JP3298972B2 - Figure processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a graphic processing method used in a mapping system that uses map data in a database.

[0002]

2. Description of the Related Art Generally, in a mapping system, it is desired to use a map created using a national reference point as a map database.

However, modern maps created based on reference points other than the national reference points are based on highly accurate surveying methods, and are partially more accurate than maps created using the national reference points. In some cases. For example, Kawasaki City, which pioneered the map, did not use the national reference point,
We have created a large scale map with a scale of 1/500 on our own,
In this map, the width of roads, boundary piles, and the like are described more accurately than a map based on a national reference point (for example, a map of 1/2500 scale of the national land).

[0004]

For this reason, it is conceivable to incorporate such partially high-accuracy maps into a mapping system. However, a map created based on reference points other than the national reference points is referred to as a map database. If you want to
Often the administrative boundaries are not consistent with other maps.

The present invention has been made in view of such a problem, and an object of the present invention is to provide a graphic processing method capable of matching one map with another map with high accuracy and efficiency. Is to do.

[0006]

FIG. 1 is a diagram illustrating the principle of the present invention. In FIG. 1, 1 is a host computer, 3
Is a map database, 5 is an application program, 7 is a work area, 9 is a display, 11 is a digitizer, and 13 is a pointing tool.

The host computer 1 performs various processes according to the application program 5. The map database 3 stores, for example, map data of a large scale with a scale of 1/500. The display 9 displays the large scale map data stored in the work area 7, for example, at a scale of 1/500. The digitizer 11 has, for example, a scale 1
/ 2500 medium scale map is pasted. This map includes an area described in the map displayed on the display 9.
The map data can be input by the pointing tool 13 in the digitizer 11. When the pointer 13 is moved, the cursor moves on the display 9 as the pointer 13 moves, and a point or line segment on the display 9 can be designated by clicking the pointer 13.

A is a flowchart showing the contents of the application program 5 and the operation of the operator.
First, a large scale map in the work area 7 is displayed on the display 9.
It is displayed above (step 21). The operator moves the pointing device 13 while watching the display 9 to display the display 9.
The cursor is positioned on, for example, a point G, and a line segment g passing through G is designated using the pointing tool 13 (step 22). Next, the operator looks at the medium-scale map affixed to the digitizer 11 and specifies the point F on the line segment g1 corresponding to the line segment g on the display 9 with the pointing tool 13 (step 2).
3). When a point F is designated by the digitizer 11, it is displayed on the display 9 as a point F '. The point F 'does not exist on the line segment g because there is a slight shift between the corresponding portions of the medium-scale map and the large-scale map, and the two do not match.

The application program 5 calculates the distance and angle between the point F 'corresponding to the point F on the display 9 and the line segment g (step 24). The operator repeats the operations from step 22 to step 24 to differentiate the points G and F, and designates a plurality of sets of the line segment g and the point F (step 25). When a plurality of sets of the line segment g and the point F are designated in this way, the application program 5 calculates affine transformation parameters using the plurality of sets of distances and angles calculated in step 24 (step 26).
Then, the large-scale map data stored in the work area 7 is affine-transformed using the affine transformation parameters (step 27). The affine-transformed map data is stored in the map database 3.

[0010]

According to the present invention, even when the large-scale map displayed on the display 9 and the corresponding portion of the medium-scale map pasted on the digitizer 11 do not match, a plurality of sets of line segments and points By specifying, the affine transformation of the large-scale map data in the work area 7 can be performed so that the large-scale map is converted to match the medium-scale map.

[0011]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 2 is a diagram showing a configuration of an apparatus used in the graphic processing method according to one embodiment of the present invention. 1 is a host computer, 3 is a map database, 5 is an application program, 9 is a display, 11 is a digitizer, 13 is a pointing tool, 15 is a keyboard, and 17 is a printer.

FIGS. 3 and 4 are flow charts showing the processing of this embodiment. First, the operator converts a large-scale map (for example, a map of Kawasaki city at a scale of 1/500) into a digitizer 11.
, And input the map data such as the block line on the map using the indicating tool 13 and store it in the work area 7.

