JP2003216023A - System for automatically generating numerical height model using topographical map information - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an automatic numerical height model generation system for automatically generating a numerical height model of high resolution by using topographical map information. <P>SOLUTION: The system 10 is provided with at least a topographical map recognition processing means 6 consisting of an image input means 6a, a contour extraction means 6b, an identification labeling means 6c, a noise information removing means 6d, and a connection restoring processing means 6e, a contour part height data generation means 7, an unknown point interpolation means 8 consisting of a falling-water line passage point detection means 8a and monotone function interpolation operation means 8b, and a shaping processing means 9. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital elevation model automatic generation system for extracting contour lines from a commercially available topographic map or a large-scale survey map and generating digital elevation information with high horizontal resolution.

[0002]

2. Description of the Related Art Conventionally, map information having contour lines covers a wide area with a 15,000 scale topographic map issued by the Geographical Survey Institute. However, even if the map data is read and three-dimensional data is created according to the elevation values of the contour lines, the data in the height direction depends on the distribution of the contour lines, and therefore, three-dimensional data having a step and approximating the actual topography can be obtained. There was no problem. In addition, although a "numerical map" issued by the Geospatial Information Authority of Japan is issued as digital data, the horizontal resolution is limited to about 50 m mesh, and there was a problem that a high-resolution numerical elevation model of several m mesh could not be obtained. .

For this reason, polynomial interpolation methods such as linear cubic spline interpolation, linear interpolation methods such as morphological interpolation, and one-dimensional interpolation methods have been proposed as methods for generating a digital elevation model from contour line data.
There are problems such as the occurrence of vibrations that cause inconvenient bumps and dents, that the topography near the contour line is not smooth, and that there are many linear noises running in eight directions. The production method was not established.

However, it is known that the above polynomial interpolation method has advantages of each method such that the terrain near the contour line is smoothly generated, and that the linear interpolation does not generate vibrations causing inconvenient bumps and dents. ing.

[0005]

The present invention has been made in view of the above problems, and combines the advantages of a polynomial interpolation method using three or more adjacent contour lines and a linear interpolation method using a monotone function. It is an object of the present invention to provide a digital elevation model automatic generation system using topographic map information for automatically generating a high-resolution digital elevation model, which is provided with an interpolating means.

[0006]

A digital elevation model automatic generation system using topographic map information of the present invention is a digital elevation model automatic generation system comprising a computer system equipped with a scanner for reading a topographic map as digital data. In the central processing control unit of the computer, topographic map recognition processing means for generating contour line vector data composed of point sequence data forming a contour line from topographic map information in which contour lines are drawn, and each of the contour line vector data. Contour line elevation data generating means for giving contour line elevation values from adjacency relations, and unknown point interpolating means for estimating the elevations of elevation unknown points other than the contour line portion from the contour line portion near the elevation unknown point to generate a digital elevation model, Shaping means for smoothing the terrain step of the generated digital elevation model, wherein the unknown point interpolation means is Substituted the coordinates and elevation values of the four closest points on the two contour lines sandwiching the unknown point for which the altitude value is desired to be sandwiched, and the closest points on the contour lines outside and adjacent to each closest point, and the coordinate value of the unknown point. The feature is that it is obtained by using a monotone function.

According to the present invention, it is possible to obtain elevation numerical information of a desired area with a desired resolution mesh by an easy operation of causing a scanner to read a commercially available topographical map covering all the terrain of the whole country. Therefore, by obtaining three-dimensional data that reproduces natural terrain, it is possible to use the data for a three-dimensional image such as a bird's-eye view that approximates the actual terrain and for civil engineering work planning data.

According to the unknown point interpolation means of the present invention,
By generating and using two monotonic functions from the coordinates / elevation values of four points on four different contour lines adjacent to an unknown altitude point, the generation of vibrations that cause inconvenient bumps and dents in the conventional method, near contour lines It is possible to eliminate the problem that a lot of linear noises running in eight directions are generated and the terrain is not smooth.

Further, it is possible to realize a high-speed processing which is located in the middle of the processing time of the high-speed polynomial interpolation method and the processing time of the linear interpolation method. Therefore, it is possible to automatically generate a digital elevation model in a practical processing time on an available commercial personal computer.

The invention according to claim 2 is the invention according to claim 1, wherein the unknown point interpolating means is the closest points P2, P3 and P2 on the two contour lines sandwiching the unknown point whose altitude value is to be obtained. the closest point P1 on adjacent contours Tensotogawa, P3 and nearest point P4 on adjacent contours and drainage line pass point detection means for obtaining a Tensotogawa, P1, P2, monotonic functions through P3 f _{1 (x)}
And the distance between the unknown point coordinates and the points P2 and P3 is proportionally calculated using the monotone function f ₂ (x) passing through P2, P3 and P4.
It is characterized by comprising a monotone function interpolation calculation means for obtaining the altitude by (x).

