in S313. First, the altitude variance σ of all points having a certain group number is obtained (S308). Next, a random number that follows a normal distribution with a variance of σ is generated (S309). Such a random number sequence is obtained as the following equation (2) using, for example, an integer uniform random number prepared as a standard library of C language. However, in Expression (2), r _i is an integer uniform random number, A is its range, and n = 3 or 4.

[0019]

(Equation 2)

A new altitude is calculated by the following formula (3) by adding the rand obtained by the formula (2) to the temporary altitude data h _m set at the time of the division processing (S311). _{h = h m + rand ... (} 3) or more groupings and altitude calculation process, all groups (S307 → S315) all points (S310 →
This is performed for S313). As a result of this loop processing, the altitude calculation processing result at the last group number is applied to the vertices having a plurality of group numbers.

Next, in S316, the number of divisions is increased by one, and S
In 317, it is determined whether or not the number of divisions after incrementing by one has reached the specified number of divisions, and if not, S
Returning to 305, the processes of S305 to S315 are repeated.
If the number of divisions after incrementing by one has reached the designated number of divisions, the process ends in S318. The configuration in which the area expansion interpolation processing means is added to the interpolation calculation processing means of the first embodiment is
It will be described later in the third embodiment.

6 to 11 are views showing comparison of processing results according to the presence / absence of grouping in the first embodiment of the present invention, in which the interpolation points are displayed in a wire frame. In FIGS. 6 to 8, the grouping is not performed, and the altitude of the interpolation point is directly obtained from the average value and the dispersion value of the closest points. FIG. 6 shows real data of 4 × 4 mesh, and this real data is mesh-divided to obtain 8 × 8, 16
Interpolation with x16 is shown in FIGS. FIG. 9 is a diagram showing an example of grouping of the actual data shown in FIG.
When using the same actual data as above, the altitude variance σ in region 0 was 83.5 (the unit is arbitrary), and in region 1 it was 12.8. 10 and 11 are obtained as a diagram showing the data after the interpolation using the grouping of FIG. The lower side of FIG. 9 corresponds to the front side in FIGS. This FIG.
Comparing 1 with FIGS. 7 and 8, it can be seen that the relief pattern appears more clearly.

According to the terrain data interpolating apparatus according to the first embodiment of the present invention, as described above, when the interpolation value of the terrain data is obtained, the data of the closest contact points which are in a mutually orthogonal positional relationship are used, and When adding a random number to the altitude average value of the contact points, divide the entire region into regions with high correlation of altitude fluctuation, calculate a random number with a distribution equal to the altitude variance in each region, and include this within that region It was applied to the interpolation points. As a result, it is possible to obtain the interpolation data reflecting this while confirming the actual topographic relief.

Embodiment 2 FIG. FIG. 13 shows a second embodiment of the present invention.
4 is a flowchart of the interpolation processing according to FIG. 3, showing another altitude calculation processing method in which the grouping processing (S306) in the interpolation processing of FIG. 3 is not performed. In addition, the symbol {} in the drawing represents all the data sets of the coordinates [0] to [3]. FIG. 14 is a diagram for explaining the altitude calculation processing method in FIG.

When the interpolation processing is started in S401 of FIG. 13, the actual data {0} is read out in S402, and S403
The initial value 0 of the number of divisions is set with. When the interpolation coordinate {1} by the first mesh division processing is set in S404 (see (b) of FIG. 2), the altitude of the interpolation coordinate {1} is calculated in S405. In step S405, the average value h _m and the variance σ of the four [0] altitudes closest to a certain interpolation coordinate [1] are obtained. Next, a random number that follows a normal distribution and has a variance of σ is generated. This random number sequence is obtained as the following expression (4) using, for example, an integer uniform random number prepared as a standard library of C language. However, in equation (4), r _i is an integer uniform random number, A is its range, and n = 3 or 4.