FIG. 5 shows a large-scale map attached to the digitizer 11. In FIG. 5, part E1 is described with the current road width, and is more accurately described in part E1 than in the basic national land map.

When this map is input by the digitizer 11, first, A1 is input as a reference point, and four corners A1, B
The user designates 1, C1, and D1 with the pointing device 13, inputs the coordinate values thereof, and calculates parameters for correcting the map by, for example, a method described in JP-A-1-320579. Then, data such as a block line is input using the indicating tool 13. Further, the input data is multiplied by the reciprocal of the scale (in this case, 500) to convert the data into local dimensions, corrected by the above-described parameters, and stored in the work area 7. Then, the map is displayed on the display 9 (step 301).

Next, the operator pastes a medium-scale map (scale map of 1/2500 national land) onto the digitizer 11 and inputs four corners A2, B2, C2, and D2 of the map, and the image is displayed on the display 9. The coordinate value of the point A at the lower left corner of the area corresponding to the area is input (step 302).

The processing in step 302 will be described in detail. FIG. 6 shows a medium-scale map attached to the digitizer 11. In this medium-scale map, ABCD is a portion planned for a large-scale map displayed on the display 9.

FIG. 7 is an enlarged view of the area ABCD of FIG. As shown in this figure, since the road width is narrow in the section E, it is drawn with a predetermined value, for example, 1 mm width.

FIG. 8 is an explanatory diagram of the input in step 302. First, four corners A2, B2, C2,
This coordinate value is input by designating D2. Then A2 of coordinates (X _A, Y _A) to enter. Then by specifying a point A by the pointing device 13, inputs the coordinates (l _x, l _y) of the point A. Here X _A, Y _A is the absolute coordinates. Also, l _x , l
_y is the coordinates of point A on the map of FIG. 6 when A2 is the origin. Thus, the coordinate value of the point A is input.

Next, the parameter i representing the number of sets is cleared (step 303), and the point G on the display 9 is input by the pointing device 13 (step 304).

The processing of step 304 will be described.
FIG. 9 shows the current display content of the display 9 and the display content assuming that the area ABCD on the digitizer 11 is displayed on the display 9 as it is, in a superimposed manner. In FIG. 9, a solid line indicates actual display contents of the display 9, and a broken line indicates display contents on the assumption that the drawing on the digitizer 11 is displayed as it is.

The operator operates the indicating tool 13 while looking at the display 9. The cursor moves on the display 9 in conjunction with the movement of the pointing tool 13. When the operator moves the cursor to a position near point G in FIG.

When the point G is designated in this way, the host computer 1 determines that the line segment g closest to the designated point is
(In the present embodiment, the point G exists on the line segment g), for example, is displayed in red (step 305). FIG. 10 shows the display contents of the display 9 at this time, and the line segment g is displayed in red.

Next, the operator looks at the digitizer 11 and inputs a point F on the line segment corresponding to the line segment g1 on the medium-scale map attached to the digitizer 11 using the pointing tool 13 (step 306). ).

After inputting the point F, the following conversion is performed.

[0025] _{X = bx + X A Y =} by + Y A coordinate where x, y are on the digitizer 11 at point F, b represents the reciprocal of the scale.

As shown in FIG. 9, the large-scale map and the medium-scale map have slight deviations even in the corresponding portions. When the point F is input on the medium-scale map attached to the digitizer 11, the point F 'is input on the display 9 as shown in FIG.

Next, the host computer 1 calculates the distance L _i and the angle θ _i between the line segment g and the point F ′ and stores them together with the coordinates (X _i , Y _i ) of the point F ′ (step 307).

The calculation of the distance L _i and the angle θ _i is performed as follows. FIG. 11 is a diagram showing a line segment g and a point F ′. The coordinates of the point F 'are now (X ₁ , Y ₁ ), and the coordinates (X ₂ , Y ₂ ) and (X ₃ , Y ₃ ) of any two points on the line segment g.
Lower the perpendicular line to the straight line g from the point F', distance the length of the vertical line L _i, the inclination relative to the horizontal axis of the perpendicular line and angle theta _i.