According to the present invention, two functions f1 (x) and f2 having four monotonicities, that is, a function passing through three points is upwardly convex or downwardly concave, and has an upward slope or a downward slope.
Since f (x) representing the elevation of the unknown point between points P2 and P3 is obtained from (x), the slope topography can be naturally reproduced. Further, in the case of the saddle part, the same effect can be obtained by setting the point P by always obtaining the intersection with the contour line in the direction in which the elevation value decreases.

A third aspect of the invention is the invention according to the second aspect, wherein the unknown point interpolating means has an unknown point whose elevation value is to be obtained in the vicinity of the summit and there is no contour line for obtaining the P1 point. , A monotonic function computing means that estimates a height value by setting a point having the maximum distance conversion value from the innermost contour line as a peak P1 point and uses f ₁ (x) as a unimodal function.

According to a fourth aspect of the present invention, the monotonic function f ₁ (x) and the monotonic function f ₂ (x) of claim 2 have a function form of Equation 1, and a monotone function for apportioning the distance between the P2 and P3. f
Formula (3) is used for (x), and Formula 2 is used for the functional form of f ₁ (x). According to the present invention, the monotone function operation means can be applied even near the top of a mountain.

The invention according to claim 5 is the invention according to claim 1, wherein the topographic map recognition processing means uses image input means for the topographic map in which contour lines are drawn, and color recognition is performed from the input image data. Contour line extraction means for extracting only contour line information, identification labeling means for thinning the extracted contour line information to identify it as point sequence data, and noise information for removing noise information including branches, branches and short-circuit lines in the image It is characterized in that it is provided with a removing means and a connecting and restoring processing means of a contour name of a topographic map, a numerical value of an altitude, and a broken contour line by displaying a topographical symbol.

According to the present invention, it is possible to obtain digital map data of contour lines from a commercially available 1 / 25,000 topographic map or a large-scale survey map, and to obtain digital map data of a finer mesh.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a schematic diagram showing the configuration of a digital elevation model automatic generation system 10 of the present invention.

In FIG. 1, reference numeral 1 is a scanner for reading a topographic map in which contour lines are drawn. This scanner uses color information that can be identified as digital information in order to identify contour lines printed in brown on a commercial topographic map. Reference numeral 2 is an image display device (display), and 3 is an input device such as a keyboard and a mouse.

In the digital elevation model automatic generation system 10 of the present invention, the database 4 and the central processing control device 5 are provided with at least the following data tables and processing means.

The database 4 includes an original topographic map digital data table 4a for recording the digital image information of the topographic map read by the scanner 1, and a contour line extraction digital data table 4b for recording contour lines extracted from the original data.
And a contour line numerical elevation table 4c in which the elevation value is added to the point sequence coordinate information of the contour line, a high-definition interpolated numerical elevation model data table 4d in which the elevation value is interpolated and recorded at the coordinates between the contour lines, and an interpolated elevation And at least the shaped digital elevation model data table 4e after shaping the unnatural terrain.

The central processing control device 5 comprises a topographic map recognition processing means 6, a contour line elevation data generating means 7, an unknown point interpolation means 8, and a shaping processing means 9.

The topographic map recognition processing means 6 is an image input means 6a, a contour line extraction means 6b, an identification labeling means 6c, a noise information removal means 6d, and a connection restoration processing means 6.
e and.

The image input means 6a is a means for reading the topographic map of the target original topography using the scanner 1 and recording it in the original topographic map digital data table 4a of the database 4.

FIG. 2 shows the result of processing by the topographic map recognition processing means. FIG. 2A is an example in which a commercially available topographic map around the top of Mt. Iwate of 1 / 25,000 is read. As shown in the figure, in addition to the contour lines, this topographic image information includes place names, elevation values, topographic symbols, and the like.

The contour line extracting means 6b is means for extracting only the contour line portion from the original topographic map digital data table 4a and recording it in the contour line extraction digital data table 4b. This is done by identifying the contour lines printed in brown in the case of a topographic map issued by the Geographical Survey Institute of Japan.

FIG. 2B shows an embodiment in which contour lines are extracted from the topographic image information shown in FIG. The contour lines for every 100 m are bolded. Also, the contour lines are cut off in the printed parts such as place names, elevation values, and topographic symbols.

The identification labeling means 6c is a means for labeling and thinning contour lines. The noise information processing means 6d is means for removing information other than contour lines such as short line segments, branches and branch points. FIG. 2C is the result of processing FIG. 2B.

The connection restoration processing means 6e is means for connecting contour lines of the same elevation at the places where the contour lines are disconnected. As this technique, a well-known technique of the extended Voronoi diagram can be used. Figure 2 (d)
2 is an example of a Voronoi diagram in which connection restoration processing is performed on FIG.

FIG. 3 is an embodiment showing the result of the contour line extraction digital data table 4b of the processing result by the topographic map recognition processing means 6. At this time, the contour line information of the contour line extraction digital data table 4b is recorded as point sequence information of one pixel.