[0026]

(Equation 3)

The rand obtained by the above equation (4) is added to the provisional altitude data h _m which is the altitude average value of the closest point of the interpolation point. In this case, it is shown in FIG. Such correction is performed according to the ratio of the distance between the interpolated coordinate [1] and four surrounding [0], and the altitude of [1] is calculated by the following equation (5).
Find more.

[0028]

(Equation 4)

In this way, the distribution of the original altitude is reflected in the newly obtained random number distribution, and correction is performed in consideration of the distance between the interpolation point and the nearest contact point, whereby the shape close to the actual topographical relief is obtained. Interpolation data can be obtained with. The above processing is performed for all [1].

When the interpolation coordinate {2} is set in S406, the altitude of the interpolation coordinate {2} is calculated in S407 as follows. . The average value h _m and the variance σ of the altitudes of [0] 2 and [1] 1 which are closest to a certain interpolation coordinate [2] are obtained. Hereafter, up to the part for obtaining rand, the above S4
It is the same as 05, but when finding the final altitude,
4 (b), correction is performed according to the ratio of the interpolated coordinate [2] and the distances [0] and [1] around it,
The altitude of [2] is calculated by the following equation (6).

[0031]

(Equation 5)

The above processing is performed for all [2].

When the interpolation coordinate {3} is set in S408, the altitude of the interpolation coordinate {3} is calculated in S409 as follows. An average value h _m and a variance σ of altitudes of [0] 2 and [1] 2 which are closest to a certain interpolation coordinate [3] are obtained. Hereafter, up to the part for obtaining rand, the above S40
It is the same as No. 5, but when finally obtaining the altitude,
Correction according to the ratio of the interpolated coordinate [3] and the distances of [0] and [1] around it as shown in (c) of
The altitude of [3] is calculated by the following equation (7).

[0034]

(Equation 6)

The above processing is performed for all [3].

According to the terrain data interpolating apparatus of the second embodiment of the present invention, as described above, when the interpolated value of the terrain data is obtained, the data of the closest contact points which are in a mutually orthogonal positional relationship are used, and When adding a random number to the altitude average value of the contact points, use a random number having a distribution equal to the dispersion of the altitude of the closest contact point, and correct the random number according to the ratio of the distance between the interpolation point and the closest contact point. Since it is applied, it becomes possible to calculate the interpolation data in a form close to the actual topographical relief.

Now, the above S405, S407 and S409
When the method described in 1) is applied to the internal rectangular area in FIG. 12 as it is, the connection at the boundary line with the adjacent area cannot be smoothly performed only by using the actual data [0] of each area. The reason is shown below using the examples of FIGS. 15 and 16. FIG. 15 is an explanatory diagram of a mesh division method in the vicinity of an internal rectangular area boundary according to the present invention, and FIG. 16 is an explanatory diagram of an area expansion method according to the present invention.

FIG. 15A shows ID0 and ID1 in FIG.
Represents the pattern near the boundary of the inner rectangular area,
The center line in the figure is the boundary line between ID0 and ID1, and corresponds to the area boundary line (thick line) in FIG. There are two methods of calculating the point sequence [2] on the boundary line: a method of calculating from the area side of ID0 and a method of calculating from the area side of ID1.
The arrows on both sides of the boundary indicate that the calculation can be performed from these two regions, respectively. FIG.
When the above equation (6) is applied to the point sequence [2] on the boundary line of (a) of, the provisional altitude h _{m obtained} from the ID0 side is [1] (0) of the ID0 region (where () is ID number) and [0], while h calculated from ID1
_m is calculated from [1] (1) and [0], and both are essentially inconsistent.

Therefore, in the third and fourth embodiments of the present invention, the interpolation processing of such topographical data is performed as shown in FIG.
As shown in (a) of, the boundary of each internal area is extended to an adjacent area (to be exact, an area sharing a rectangular vertex),
Interpolation processing is performed by regarding this as actual data (hereinafter, this area is referred to as an extended area). Here, the extension to the adjacent area may be performed by the required number of meshes.
In the example of, the case of expanding by one mesh is shown.