Assuming that a = Y ₂ −Y ₃ and b = X ₃ −X ₂ , θ _i = sgn (b / a) · tan ⁻¹ | b / a | (where −π / 2 <θ _i <π / 2. When a = 0, θ _i =
π / 2). Here, sgn (b / a) indicates that the vector direction of the line segment g is positive when the vector direction is in the second and fourth quadrants, and negative when the vector direction is in the first and third quadrants.

On the other hand, the distance L _i to seek the _{c = (X 3 -X 2)} · Y 2 + (Y 2 -Y 3) · X 2 to the _{L i = sgn (a) ·} (aX 1 + bY 1 −c) / (a ²
+ B ² ) ^1/2 .

Next, the parameter i is incremented by "1" (step 308), and the processing from step 304 to step 308 is repeated until the operator instructs termination (step 309). That set of different G and the point F is a plurality of sets specified, for each set, the distance L _i and the angle theta _i is calculated.

The number of sets at this time is substituted into N (step 310), and affine transformation parameters are calculated (step 311).

FIG. 12 is an explanatory diagram of calculation of affine transformation parameters. And _{l i = L i -x i cosθ} i -y i sinθ i , as shown in FIG. 12, consider the following equation of the affine transformation for x _i, y _i. In FIG. 12, for simplicity, the vector l
_{Although i} , x _i , and y _i are shown as being translated, the vector L _i and the vector l _i actually overlap.

X_i= AX_i+ BY_i+ C y_i= DX_i+ EY_i+ F As a result l_i= L_i-Cosθ_i(AX_i+ BY_i+ C) -si
nθ_i(DX_i+ EY _i+ F). This l_iThe condition for minimizing the sum of squares of

[0035]

(Equation 1)

Is as follows. The calculation result is a simultaneous equation in which the non-constant parameters are A, B, C, D, E, and F, and is represented by a matrix as follows.

[0037] _{[K hj] [A, B} , C, D, E, F, l] = 0 [K hj] is _{K 11, K 12, ... K} 17, ... K 61, ... and so K ₆₇ 42 It becomes a matrix of number of coefficients, for example, shows a K ₁₇ from K ₁₁ in this is as follows.

[0038]

(Equation 2)

To solve this simultaneous equation, for example, “F
If a scientific calculation subroutine library by ORTRAN 77 ”(Kigaku Publishing, pp. 84-86) is prepared in a program, the calculation can be performed by the host computer 1. In the affine transformation parameters, A is the expansion and contraction ratio in the X direction, (B + D) / 2 is the flattening ratio, C is the movement amount in the X direction, E is the expansion and contraction ratio in the Y direction, and (BD) / 2 is Rotation amount,
F relates to the amount of movement in the Y direction.

Next, the data in the work area 7 is converted using the affine conversion parameters obtained in step 311 (step 312).

That is, assuming that the coordinates of the data in the work area 7 are X and Y and the coordinates converted using the affine transformation parameters are X ^* and Y ^* , X ^* = X + (AX + BY + C) Y ^* = Y + (DX + EY + F) ). This coordinate conversion is performed on map data in the work area 7.

Then, the converted data is stored in the map database 3 (step 313). In the present embodiment, the correction is performed while a large-scale map is input. However, the present invention is also applicable to a case where the large-scale map has already been input and is stored in a database. As described above, in this embodiment, a large-scale map (for example, a map of Kawasaki city with a scale of 1/500) having high accuracy is partially matched with a medium-scale map (for example, a basic map of Japan) using affine transformation. Can be converted as follows. A map converted using the affine transformation has high accuracy in part, and thus is suitable as a map database.