The contour line elevation data generation means 7 reads out contour line point sequence information from the contour line extraction digital data table 4b, and converts it into point sequence coordinate data of all the contour lines based on the elevation value given to the reference contour line from the outside. It is means for giving an elevation value and recording it in the contour line numerical elevation table 4c.

Next, a means for obtaining an altitude unknown point X between contour lines will be described. In addition, the setting of the altitude unknown point X is
It is set depending on how many meshes the accuracy (resolution) of the digital elevation model is. In the following description, the interpolation process performed for each one of the obtained altitude unknown points X (points of interest) will be described.

The unknown point interpolating means 8 is composed of a drop line passing point detecting means 8a and a monotone function interpolation calculating means 8b, and calculates an altitude value z for the coordinates (x, y) of the altitude unknown point X. To do.

The falling line passing point detecting means 8a finds the closest points P2 and P3 on two contour lines sandwiching the altitude unknown point X. Next, it is means for obtaining the closest point P1 on the adjacent contour line further outside from the point P2 and the closest point P4 on the adjacent contour line further outside from the point P3. This corresponds to approximately obtaining the flow line (falling line) of precipitation passing through the altitude unknown point X.

FIG. 4 is an explanatory diagram of the falling line passing point detecting means of the present invention. P1 to P4 indicate the intersections of the falling line and the contour line. FIG. 4A shows a sloped portion, FIG. 4B shows a saddle portion, FIG. 4C shows the vicinity of the summit, and FIG. 4D shows the summit.

As shown in FIG. 4 (a), first, the closest points P2 and P3 on the contour lines T2 and T3 that sandwich the altitude unknown point X are found, and the closest point from the point P2 on the contour line T1 outside the point P2. P1 is obtained, and the closest point P4 from the point P3 on the contour line T4 outside the point P3 is obtained.

In the saddle topography shown in FIG. 4 (b), the intersections with the contour lines are always obtained in the direction of decreasing elevation values P1 to P4.
To set.

In the topography near the summit shown in FIG. 4C, the peak position of the peak is set to P1. For the peak position, the point with the maximum distance conversion value (arbitrary one when there are multiple points) is determined from the innermost contour line, and the elevation value is obtained by the elevation value estimation method (method such as cubic curve fitting) to obtain P1. Coordinates and elevation.

In the topography of the mountain top shown in FIG. 4D, P
Only 1, P2 and P3 are set and P4 is omitted.

The a, b, and c are determined from the coordinates of the points P1, P2, P3, and P4 on the contour line obtained by the drop line passing point detection means 8a and the altitude value, and a monotone function f passing through P1, P2, and P3
₁ (x) and a monotone function f passing through P2, P3, and P4
₂ (x) is generated. Where f ₁ (x), f ₂ (x)
The following function 1 is used for the function form of

[Formula 1]

Next, using the altitude unknown point X, the distance α to the P2 point and the distance β to the P3 point, the altitude value of the unknown point X is proportionally calculated by the function f (x) of the following mathematical formula 2. .

[Formula 2]

FIG. 5 is a schematic diagram showing a calculation formula of the monotone function interpolation calculating means 8b. FIG. 5A shows a monotone function f ₁ (x) that passes through P1, P2, and P3, and an f ₂ that passes through P2, P3, and P4, where the altitude unknown point X of the inclined portion is obtained.
It is a graph of (x). It indicates that the altitude of the altitude unknown point X is calculated as a value obtained by apportioning the value of f ₁ (x) and the value of f ₂ (x) at the distances α and β.

When P1 is obtained at the peak position near the peak shown in FIG. 4 (c), the functional form of f1 (x) is given by the following mathematical expression 3.

[Formula 3]

FIG. 5B is a schematic diagram showing a calculation formula of the topography near the summit.

The above means is repeated and the result is recorded in the high-definition interpolated numerical elevation model data table 4d as three-dimensional information.

Next, problems of the generated high-definition interpolated numerical elevation model data table 4d will be described. The arrow in the topographic map shown in FIG. 6 indicates the ridge direction, and the result of the monotone function interpolation of this topographical section is shown in FIG. 7 (a). In the monotone function interpolation of the present invention, since the elevation values are calculated independently of each other on both sides of the ridge, a step is likely to occur.

The arrows in the topographical map shown in FIG. 6 indicate the direction of the cross section that obliquely cuts the sloping ground. The result of the monotone function interpolation of this topographical section is shown in FIG. In the monotone function interpolation of the present invention, adjacent waterfall lines are calculated independently of each other, and thus a small step is likely to occur in a direction orthogonal to (or largely intersecting with) the waterfall line.

The shaping processing means 9 reads the high-definition interpolated numerical elevation model data table 4d obtained for smoothing the above-described stepped data, and smoothes it using a known regularization method, This is means for recording in the shaped digital elevation model data table 4e.

FIG. 7 (b) shows the shaping processing result of FIG. 7 (a), and FIG. 7 (d) shows the shaping processing result of FIG. 7 (c). In the shaping process, it is desirable to thin out and designate pixels to be processed on the contour line in order to improve the shaping effect near the contour line.