In FIG. 16A, an area A shown by a broken line represents an expanded area obtained by expanding ID4, and when interpolating ID4, a part of the data of ID0 to ID3 and ID5 to ID8 is added. Further, with respect to the internal region which is in contact with the periphery of the entire region, naturally the expansion in the peripheral direction is not performed. ID0 in (a) of FIG. 16 corresponds to this case, and the area after expansion becomes as indicated by the broken line B. Ie ID
Since there is no internal area adjacent to the upper and left sides of 0, ID
A part of the data of ID1, ID3, and ID4 is added to the 0 data. When area expansion is performed as described above, FIG. 15 (a) becomes as shown in FIG. 15 (b). In FIG. 15B, the boundary line after the region expansion from ID0 to ID1 is the broken line on the right side of the boundary line before the expansion, and conversely after the region expansion from ID1 to ID0. The boundary line of is the broken line on the left side of the boundary line before expansion. The boundary lines after expansion are shifted by 1 mesh.

When attention is paid to the thick broken rectangular area written as B in FIG. 16A, the right side of this rectangle is ID0 to I.
This is a boundary line of the expansion area after expansion in the D1 direction, which is the same as the broken line on the right side in FIG. Now, when obtaining the interpolation points around the area ID0,
If only the actual data on the periphery and in the same area is used, the actual data of the altitude variation outside the same area is missing, and thus the interpolation accuracy decreases accordingly. Therefore, the area is appropriately expanded by a few meshes and the actual data of that part is also taken in, and the interpolation processing is performed again for the expanded area.

At this time, FIG. 15B shows how interpolation is performed on the periphery of the original area ID0. In the figure, [2] is the interpolation point on the periphery, but it can be seen that the altitude is obtained from the average value of the data of the four surrounding points. However, in order to guarantee the connection on the boundary line, the altitude data is calculated by the following equation (8) by removing the random number term from the equation (7). h = h _m ... (8) to use the average value is not added here random number, in order to ensure that the same value even when seeking high [2] from ID1 side.

In order to perform interpolation using such an extended area, it is necessary to increase the terrain data memory.
For example, when the actual data consists of m × m meshes, the actual data + the number of expanded area meshes is (when expanded in all directions) (m + 2) × (m + 2), and the data size is D _{m + 2, m If +2} (unit is arbitrary) is written, the maximum terrain data memory size required for halving the mesh size by the interpolation process is D _{m + 2, m + 2} × 4.

Embodiments 3 and 4. FIG. 17 is a flowchart of interpolation processing including area expansion processing according to the third and fourth embodiments of the present invention. In addition, here, what applied the above-mentioned area | region expansion process to Embodiment 1 is called Embodiment 3, and what was applied to Embodiment 2 is called Embodiment 4. Further, it is assumed that the plurality of internal rectangular areas are given numbers from ID0 to ID (last). S501 in FIG.
When the interpolation process is started in step S502, the initial value 0 of the ID number is set in step S502. In S503, all I
The actual data [0] corresponding to each ID is read for D (the IDs in S502 to S508), and the number IDn of the adjacent internal area is obtained in S504.

Subsequently, in S505, the actual data {0} (IDn) of all the extension areas are read out. This is shown in FIG.
This can be easily realized by dividing and managing the data of the area that can be shared, such as a, b, c, d, e, f, g, and h in (b) of FIG. In the interpolation processing of S506, the same processing as the interpolation processing in the first embodiment of FIG. 3 or the second embodiment of FIG. 13 is performed, but the altitude data of the interpolation points on the periphery of the corresponding area is obtained by the equation (8). ) Use the altitude average value without adding random numbers.
In addition, unless the periphery of the internal region matches the periphery of the entire region, the setting of the interpolation coordinate [2] is unnecessary. . Then S50
In step 7, the ID number is incremented by 1, and in step S508, it is determined whether the ID number has become ID (last) +1.
509), and if not, steps S503 to S508 are repeated.