[Second Embodiment] Next, a second embodiment of the present invention will be described with reference to the flow charts shown in FIGS. In the second embodiment, corner cutting of an intersection and the like are performed.
Despite the fact that the block has changed over time, if the operator inputs and specifies the block line related to change over time without noticing that fact, the accuracy of the affine transformation parameters may deteriorate, so The affine transformation parameters are obtained by excluding possible data.

That is, in step 311, the host computer 1 performs the following processing.

That is, the stored coordinates X _i , Y _i ,
All the pairs of the distance L _i and the angle θ _i are extracted (step 1301).

Next, l _i = L _i -cos θ _i (AX _i + BY _i + C) -si
In accordance with nθ _i (DX _i + EY _i + F), distances l _i after affine transformation are obtained for the number of inputs (step 1302). Then, the calculated standard deviation S of each l _i is calculated (step 1303).

Next, assuming that the threshold value is α (for example, 2 or 3), all l _i satisfying | l _i |> Sα are deleted assuming that they correspond to the aged line data (step 1304).

Then, assuming that the number of erased l _i is n, N−
Let n be a regular input number N (step 1305), and based on l _i of the regular input number N, obtain affine transformation parameters A, B, C, D and F by the least squares method (step 1306). .

FIG. 14 is a flowchart showing a process for semi-automatically excluding such data. Since this processing has many parts in common with the processing shown in FIG. 13, only the differences will be briefly described.

After calculating the standard deviation S of each l _i , these l _i have changed over time for all l _i satisfying | l _i |> Sα with a threshold value α (eg, 2 or 3). A flag is set assuming that it corresponds to the block line data (step 1304a). Then, for each pair, a vector having a length l _i and an angle θ _i starting from the coordinates X _i and Y _i is displayed on the display 9 (step 1304).
b). At this time, l _i is a small value, and it is difficult to see the vector, so that it is enlarged and displayed several tens of times, and the flagged flag is likely to be related to the aging line data. To indicate to the operator that something is, it is displayed in a different color from the others.

Therefore, the operator specifies a vector to be excluded (step 1304c). Next, as in the case of FIG. 13, affine transformation parameters A, B, C, D, E, and F are obtained by the least squares method based on the remaining l _i (steps 1305 and 1306).

As described above, the inappropriate data is automatically or semi-automatically excluded and the affine transformation parameters are determined, thereby improving the reliability of the transformation result without increasing the burden on the operator. it can.

[Third Embodiment] Next, a third embodiment will be described with reference to FIG.
This will be described according to the flowchart of FIG. The purpose of the third embodiment is as follows. For example, recently, if the road is re-measured in an area where sewerage work has been performed over a large area, and the map data has been partially but extensively corrected, there will be a large difference between the corrected and uncorrected areas. The error between the block line of the small-scale map and the block line of the medium-scale map may be largely different. In such a case, unless the affine transformation parameters are determined for each area, the accuracy of the affine transformation parameters deteriorates. For this reason, in the third embodiment, affine transformation parameters are determined for each area.

That is, in the third embodiment, after the step 310 in FIG. 4, the host computer 1 extracts all the stored sets of the coordinates X _i , Y _i , distance L _i , and angle θ _i (step 1601). .

Next, l _i = L _i -cos θ _i (AX _i + BY _i + C) -si
In accordance with nθ _i (DX _i + EY _i + F), distances l _i after affine transformation are obtained for the number of inputs (step 1602). Then, for each pair, the coordinates X _i ,
A vector having a length l _i and an angle θ _i starting from Y _i is enlarged several tens times and displayed on the display 9 (step 1).
603).

Then, the operator inputs the range of the vector having the same tendency while watching the distribution state of the vector to divide the area on the display screen (step 160).
4). Then, the host computer 1 reads out all l _i for one sectioned area, and determines affine transformation parameters for the area by the least squares method (step 1605). Then, affine transformation is performed on all coordinate data of the area according to the determined affine transformation parameters (step 1606). Next, the same processing as steps 1605 and 1606 is performed on the remaining areas (step 1607). Finally, the operator performs joint editing on the coordinate data of the boundary between the areas (step 1608), and ends the processing.