[0049]

According to the digital elevation model automatic generation system using the topographic map information of the present invention, the following effects are exhibited.

It is possible to obtain elevation numerical information of a desired area with a desired resolution mesh by an easy operation of reading a commercially available topographic map covering all the terrain of the whole country with a scanner. Therefore, by obtaining three-dimensional data that reproduces natural terrain, it is possible to use the data for a three-dimensional image such as a bird's-eye view that approximates the actual terrain and for civil engineering work planning data.

According to the unknown point interpolation means of the present invention,
By using two monotone functions that substitute the coordinates and elevation values of four points on four different contour lines adjacent to the unknown elevation point, the generation of vibrations that cause inconvenient bumps and dents in the conventional method, near the contour line It is possible to eliminate the problem that a lot of linear noises running in eight directions are generated, in which the terrain is not smooth.

Furthermore, it is possible to realize a high-speed processing which is located in the middle of the processing times of the high-speed polynomial interpolation method and the linear interpolation method. Therefore, it is possible to automatically generate a digital elevation model in a practical processing time on an available commercial personal computer.

[Brief description of drawings]

FIG. 1 is a system for automatically generating a digital elevation model 10 according to the present invention.
It is a schematic diagram which shows the structure of.

FIG. 2 is an explanatory diagram showing a process of topographic map recognition processing of the present invention, (a) is a topographic map reading process, (b) is a contour line extracting process, (c) is a labeling / thinning process, and (d) is a process. FIG. 7 is a diagram showing a Voronoi diagram processing process.

FIG. 3 is a diagram showing contour line data as a result of topographic map recognition processing according to the present invention.

FIG. 4 is an explanatory view of a falling line passing point detecting means of the present invention. (A) shows an inclined part, (b) shows a saddle part, (c) shows a mountain top vicinity, (d) shows a mountain top part.

FIG. 5 is a schematic diagram showing a graph of a functional expression of the monotone function interpolation calculation means of the present invention. (A) shows an inclined part and (b) shows the vicinity of the summit.

FIG. 6 shows a topographic map. It is explanatory drawing which shows the direction of the topographical cross section shown in FIG.

FIG. 7 is an explanatory diagram showing a result of the monotone function interpolation of the terrain cross section in the direction shown in FIG. 6 and a result after the shaping process. (A),
(C) shows a monotone function interpolation result, and (b) and (d) show a shaping process result.

[Explanation of symbols]

1 Scanner 2 Image Display Device (Display) 3 Input Device (Keyboard, Mouse) 4 Database 4a Original Topographic Map Digital Data Table 4b Contour Line Extraction Digital Data Table 4c Contour Line Elevation Table 4d High Definition Interpolation Elevation Model Data Table 4e After Shaping Digital elevation model data table 5 Central processing control device 6 Topographic map recognition processing means 6a Image input means 6b Contour line extraction means 6c Identification labeling means 6d Noise information removal means 6e Connection restoration processing means 7 Contour line elevation data generation means 8 Unknown point interpolation means 8a Falling-water passing point detecting means 8b Monotonic function interpolation calculating means 9 Shaping processing means 10 Numerical elevation model automatic generation system P1, P2, P3, P4 Coordinates of intersection points X between falling water lines and contour lines Contour lines between height unknown points T2, T3 X Outside T1, T4 T2, T3 Side contour lines

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] January 24, 2002 (2002.1.2
4)

[Procedure Amendment 1]

[Document name to be amended] Statement

[Correction target item name] Full text

[Correction method] Change

[Correction content]

[Document name] Statement

[Title of Invention] Digital elevation model automatic generation system using topographic map information

[Claims]

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital elevation model automatic generation system for extracting contour lines from a commercially available topographic map or a large-scale survey map and generating digital elevation information with high horizontal resolution.

[0002]

2. Description of the Related Art Conventionally, map information having contour lines covers a wide area with a 15,000 scale topographic map issued by the Geographical Survey Institute. However, even if the map data is read and three-dimensional data is created according to the elevation values of the contour lines, the data in the height direction depends on the distribution of the contour lines, and therefore, three-dimensional data having a step and approximating the actual topography can be obtained. There was no problem. In addition, although a "numerical map" issued by the Geospatial Information Authority of Japan is issued as digital data, the horizontal resolution is limited to about 50 m mesh, and there was a problem that a high-resolution numerical elevation model of several m mesh could not be obtained. .

For this reason, polynomial interpolation methods such as linear cubic spline interpolation, linear interpolation methods such as morphological interpolation, and one-dimensional interpolation methods have been proposed as methods for generating a digital elevation model from contour line data.
There are problems such as the occurrence of vibrations that cause inconvenient bumps and dents, that the topography near the contour line is not smooth, and that there are many linear noises running in eight directions. The production method was not established.

However, it is known that the above polynomial interpolation method has advantages of each method such that the terrain near the contour line is smoothly generated, and that the linear interpolation does not generate vibrations causing inconvenient bumps and dents. ing.

[0005]

The present invention has been made in view of the above problems, and combines the advantages of a polynomial interpolation method using three or more adjacent contour lines and a linear interpolation method using a monotone function. It is an object of the present invention to provide a digital elevation model automatic generation system using topographic map information for automatically generating a high-resolution digital elevation model, which is provided with an interpolating means.

[0006]

A digital elevation model automatic generation system using topographic map information of the present invention is a digital elevation model automatic generation system comprising a computer system equipped with a scanner for reading a topographic map as digital data. In the central processing control unit of the computer, topographic map recognition processing means for generating contour line vector data composed of point sequence data forming a contour line from topographic map information in which contour lines are drawn, and each of the contour line vector data. Contour line elevation data generating means for giving contour line elevation values from adjacency relations, and unknown point interpolating means for estimating the elevations of elevation unknown points other than the contour line portion from the contour line portion near the elevation unknown point to generate a digital elevation model, Shaping means for smoothing the terrain step of the generated digital elevation model, wherein the unknown point interpolation means is Substituted the coordinates and elevation values of the four closest points on the two contour lines sandwiching the unknown point for which the altitude value is desired to be sandwiched, and the closest points on the contour lines outside and adjacent to each closest point, and the coordinate value of the unknown point. The feature is that it is obtained by using a monotone function.

According to the present invention, it is possible to obtain elevation numerical information of a desired area with a desired resolution mesh by an easy operation of causing a scanner to read a commercially available topographical map covering all the terrain of the whole country. Therefore, by obtaining three-dimensional data that reproduces natural terrain, it is possible to use the data for a three-dimensional image such as a bird's-eye view that approximates the actual terrain and for civil engineering work planning data.

According to the unknown point interpolation means of the present invention,
By generating and using two monotonic functions from the coordinates / elevation values of four points on four different contour lines adjacent to an unknown altitude point, the generation of vibrations that cause inconvenient bumps and dents in the conventional method, near contour lines It is possible to eliminate the problem that a lot of linear noises running in eight directions are generated and the terrain is not smooth.

Further, it is possible to realize a high-speed processing which is located in the middle of the processing time of the high-speed polynomial interpolation method and the processing time of the linear interpolation method. Therefore, it is possible to automatically generate a digital elevation model in a practical processing time on an available commercial personal computer.

The invention according to claim 2 is the invention according to claim 1, wherein the unknown point interpolating means is the closest points P2, P3 and P2 on the two contour lines sandwiching the unknown point whose altitude value is to be obtained. the closest point P1 on adjacent contours Tensotogawa, P3 and nearest point P4 on adjacent contours and drainage line pass point detection means for obtaining a Tensotogawa, P1, P2, monotonic functions through P3 f _{1 (x)}
And the distance between the unknown point coordinates and the points P2 and P3 is proportionally calculated using the monotone function f ₂ (x) passing through P2, P3 and P4.
It is characterized by comprising a monotone function interpolation calculation means for obtaining the altitude by (x).

According to the present invention, a function passing through three points is two functions f ₁ (x) and f _{1 which} are convex or concave downward and have four monotonicities of upward rising or rightward bending.
_{Since 2} (x) is used to obtain f (x) that represents the altitude of the unknown point between the points P2 and P3, the slope topography can be naturally reproduced. Further, in the case of the saddle part, the same effect can be obtained by setting the point P by always obtaining the intersection with the contour line in the direction in which the elevation value decreases.

A third aspect of the invention is the invention according to the second aspect, wherein the unknown point interpolating means has an unknown point whose elevation value is to be obtained in the vicinity of the summit and there is no contour line for obtaining the P1 point. , A monotonic function computing means that estimates a height value by setting a point having the maximum distance conversion value from the innermost contour line as a peak P1 point and uses f ₁ (x) as a unimodal function.

According to a fourth aspect of the present invention, the monotonic function f ₁ (x) and the monotonic function f ₂ (x) of claim 2 have a function form of Equation 1, and a monotone function for apportioning the distance between the P2 and P3. f
Formula (3) is used for (x), and Formula 2 is used for the functional form of f ₁ (x). According to the present invention, the monotone function operation means can be applied even near the top of a mountain.

The invention according to claim 5 is the invention according to claim 1, wherein the topographic map recognition processing means uses image input means for the topographic map in which contour lines are drawn, and color recognition is performed from the input image data. Contour line extraction means for extracting only contour line information, identification labeling means for thinning the extracted contour line information to identify it as point sequence data, and noise information for removing noise information including branches, branches and short-circuit lines in the image It is characterized in that it is provided with a removing means and a connecting and restoring processing means of a contour name of a topographic map, a numerical value of an altitude, and a broken contour line by displaying a topographical symbol.

According to the present invention, it is possible to obtain digital map data of contour lines from a commercially available 1 / 25,000 topographic map or a large-scale survey map, and to obtain digital map data of a finer mesh.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a schematic diagram showing the configuration of a digital elevation model automatic generation system 10 of the present invention.

In FIG. 1, reference numeral 1 is a scanner for reading a topographic map in which contour lines are drawn. This scanner uses color information that can be identified as digital information in order to identify contour lines printed in brown on a commercial topographic map. Reference numeral 2 is an image display device (display), and 3 is an input device such as a keyboard and a mouse.

In the digital elevation model automatic generation system 10 of the present invention, the database 4 and the central processing control device 5 are provided with at least the following data tables and processing means.

The database 4 includes an original topographic map digital data table 4a for recording the digital image information of the topographic map read by the scanner 1, and a contour line extraction digital data table 4b for recording contour lines extracted from the original data.
And a contour line numerical elevation table 4c in which the elevation value is added to the point sequence coordinate information of the contour line, a high-definition interpolated numerical elevation model data table 4d in which the elevation value is interpolated and recorded at the coordinates between the contour lines, and an interpolated elevation And at least the shaped digital elevation model data table 4e after shaping the unnatural terrain.

The central processing control device 5 comprises a topographic map recognition processing means 6, a contour line elevation data generating means 7, an unknown point interpolation means 8, and a shaping processing means 9.

The topographic map recognition processing means 6 is an image input means 6a, a contour line extraction means 6b, an identification labeling means 6c, a noise information removal means 6d, and a connection restoration processing means 6.
e and.

The image input means 6a is a means for reading the topographic map of the target original topography using the scanner 1 and recording it in the original topographic map digital data table 4a of the database 4.

FIG. 2 shows the result of processing by the topographic map recognition processing means. FIG. 2A is an example in which a commercially available topographic map around the top of Mt. Iwate of 1 / 25,000 is read. As shown in the figure, in addition to the contour lines, this topographic image information includes place names, elevation values, topographic symbols, and the like.

The contour line extracting means 6b is means for extracting only the contour line portion from the original topographic map digital data table 4a and recording it in the contour line extraction digital data table 4b. This is done by identifying the contour lines printed in brown in the case of a topographic map issued by the Geographical Survey Institute of Japan.

FIG. 2B shows an embodiment in which contour lines are extracted from the topographic image information shown in FIG. The contour lines for every 100 m are bolded. Also, the contour lines are cut off in the printed parts such as place names, elevation values, and topographic symbols.

The identification labeling means 6c is a means for labeling and thinning contour lines. The noise information processing means 6d is means for removing information other than contour lines such as short line segments, branches and branch points. FIG. 2C is the result of processing FIG. 2B.

The connection restoration processing means 6e is means for connecting contour lines of the same elevation at the places where the contour lines are disconnected. As this technique, a well-known technique of the extended Voronoi diagram can be used. Figure 2 (d)
2 is an example of a Voronoi diagram in which connection restoration processing is performed on FIG.

FIG. 3 is an embodiment showing the result of the contour line extraction digital data table 4b of the processing result by the topographic map recognition processing means 6. At this time, the contour line information of the contour line extraction digital data table 4b is recorded as point sequence information of one pixel.

The contour line elevation data generation means 7 reads out contour line point sequence information from the contour line extraction digital data table 4b, and converts it into point sequence coordinate data of all the contour lines based on the elevation value given to the reference contour line from the outside. It is means for giving an elevation value and recording it in the contour line numerical elevation table 4c.

Next, a means for obtaining an altitude unknown point X between contour lines will be described. In addition, the setting of the altitude unknown point X is
It is set depending on how many meshes the accuracy (resolution) of the digital elevation model is. In the following description, the interpolation process performed for each one of the obtained altitude unknown points X (points of interest) will be described.

The unknown point interpolating means 8 is composed of a drop line passing point detecting means 8a and a monotone function interpolation calculating means 8b, and calculates an altitude value z for the coordinates (x, y) of the altitude unknown point X. To do.

The falling line passing point detecting means 8a finds the closest points P2 and P3 on two contour lines sandwiching the altitude unknown point X. Next, it is means for obtaining the closest point P1 on the adjacent contour line further outside from the point P2 and the closest point P4 on the adjacent contour line further outside from the point P3. This corresponds to approximately obtaining the flow line (falling line) of precipitation passing through the altitude unknown point X.

FIG. 4 is an explanatory diagram of the falling line passing point detecting means of the present invention. P1 to P4 indicate the intersections of the falling line and the contour line. FIG. 4A shows a sloped portion, FIG. 4B shows a saddle portion, FIG. 4C shows the vicinity of the summit, and FIG. 4D shows the summit.

As shown in FIG. 4 (a), first, the closest points P2 and P3 on the contour lines T2 and T3 that sandwich the altitude unknown point X are found, and the closest point from the point P2 on the contour line T1 outside the point P2. P1 is obtained, and the closest point P4 from the point P3 on the contour line T4 outside the point P3 is obtained.

In the saddle topography shown in FIG. 4 (b), the intersections with the contour lines are always obtained in the direction of decreasing elevation values P1 to P4.
To set.

In the topography near the summit shown in FIG. 4C, the peak position of the peak is set to P1. For the peak position, the point with the maximum distance conversion value (arbitrary one when there are multiple points) is determined from the innermost contour line, and the elevation value is obtained by the elevation value estimation method (method such as cubic curve fitting) to obtain P1. Coordinates and elevation.

In the topography of the mountain top shown in FIG. 4D, P
Only 1, P2 and P3 are set and P4 is omitted.

The a, b, and c are determined from the coordinates of the points P1, P2, P3, and P4 on the contour line obtained by the drop line passing point detection means 8a and the altitude value, and a monotone function f passing through P1, P2, and P3
₁ (x) and a monotone function f passing through P2, P3, and P4
₂ (x) is generated. Where f ₁ (x), f ₂ (x)
The following function 1 is used for the function form of

[Formula 1]

Next, using the altitude unknown point X, the distance α to the P2 point and the distance β to the P3 point, the altitude value of the unknown point X is proportionally calculated by the function f (x) of the following mathematical formula 2. .

[Formula 2]

FIG. 5 is a schematic diagram showing a calculation formula of the monotone function interpolation calculating means 8b. FIG. 5A shows a monotone function f ₁ (x) that passes through P1, P2, and P3, and an f ₂ that passes through P2, P3, and P4, where the altitude unknown point X of the inclined portion is obtained.
It is a graph of (x). It indicates that the altitude of the altitude unknown point X is calculated as a value obtained by apportioning the value of f ₁ (x) and the value of f ₂ (x) at the distances α and β.

When P1 is determined at the peak position near the peak shown in FIG. 4 (c), the functional form of f ₁ (x) is given by the following mathematical expression 3.

[Formula 3]

FIG. 5B is a schematic diagram showing a calculation formula of the topography near the summit.

The above means is repeated and the result is recorded in the high-definition interpolated numerical elevation model data table 4d as three-dimensional information.

Next, problems of the generated high-definition interpolated numerical elevation model data table 4d will be described. The arrow in the topographic map shown in FIG. 6 indicates the ridge direction, and the result of the monotone function interpolation of this topographical section is shown in FIG. 7 (a). In the monotone function interpolation of the present invention, since the elevation values are calculated independently of each other on both sides of the ridge, a step is likely to occur.

The arrows in the topographical map shown in FIG. 6 indicate the direction of the cross section that obliquely cuts the sloping ground. The result of the monotone function interpolation of this topographical section is shown in FIG. In the monotone function interpolation of the present invention, adjacent waterfall lines are calculated independently of each other, and thus a small step is likely to occur in the direction orthogonal to (or largely intersecting) the waterfall line.

The shaping processing means 9 reads the high-definition interpolated numerical elevation model data table 4d obtained for smoothing the above-described stepped data, and smoothes it using a known regularization method, This is means for recording in the shaped digital elevation model data table 4e.

FIG. 7 (b) shows the shaping processing result of FIG. 7 (a), and FIG. 7 (d) shows the shaping processing result of FIG. 7 (c). In the shaping process, it is desirable to thin out and designate pixels to be processed on the contour line in order to improve the shaping effect near the contour line.

[0049]

According to the digital elevation model automatic generation system using the topographic map information of the present invention, the following effects are exhibited.

It is possible to obtain elevation numerical information of a desired area with a desired resolution mesh by an easy operation of reading a commercially available topographic map covering all the terrain of the whole country with a scanner. Therefore, by obtaining three-dimensional data that reproduces natural terrain, it is possible to use the data for a three-dimensional image such as a bird's-eye view that approximates the actual terrain and for civil engineering work planning data.

According to the unknown point interpolation means of the present invention,
By using two monotone functions that substitute the coordinates and elevation values of four points on four different contour lines adjacent to the unknown elevation point, the generation of vibrations that cause inconvenient bumps and dents in the conventional method, near the contour line It is possible to eliminate the problem that a lot of linear noises running in eight directions are generated, in which the terrain is not smooth.

Furthermore, it is possible to realize a high-speed processing which is located in the middle of the processing times of the high-speed polynomial interpolation method and the linear interpolation method. Therefore, it is possible to automatically generate a digital elevation model in a practical processing time on an available commercial personal computer.

[Brief description of drawings]

FIG. 1 is a system for automatically generating a digital elevation model 10 according to the present invention.
It is a schematic diagram which shows the structure of.

FIG. 2 is an explanatory diagram showing a process of topographic map recognition processing of the present invention, (a) is a topographic map reading process, (b) is a contour line extracting process, (c) is a labeling / thinning process, and (d) is a process. FIG. 7 is a diagram showing a Voronoi diagram processing process.

FIG. 3 is a diagram showing contour line data as a result of topographic map recognition processing according to the present invention.

FIG. 4 is an explanatory view of a falling line passing point detecting means of the present invention. (A) shows an inclined part, (b) shows a saddle part, (c) shows a mountain top vicinity, (d) shows a mountain top part.

FIG. 5 is a schematic diagram showing a graph of a functional expression of the monotone function interpolation calculation means of the present invention. (A) shows an inclined part and (b) shows the vicinity of the summit.

FIG. 6 shows a topographic map. It is explanatory drawing which shows the direction of the topographical cross section shown in FIG.

FIG. 7 is an explanatory diagram showing a result of the monotone function interpolation of the terrain cross section in the direction shown in FIG. 6 and a result after the shaping process. (A),
(C) shows a monotone function interpolation result, and (b) and (d) show a shaping process result.

[Explanation of Codes] 1 Scanner 2 Image display device (display) 3 Input device (keyboard, mouse) 4 Database 4a Original topographic map digital data table 4b Contour line extraction digital data table 4c Contour line elevation table 4d High-definition interpolated numeric elevation model data Table 4e Digital elevation model data table after shaping 5 Central processing control device 6 Topographic map recognition processing means 6a Image input means 6b Contour line extraction means 6c Identification labeling means 6d Noise information removal means 6e Connection restoration processing means 7 Contour line elevation data generation means 8 Unknown point interpolating means 8a Falling line passing point detecting means 8b Monotone function interpolating calculating means 9 Shaping processing means 10 Numerical elevation model automatic generation system P1, P2, P3, P4 Coordinates X of falling water line and contour line X Height unknown point T2 Contour lines T1 and T4 that sandwich T3 X Contours outside T2 and T3

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 2C032 HB05                 5B050 AA01 BA17 DA06 EA04 EA13                       EA28                 5B057 AA13 BA02 CA01 CA12 CA16                       CB02 CB06 CB13 CB17 CC04                       CF01 DA08 DA17 DB02 DB06                       DC02 DC14 DC25 DC36                 5L096 AA02 BA12 CA14 DA02 EA04                       EA22 FA03 FA15 FA73 GA34

Claims (5)

[Claims] 1. A digital elevation model automatic generation system comprising a computer system equipped with a scanner for reading a topographic map as digital data, wherein the central processing control unit of the computer uses topographic map information in which contour lines are drawn. Topographic map recognition processing means for generating contour line vector data consisting of point sequence data forming contour lines, contour line portion altitude data generating means for giving contour line elevation values from the respective adjacency relations of the contour line vector data, and other than the contour line portion. An unknown point interpolating means for estimating the altitude of an unknown altitude point from the contour line portion in the vicinity of the unknown altitude point, and a shaping processing means for smoothing the terrain step of the generated numerical elevation model, The unknown point interpolating means interposes unknown points for which elevation values are to be obtained 2
The feature is that it is obtained by using a monotone function in which the coordinates / elevation values of four points of the closest point on the contour line of the book and the closest point on the outer adjacent contour line of each closest point and the coordinate value of the unknown point are substituted. Digital elevation model automatic generation system using active topographic map information. 2. The unknown point interpolating means includes two closest points P2 and P3 on two contour lines sandwiching an unknown point whose altitude value is to be obtained.
Falling-water passing point detection means for obtaining the closest contact point P1 on the adjacent contour line outside the P2 point and the closest contact point P4 on the adjacent contour line outside the P3 point, and a monotone function f passing through P1, P2, and P3.
₁ (x) and a monotone function f passing through P2, P3, and P4
_2. The terrain according to claim 1, characterized by comprising a monotone function interpolation calculation means for obtaining an altitude by f (x) which proportionally calculates the distance between the unknown point coordinates and P2, P3 points using ₂ (x). Digital elevation model automatic generation system using figure information. 3. The unknown point interpolating means, when the unknown point whose altitude value is to be obtained is near the top of the mountain and there is no contour line for obtaining the P1 point, finds the point whose distance conversion value from the innermost contour line is the largest. 3. A digital elevation model automatic generation system using topographic map information according to claim 2, further comprising: a monotone function computing means for estimating an elevation value with P1 point and f ₁ (x) as a single peak function. . 4. The monotone function f ₁ (x) according to claim 2.
And the function form of the monotone function f ₂ (x) is Equation 1, P2, P
3. A monotone function f (x) for apportioning the distance from 3 is used in Equation 3, and a functional form of f ₁ (x) in claim 3 is in Equation 2. 3. Digital elevation model automatic generation system using the described topographic map information. 5. The topographic map recognition processing means, a topographic map image input means in which contour lines are drawn, a contour line extraction means for extracting only contour line information from the input image data by color recognition, and the extracted contour lines. Identification labeling means for identifying information as point sequence data by thinning the information, noise information removal means for removing noise information including branches, branches and short-circuit lines in the image, topographic map place name, elevation numerical value, topographic symbol display 3. A digital elevation model automatic generation system using topographic map information according to claim 1, further comprising a connection restoration processing means for disconnection contour lines according to.