The interpolation processing of the third and fourth embodiments for carrying out the area expansion processing in FIG. 17 is summarized as follows.
That is, in the third embodiment, the terrain data is divided into rectangular areas having a high degree of correlation with altitude variation, and when interpolation is performed for each rectangular area, the corresponding area is in contact with the corresponding rectangular area or in the direction of all rectangular areas sharing a vertex. Area is expanded by a predetermined number of meshes, the interpolation calculation processing described in the first embodiment is performed, and the area expansion interpolation using the altitude average value of the nearest point of this interpolation point as the altitude data of the interpolation point on the periphery of the relevant area. Including processing. Further, in the fourth embodiment, when the terrain data is composed of an aggregate of a plurality of rectangular areas, when performing interpolation of each rectangular area, the directions of all the rectangular areas that are in contact with the corresponding rectangular area or share the apex are shared. After expanding the corresponding area by a predetermined number of meshes, the interpolation calculation process described in the second embodiment is performed, and the altitude average value of the closest point of this interpolation point is used as the altitude data of the interpolation point on the periphery of the area. Includes extended interpolation processing.

FIG. 18 is a diagram showing an interpolation processing result according to the fourth embodiment of the present invention, in which the interpolation result is displayed in a wire frame. FIG. 18A shows actual data composed of four internal rectangular areas of 2 × 2 mesh before interpolation.
The mesh of each area in FIG. 18A is divided into two and interpolated into a 4 × 4 mesh, and the entire area is combined and displayed in FIG. 18B.

According to the terrain data interpolating apparatus according to the third and fourth embodiments of the present invention, as described above, when the terrain data is an aggregate of a plurality of rectangular areas, the area to be interpolated is expanded and dealt with. By doing so, each area can be interpolated without losing the continuity between the areas.

Although the first to fourth embodiments have been described with respect to the interpolation of three-dimensional data, it is clear that the present invention can be applied to two-dimensional data by simply lowering the order. is there. Also, as the mesh division processing method, the method of dividing each side of the rectangle forming the mesh into two equal parts has been described, but it is also obvious that this can be changed into three equal parts, five equal parts and the like. Further, as the random numbers used in the altitude calculation, those which follow a geometric distribution, a triangular distribution, a chi-square distribution, a gamma distribution, a beta distribution, an F distribution, a binomial distribution, etc. can be used. Further, the coordinate system that corrects the correction value when performing the correction according to the distance between the interpolation point and the closest point can be extended to the non-orthogonal coordinate system.

[0050]

As described above, according to the present invention, in the terrain data interpolating device for dividing the terrain into the mesh structure and interpolating the given altitude data, the mesh of the altitude data is subjected to mesh division processing to determine the mesh size. The interpolation point coordinates to be divided into two are obtained, the nearest contact point having a positional relationship orthogonal to each of the interpolation point coordinates is selected, and then the entire area to be interpolated is divided into areas with high correlation of altitude fluctuation,
The altitude of the interpolation point is calculated by calculating a random number having a distribution equal to the altitude dispersion in each of the divided areas, and adding the value of this random number to the average height value of the closest points of the interpolation points in the corresponding area. In addition, an interpolation calculation means for repeating the previous upper altitude calculation processing for the re-interpolation points obtained by re-dividing the mesh size by the preset number of divisions is provided, so when calculating the interpolation value of the terrain data In addition to using the data of the closest contact points that are orthogonal to each other, when adding a random number to the altitude average value of this closest contact point, the entire region is divided into regions with high correlation of altitude fluctuations and the altitude variance in each region is calculated. By calculating a random number with equal distribution and applying it to the interpolation points included in that area, this can be done while checking the actual topographic relief. Interpolated data reflects it is possible to obtain.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration example of a terrain data drawing apparatus using a terrain data interpolation apparatus of the present invention.

FIG. 2 is a diagram illustrating a mesh division method according to the present invention.

FIG. 3 is a flowchart of all interpolation processing including mesh division processing according to the first embodiment of the present invention.

FIG. 4 is a diagram illustrating a grouping processing method according to the present invention.

FIG. 5 is a diagram showing a structure of topographical data according to the present invention.

FIG. 6 is a diagram showing actual data when grouping is not used.

FIG. 7 is a diagram showing data (8 × 8) after interpolation when grouping is not used.

FIG. 8 is a diagram showing data (16 × 16) after interpolation when grouping is not used.

9 is a diagram showing an example of grouping of the actual data of FIG.

FIG. 10 is a diagram showing data (8 × 8) after interpolation when grouping is used.

FIG. 11 is a diagram showing data (16 × 16) after interpolation when grouping is used.

FIG. 12 is a diagram showing a topographic map composed of a plurality of rectangular areas.

FIG. 13 is a flowchart according to the second embodiment of the present invention.

14 is a diagram illustrating an altitude calculation processing method in FIG.

FIG. 15 is an explanatory diagram of a mesh division method in the vicinity of an internal rectangular area boundary according to the present invention.

FIG. 16 is an explanatory diagram of a region expansion method according to the present invention.

FIG. 17 is a flowchart of interpolation processing according to the third and fourth embodiments of the present invention.

FIG. 18 is a diagram showing an interpolation processing result according to the fourth embodiment of the present invention.

FIG. 19 is an explanatory diagram of a conventional interpolation method disclosed in the patent publication.

[Explanation of symbols]

 101 display device 102 display control unit 103 terrain database 104 terrain data interpolation device 105 data bus 106 input / output unit 107 interpolation processing control unit 108 interpolation calculation processing unit 109 terrain data memory

Claims (4)

[Claims] 1. A terrain data interpolating device for interpolating altitude data provided by dividing a terrain into a mesh structure, and obtaining interpolation point coordinates for dividing a mesh of the altitude data into two mesh sizes by mesh division processing. Select the closest contact point that is in a positional relationship orthogonal to each interpolation point coordinate, then divide the entire interpolation target area into areas with high correlation of altitude fluctuation, and equalize the altitude distribution in each of these divided areas. A random number having a distribution is calculated, and the altitude of the interpolation point is calculated by adding the value of this random number to the altitude average value of the closest points of the interpolation point in the corresponding area,
Further, the terrain data interpolating device is provided with an interpolation calculation processing means for repeating the altitude calculation processing for the re-interpolation points obtained by re-dividing the mesh size by a preset number of divisions. 2. The terrain data is divided into rectangular areas having a high degree of correlation with altitude variation, and when the rectangular areas are interpolated, the corresponding areas are prescribed in the directions of all rectangular areas that are in contact with the corresponding rectangular areas or share vertices. After expanding by the number of meshes, claim 1.
The area expansion interpolation processing means for performing the interpolation calculation processing according to claim 1 and using, as altitude data of interpolation points on the periphery of the relevant area, an average height value of the nearest points of the interpolation points. Topographical data interpolating device according to 1. 3. A terrain data interpolating device for dividing terrain into a mesh structure to interpolate given altitude data, and obtains interpolation point coordinates for dividing the mesh of the altitude data into two mesh sizes by mesh division processing. The closest contact point having a positional relationship orthogonal to each interpolation point coordinate was selected, and the distribution was equal to the dispersion of the altitude of the closest contact point, and the correction was made according to the distance between the interpolation point and the closest contact point. A random number is calculated, and the altitude of the interpolation point is calculated by adding the value of this random number to the altitude average value of the closest points of the interpolation points in the corresponding area, and further, the mesh size is set by the preset number of divisions. A topographical data interpolating device comprising an interpolation calculation processing means for repeating the altitude calculation processing for a re-interpolation point obtained by re-dividing. 4. When the terrain data is composed of an aggregate of a plurality of rectangular areas, when performing interpolation of each rectangular area, the corresponding area is in contact with the corresponding rectangular area or in the direction of all rectangular areas sharing vertices. After expanding by a predetermined number of meshes,
The interpolation expansion processing according to claim 3 is performed, and area expansion interpolation processing means that uses an altitude average value of the nearest contact points of the interpolation points as altitude data of the interpolation points on the periphery of the relevant area is added. The terrain data interpolation device according to claim 3.