[Fourth Embodiment] Next, a fourth embodiment will be described. In the fourth embodiment, the type of block line, for example, national road,
Affine transformation parameters are determined by weighting according to city roads, ward roads, city roads, private roads, sidewalks, and the like. For example, assuming that one block line is designated for each of the national road, the metropolitan road, the ward road, the city road, the private road, and the sidewalk, these are respectively 6, 5, 4, 3, 2, and 1 each. Determine affine transformation parameters as specified. In this manner, the affine transformation is performed with higher accuracy for the coordinate data along the block line having higher importance.

[Fifth Embodiment] Next, a fifth embodiment will be described. In the fifth embodiment, instead of obtaining the distance and angle between the line segment g and the point F ′, the coordinate position data is divided into an X component and a Y component between the point G and the point F ′. The processing of FIGS. 3 and 4 for determining the parameters for the affine transformation based on the distance and the angle between the line data is executed exactly in the same manner.

That is, the coordinate position data of the designated point G on the large scale map and the point F 'on the medium scale map are (X ₁ , Y ₁ ) and (X ₂ , Y ₂ ) as shown in FIG. For example, in step 304 of FIG. 3, it is considered that the X axis and the Y axis are specified, and in the (i + 1) -th step 307, the distance _{Li + 1 is set} to (X ₁ −X ₂ ),
Step 30 in the (i + 2) -th time is set to 0 degree for _{i + 1.}
In 7 the distance _{L i + 2 (Y 1 -Y} 2), an angle θ _{i + 2} 90
It is processed as a degree.

Sixth Embodiment Next, a sixth embodiment will be described. In the sixth embodiment, when the four corners of the map are input in step 302 of FIG. 3, the coordinate values are converted using affine transformation parameters, and the data is converted to other medium-scale map data (basic land map). ) And keep it in a safe place.
The coordinates and a straight line connected by the coordinates are displayed.

FIG. 17 is a diagram showing the contents displayed on the display 9 at this time. In FIG. 17, the two-dot chain line is a straight line connecting the four corners of the map of the medium-scale map. The two-dot chain lines protruding from the display 9 are stored in correspondence with other meshes.

That is, A1 ', B1', C1 ', D1
′ Are (x _A , y _A ), (x _B , y _B ), (x _C ,
Assuming that y _C ) and (x _D , y _D ), each coordinate value is affine-transformed as in the following equation and stored together with the converted coordinate data and the like.

[0063] _{X = x A + (Ax A} + By A + C) Y = y A + (Dx A + EY A + F) X = x B + (Ax B + By B + C) Y = y B + (Dx B + EY B + F _{) X = x C + (Ax} C + By C + C) Y = y C + (Dx C + EY C + F) X = x D + (Ax D + By D + C) Y = y D + (Dx D + EY D + F)

[0064]

As described in detail above, according to the present invention, one map can be matched with another map with high accuracy and efficiency.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is an apparatus configuration diagram for executing a graphic processing method according to an embodiment of the present invention;

FIG. 3 is a flowchart showing the operation of the embodiment.

FIG. 4 is a flowchart showing the operation of the embodiment.

FIG. 5 is a diagram showing a large-scale map displayed on a display 9;

FIG. 6 is a diagram showing a medium-scale map pasted on the digitizer 11;

FIG. 7 is an enlarged view of rectangles A, B, C, and D in FIG. 6;

FIG. 8 is a diagram illustrating input of rectangles A, B, C, and D in FIG. 6;

FIG. 9 shows the display contents of the display 9 and the digitizer 11.
Explanatory drawing when it is assumed that the corresponding portion of the above drawing is displayed on the display 9 as it is.

FIG. 10 is a diagram showing display contents of the display 9 in which a line segment g is highlighted in step 305.

FIG. 11 is a diagram showing a positional relationship between a line segment g and a point F;

FIG. 12 is a diagram illustrating an affine transformation

FIG. 13 is a flowchart illustrating an operation when automatically determining a determinant of an affine transformation parameter;

FIG. 14 is a flowchart showing an operation when semi-automatically excluding determinants of affine transformation parameters;

FIG. 15 is a flowchart showing an operation when affine transformation parameters are determined for each area.

FIG. 16 is a view for explaining processing when a reference point on each map is designated;

FIG. 17 is a diagram showing a map of a corresponding portion of a large scale on a medium scale map.

[Explanation of symbols]

 1 Host computer 3 Map database 5 Application program 7 Work area 9 Display 11 Digitizer 15 Keyboard 17 Printer

Claims (8)

(57) [Claims] 1. A system comprising a digitizer, a display, and a storage device, comprising: (a) inputting a first map from the digitizer and storing the first map in the storage device; and (b) the storage device. Displaying the first map stored on the display on the display; (c) designating a first line segment on the display; and (d) pasting on the digitizer, Specifying a point on a second line segment corresponding to the first line segment on a second map having a portion corresponding to the first map; and (e) specifying a point on the second line segment corresponding to the point. 1 reference point on the map,
(F) repeating steps (c) to (e) to calculate a distance and an angle to the first line segment;
Calculating an affine transformation parameter using a distance and an angle between the line segment and the reference point; and (g) using the affine transformation parameter to convert the first map data stored in the storage device. And performing an affine transformation. 2. A system comprising a digitizer, a display, and a storage device, comprising: (a) inputting a first map from the digitizer and storing the first map in the storage device; and (b) the storage device. Displaying the first map stored on the display on the display; (c) designating a first point on the display; and (d) pasting on the digitizer, Specifying a second point corresponding to the first point on a second map having a portion corresponding to the first map; and (e) diagonally connecting the first point and the second point. Calculating the length and inclination of each of the long side and the short side of the rectangle as a distance and an angle, and (f) repeating steps (c) to (e) to obtain a plurality of sets of the first set.
Calculating an affine transformation parameter using a distance and an angle between the point and the second point; and (g) converting the first map data stored in the storage device into the affine transformation parameter. Using the affine transformation. 3. A system comprising a digitizer, a display, and a storage device, comprising: (a) inputting a first map from the digitizer and storing the first map in the storage device; and (b) the storage device. Displaying the first map stored on the display on the display; (c) designating a first line segment or a third point on the display; and (d) pasting on the digitizer. A point on a second line segment corresponding to the first line segment on a second map having a portion corresponding to the first map,
Designating a fourth point corresponding to the point of (e), a reference point on the first map corresponding to the point,
A distance and an angle with the first line segment are calculated, or a length and a slope of a long side and a short side of a rectangle having the third point and the fourth point as diagonal lines are defined as a distance and an angle, respectively. (F) repeating steps (c) to (e) to calculate a plurality of first sets
Distance and angle between the line segment and the reference point, or the third point and the fourth point
Calculating an affine transformation parameter using a distance and an angle to the point; and (g) affine transforming the first map data stored in the storage device using the affine transformation parameter. And a graphic processing method comprising: 4. In the step (e), a length of a perpendicular drawn from the reference point to the first line segment is set as a distance, and an inclination of the perpendicular with respect to a horizontal axis is set as an angle. The graphic processing method according to claim 1. 5. The affine transformation parameter is calculated in step (f) by excluding a set of distances and angles whose distances deviate greatly from the standard deviation. The graphic processing method according to any one of the above. 6. The method according to claim 1, wherein the step (f) calculates a parameter for affine transformation by weighting a designated line type. Described graphics processing method. 7. Inputting the four corners of the map of the first map, affine-transforming the coordinate values using the affine-transformation parameters calculated in step (f); 4. The method according to claim 1, further comprising: a step of holding the coordinate values, and a step of displaying a straight line connected by the coordinate values together with other map data when performing drawing. Described graphics processing method. 8. The first map is a large-scale map created based on reference points other than a national reference point, and the second map is a medium-scale map created based on a national reference point. The graphic processing method according to claim 1, wherein: