JPH06309469A - Method for plotting outline character - Google Patents

Abstract

PURPOSE:To quickly execute the painting-out of a rotated outline character. CONSTITUTION:The internal area of a character is divided into plural polygonal areas and plotting processing is executed by means of intermediate font data obtained by integrating these polygons. The shape of each polygon is selected so as to have two entering vertexes or less at the time of observing it from an optional direction. The coordinate transformation of the intermediate font data at the time of plotting includes the operation of parallel movement, magnification/reduction and rotation. This transformation is executed to deform a scquare surrounding the character font to a parallelogram. Although each polygon is also deformed, the property having two entering vertexes or less at the time of observing it from an optional direction is maintained. Since the entering vertexes of each polygon is restricted only to one or two also in the y-axis direction vertcial to scanning lines on the display screen, painting- out operation can quickly be executed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an outline character drawing method for drawing a character on a display, a printer or the like using an outline character font.

[0002]

2. Description of the Related Art An outline character font expresses the shape of a character by an outline or outline.
It is possible to display a character by enlarging, reducing and rotating it at an arbitrary magnification using the same font data. To actually draw a character with an outline character font, the vertex coordinates that define the contour shape are transformed on the display surface and then the area surrounded by the contour line is filled. Coordinate transformations usually consist of a combination of translation, stretching and rotation. The process of filling the pixels belonging to the character area on the display surface using the contour line data is called scan conversion or raster conversion. In the following description, the entire character display process including the reading of font data and coordinate conversion may be referred to as scan conversion. This type of scan conversion technique is introduced, for example, in Japanese Patent Application No. 4-205988.

The conventional scan conversion algorithm has two processing methods disclosed in the above document. Hereinafter, these processing methods will be described. FIG. 2 is an explanatory diagram of the outline character font drawing method according to the first conventional method. FIG. 2 shows a drawing process of a character of "love" by a conventional method. This "love" character uses a total of 123 vertices 7
Each polygon is used as a contour line. down the y coordinate axis, x
If the coordinate axis is directed to the right, the scanning line y = ymin passing through the top vertex +1 of the contour line to the scanning line y = ymax passing through the bottom vertex.
Find all intersections with the character contour for each scan line up to
The line segment corresponding to the internal area of the character is filled from left to right.

In FIG. 2, the state in which the character "love" is filled up to the scanning line y = y is shown by the hatching of the horizontal line. 3 and 4 are flowcharts of scan conversion according to the first conventional method. These figures show a flowchart for drawing a plurality of characters (character strings) by this method. This flowchart has the following triple loop structure. The outermost loop shows the repeated processing of drawing one character. The loop indicates the repeating processing of each scanning line from the top vertex to the bottom vertex when drawing one character. The loop traces the intersection of the scan line and the contour line on one scan line from left to right,
This represents a drawing process in which line segment drawing is repeated.

As the horizontal scanning line moves downward from above in FIG. 2, the number of intersections with the contour line increases or decreases in units of two. The apex that increases the number of intersections by two is the entry vertex. The vertex that reduces the number of intersections by 2 is called the exit vertex. These vertices, when advanced along the contour line,
The y-coordinate is a vertex having a minimum value and a maximum value. +1 in the figure
The 14 vertices indicated by ˜ + 14 are the entry vertices of the contour line of the character “love”. The other loop shown in FIG. 3 represents the entry processing executed on the scan line where the entry vertex exists. The entry vertices are sorted in ascending order of y coordinate,
In this flowchart, the approach vertices are detected using the difference between the y coordinates of the approach vertices adjacent to each other (the approach difference). An active edge list (hereinafter referred to as AEL) is used to efficiently execute the above-described entry processing and drawing processing (steps S101 to S107).

FIG. 5 shows an active edge list according to the conventional method. AEL is a data structure in which ridges (called active edges) that intersect the current scanning line are sorted in ascending order of x coordinates. The ridge line (edge) is a chain of line segments from the entry vertex to the exit vertex, or one of the line segments. The line segment data consists of an x coordinate, a height h, and an inclination Δx.
In the approach processing of step S108 of FIG. 3, the left edge line and the right edge line emerging from the entry vertex are inserted into the AEL position such that the x coordinate is in ascending order. In the drawing process of the loop, the active edge pointer (AEP) is AE
Traversing on L from left to right, the active edge pointed to by AEP is called in step S118, and step S119
Determines whether it is the left edge line or the right edge line.

Since passing the left edge line indicates entering from the outside to the inside of the character area, step S1 is performed in advance.
The integer variable w set to 0 in 15 is counted up in step S120, and the right ridge line indicates that the character region goes out from the inside of the character area. Therefore, w is counted down in step S123. It is assumed that w is 0 in the outer area of the character and 1 or more in the inner area, and is not negative. In the processing procedure of FIG. 4, the area from the left ridgeline that changes the value of w from 0 to 1 to the right ridgeline that changes from 1 to 0, that is, the range in which w is not 0 is filled. That is, the fill start point x1 is set in step S122, the fill end point xr is set in step S125, and pixels from x1 to xr are collectively filled.

In step S126, the x coordinate of the ridge is updated (increment Δx is added), the height h is counted down (1 is subtracted), and the next scanning line is updated. Then, when h = 0 in step S127, the active edge is updated to the next ridge line in step S128, and x and h are reset. Here, when the exit vertex is reached, the active edge is removed from the AEL. When the processing for all active edges included in the AEL is completed,
Exit the loop from step S116 to step S11
Proceed to 7 to enter the next scan line. As described above, the first conventional method is characterized in that the AEL is used to repeat the entry process and the filling of a plurality of line segments for each scanning line as needed. Next, the second conventional method will be described with reference to FIG. 6 by exemplifying a case where the characters of "love" are similarly painted.

In the second conventional method, the internal area of a character is set to 1
Divide into 0 polygons, and make the logical sum of the internal regions of these polygons the internal region of the entire character, so that the polygon alone has only one entry vertex when viewed in the y-axis direction. To do. Then, the intermediate font data in which the left and right edge data of the polygon are accumulated is created in advance, and when the character is drawn, the polygon-divided intermediate font data is read out, and coordinate conversion consisting of only movement and expansion / contraction without rotation is performed. The coordinate conversion is performed individually for each of the divided polygons. Figure 6
Indicates a state in which four polygons whose entry vertices are +1 to +4 are filled.

FIG. 6 shows a flow chart of scan conversion by the second conventional method. Is a character loop, is a polygon loop, and is a scan line loop. The second conventional method is AE
Since L is not used, the processing related to this in the first conventional method is unnecessary, and the drawing processing is also simplified, so that it is faster than the first conventional method. First, step S in FIG.
16 interprets the command for displaying the character string and prepares for execution. In step S17, the intermediate font data for one character is read in, and in step S18, all the vertex coordinates are subjected to coordinate conversion to the display surface. Up to this point, the procedure is the same as in the first conventional method, but in the first conventional method of FIG.
3 is followed by a vertex and edge classification process S104 and an entry vertex list and left and right edge line list creation process S105. However, in the second conventional example of FIG. It is unnecessary.

The creation of the right and left ridge line data in the next step S19 is the same as that of the first conventional method, but the drawing processing thereafter includes the processing relating to AEL and AEP in the first conventional method at all in the second conventional method. Not at all. This is because only one drawing line segment is generated at each scanning line from the entry vertex to the exit vertex of the polygonal shape obtained by the division. When entering the polygon loop, in step S20, the first edge data of the left and right edge list of the divided polygon is taken out. Here, the x-coordinate xr and the height of the starting point of the left edge data are set to hr and hr, respectively. In step S21, the y coordinate y1 of the entry vertex is set to y. When entering the scan line loop, in steps S22, S23, and S24, (xl, y)
Pixels up to (xr, y) are filled.

In step S25, the x coordinate and height of the left edge line are updated to the next scanning line, and the y coordinate is updated to the next scanning line in step S32. When the height hl or hr becomes 0, the next ridge line data is taken out in step S28 or S31, and the x coordinate and the height h are reset. When the exit vertex is reached, it is found in step S27 that the next left edge line does not exist, so the processing of that polygon is terminated. Since the inside of the polygon is filled in as described above, this may be executed for all the polygons in order. Areas shared by two or more polygons are naturally filled. Further, it is necessary to paint two polygons that are in contact with each other on a vertical line segment so that no space is left on the vertical line segment. When drawing of one character is completed in this way, the process returns to step S17 to read the polygonal division intermediate font pointed to by the next character code, and coordinate conversion is performed so as to change the display position in step S18, so that the next character can be similarly drawn. The character string drawing is completed by repeating this for the specified number of characters.

[0013]

However, the above-mentioned conventional technique has the following problems. First, in the first conventional method, it is necessary to execute the following many processing items each time one character is drawn or every scanning line. a) coordinate conversion (movement, expansion / contraction, rotation) b) creation of left and right edge line data c) entry vertex sort d) AEL entry process e) AEL sort f) AEL update exit process g) Drawing process using AEL

Of these, the processes concerning the entry vertex sort, AEL entry, AEL sort, and AEL update exit of c), d), e), and f) are directly necessary to obtain the coordinate value of the pixel to be drawn. It is not an extra calculation, but an additional process for efficiently supplying the data necessary for the drawing process using the AEL of g). These additional processes have a large number of intersections with the scanning lines because the outline of the outline character has a complicated shape, and are required to perform efficient processes corresponding to these. However, there is a problem that most of the scan conversion time is spent on these incidental processes.

On the other hand, the second conventional method cannot be applied to the case of rotating and displaying characters. For example, a polygon having the +4 vertex as the entry vertex in FIG. 6 has one entry vertex and one exit vertex when viewed in the y-axis direction (considering only the y coordinate value), but viewed in the x-axis direction. And (when the character is rotated 90 degrees and viewed in the y-axis direction), four pairs of entry and exit vertices are generated, and even if only one polygon is drawn, the processing using the AEL is required. Therefore, when the character is rotated, the processing time becomes long even when the character is divided into polygons. The present invention has been made by paying attention to the above points, and is a high-speed outline character drawing capable of coordinate conversion including rotation without generating an active edge list during scan conversion of outline characters. It is intended to provide a method.

[0016]

In the outline character drawing method of the first invention of the present invention, a contour line of a character is formed by a set of a plurality of line segments, and a plurality of internal areas surrounded by the plurality of line segments are formed. In the case of division into polygons, the logical sum of the internal areas of the polygons is the entire internal area of the character, and the shapes of all the polygons are related to coordinate axes set in arbitrary directions. Select each of the polygons so that the number of entry vertices at which the position coordinates are the minimum is less than or equal to a predetermined number,
The line segment data of each polygon is prepared in advance as the intermediate font data of the character, each polygon which is the intermediate font data is read out, the coordinates are converted, and the intersection of each coordinate-converted polygon and the scanning line is read. Is recognized and the space between the intersections is filled.

The outline character drawing method of the second invention is
When the contour line of a character is formed by a plurality of line segments and the contour lines themselves do not intersect, all the entry vertices where the y coordinate of the vertex on the contour line with respect to the reference coordinate y-axis becomes a minimum,
y so that the y coordinates are in ascending order, and if the y coordinates are equal, y
Obtain an entry vertex list ordered such that the x-coordinates of the reference coordinate x-axis rotated 90 degrees counterclockwise are in descending order, and the inner region of the character is left from each of the entry vertices along the contour line. In the direction to look in and the direction to the right, the left edge line and the right edge line are obtained as a series of line segments reaching the exit vertex where the y-coordinate becomes maximum, and the left edge line and the left edge line starting from the unused entry vertex in the entry vertex list When the area sandwiched between the right ridgelines is filled in scan line units and the scan line reaches the entrance apex, and another left ridgeline and a region sandwiched by the right ridgelines occur, the current left ridgeline and right ridgeline and another ridgeline are separated. The left and right edge data of one of the combinations of the left edge and the right edge of is stored in the stack memory, and the area sandwiched between the left edge and the right edge of the other combination is painted in scan line units. However, when the left ridge line or the right ridge line exits due to the scanning line reaching the exit vertex, the left ridge line and the right ridge line stored in the stack memory are restored, and the restored left ridge line and the right ridge line are restored. It is characterized in that the area sandwiched between the two is filled in every scanning line.

[0018]

In the outline character drawing method according to the first aspect of the present invention, the internal area of the character is divided into a plurality of polygonal areas, and the drawing processing is performed using the intermediate font data in which the polygons are integrated. In this case, those polygonal shapes are selected to have, for example, at most two entry vertices when viewed from an arbitrary direction. The coordinate conversion of the intermediate font data is performed at the time of drawing, and this coordinate conversion is a conversion including operations of parallel movement, expansion and contraction, and rotation. These transformations are transformations that transform a square surrounding a character font into a parallelogram. Each polygon is also deformed, but at most 2 when viewed from any direction.
The property of having one entry vertex is maintained. Therefore,
As a result of the coordinate conversion, there are either one or two polygonal entry vertices in the y-axis direction perpendicular to the scanning line on the display surface. Therefore, the scan conversion does not include the generation processing of the active edge list which has been conventionally required when the character is rotated, so that the processing speed is increased.

In the outline character drawing method according to the second aspect of the present invention, the filling process is performed using the entry vertex list sorted by the y coordinate and the left and right edge line data. In this case, the scan conversion is not performed using the active edge list, and the filling process is performed in units of a region sandwiched between a pair of left and right ridge lines and a pair of upper and lower scan lines. When a plurality of drawing line segments occur on the same scanning line, the area including the leftmost line segment (small x coordinate) is preferentially filled. That is, there is an entry vertex in the area being drawn, and 2
When separated into two drawing areas, the area on the left is preferentially filled. Then, the left and right ridge lines sandwiching the right area are paired and stored in the first stack memory so that the filling of the right area can be resumed after the end. If only the right edge of the left area that has been preferentially processed exits and the right area has not been drawn yet, the left side will be merged so that the left and right areas can merge after drawing the right area. The edge line is stored in the second stack memory. As described above, the entire inner area of the character is filled.

[0020]

Embodiments of the present invention will now be described in detail with reference to the drawings. (First Invention) FIG. 1 is an explanatory diagram of an example of polygon division of a character according to the first invention. In FIG. 1, the letters "love" are divided into polygons to suit the method of the present invention. That is, the internal area of the character is divided into a total of 15 polygons +1 to +15 that do not overlap each other. The continuous internal region is divided using a total of 14 dividing line segments in the figure, and is a shared side of two adjacent polygons. These polygons are divided so as to have a shape having one or two entry vertices when viewed from an arbitrary direction. For example, when viewed from the y-axis direction, all polygons other than the +5 polygon have a shape with only one entry vertex. The +5 polygon has two entry vertices when viewed in the y direction, but has only one entry vertex when viewed in the x-axis direction, and has two entry vertices in the other directions. A convex polygon such as +9 (in the illustrated example, a substantially parallelogram) is a polygon other than +9, which is a concave polygon and may have two entry vertices depending on the viewing direction, but three entry vertices. It is divided so that it has no vertices.

In the following, first, regarding the first invention, a method of dividing into a convex polygon or a concave polygon having two entering vertices, a coordinate conversion method of these polygons, and a drawing method of these polygons will be described in order. . [Polygon dividing method] When the outline of a character is formed by a plurality of polygons, a method of dividing by a plurality of dividing line segments that cross the internal area of the character will be described. To prevent the increase of intermediate font data, it is necessary to keep the number of divisions to a minimum. The polygon surrounding each of the internal regions separated by the dividing line segment must be a convex polygon or a concave polygon with two entry vertices as an essential condition. These polygons are coordinate-transformed and drawn. The convex polygon is drawn at high speed by the second conventional method. Even a concave polygon having two entry vertices can be drawn at a speed similar to that of a convex polygon, but in order to improve the drawing speed, a shape close to a convex polygon is desirable. The division procedure satisfying the above conditions is shown below.

A) The unevenness is discriminated based on whether the interior angle of the interior of the character for each vertex on the contour line is larger or smaller than 180 degrees. b) A segment line segment candidate is selected depending on whether the line segment connecting the concave vertices is completely included in the internal area of the character. c) Since the dividing line segment divides the interior angles of both vertices into two, the one that eliminates the concave vertex by this division is a strong candidate. This divides the convex polygon or the concave vertex into two concave polygons. d) The number of divided polygons is small, and the shape is a convex polygon or a concave polygon with two entry vertices, and the actual dividing line segment is determined so that the area of each polygon is divided as evenly as possible. e) When the divided polygon includes a concave polygon, it is confirmed that the concave polygon has two or less entry vertices in any direction.

The division result shown in FIG. 1 is obtained by artificially determining the division line segment according to the above procedure. This gives a small set of simple shaped polygons. Figure 8
Shows the method of confirming the condition e) and the method of division described above.
In the concave polygon of FIG. 8A, the concave portion between the vertices P1 to P7 (a chain of concave vertices) has an outward normal direction n which is erected at both ends thereof.
When viewed from the direction between 1 and n6 (θ1), P1 and P7 are both entry vertices, and the recess between P7 and P9 is P7 and P9 when viewed from the direction (θ2) between n7 and n8. And Therefore, when viewed from the direction in which both θ1 and θ2 are overlapped with each other, there are three entry vertices P1, P7, and P9, and the condition e) is not satisfied. However, as shown in FIG. 8B, if the polygon is divided by the dividing line segment P4 Q passing through the concave vertex P4 into two polygons, θ1 and θ2 do not overlap in the polygon on the right side. Therefore, the condition e) is satisfied. θ1 and θ2 do not overlap in the same direction or in the opposite direction, and their sum does not exceed 180 degrees. The closer the values of θ1 and θ2 are to 0, the closer the polygon is to the convex polygon.

[Polygon Coordinate Conversion Method] FIG. 9A shows a polygon inscribed in a rectangle ABCD in the outline font definition coordinate system O-xy. FIG. 9B shows the result of conversion into the screen coordinate system O'-x'y 'by the following equation. x ′ = ax + by + e y ′ = cx + dy + f The transformation represented by this equation is called an affine transformation. As a result, the rectangle ABCD is mapped to the parallelogram A'B'C'D ', and the polygonal map is mapped to this parallelogram A'B'C'D'.
Inscribe. For example, as shown in FIG. 9B, the mapping of this polygon has the entry vertices V′1 and V ′ at which the y ′ coordinates are minimal.
Exit vertices V'3, V'which have 2 and have a maximum y'coordinate.
4 and their original images are V1, V2, V3,
Set to V4. If the y'coordinate values of V'1, V'2, V'3, V'4 are y'1, y'2, y'3, y'4 respectively,
The original image of the horizontal scanning line passing through these is four inclined parallel lines shown in FIG. The fact that these four straight lines are parallel is clear from the form of the equation shown in FIG. Further, when the coordinate axes are appropriately provided in the direction perpendicular to these four straight lines (the y'direction), the coordinate value can be made to coincide with cx + dy + f, that is, y '. Therefore, V1, V2,
V3 and V4 are the entrance vertex or the exit vertex with respect to this coordinate axis. If the original image of the polygon has two or less entry vertices in an arbitrary direction, it can be seen that the coordinate value y ′ of this coordinate axis also has two or less entry vertices. Therefore, the affine transformation does not result in more than three entry vertices.

[Polygon drawing method] FIGS.
3 shows a scan conversion flowchart of a character string according to the present invention.
In FIG. 10, the loop is a loop executed once for each character drawn, and the loop is a loop executed once for each polygon drawn. Here, each polygonal shape has one or two entry vertices when viewed in the downward direction (y direction) perpendicular to the scanning line on the display surface after coordinate conversion. This becomes clear when the left and right ridge line data is created while checking the increase and decrease of the y coordinate in step S204. If there is one entry vertex, the process proceeds to step S206, and the y coordinate of the entry vertex is set. Then, in the next step S207, the first line segment data of the left and right ridgelines that exit from the approach vertex is (xl, h
l) and (xr, hr). Step S209
Executes the drawing process of FIG. The drawing process of FIG. 12 is a region drawing process between a pair of left and right ridges. In this process, whether any of the left and right ridgelines is leaving (step S24)
9) or when the drawing is performed by the height h corresponding to the designated number of scanning lines (step S250), the called processing is returned to. Then, this process executes the same process as the drawing part of the conventional second method at a similar speed (step S2).
41-S248). In this way, a polygon having one entry vertex is drawn. When there are two entry vertices, the scan conversion processing unique to the present invention is performed.

FIG. 13 is an explanatory view illustrating a method of drawing a concave polygon having two entry vertices in different cases. The y-coordinates of the two approach vertices V1 and V2 are y1 and y2 (y1 ≦ y
2), first, the height h from y = y1 to y = y2 h =
The area of y2-y1 is unconditionally filled. This state is shown by hatching. This processing is executed by calling the area drawing processing of FIG. 12 as drawing A1 in step S215 (only when h> 0).

Next, the magnitude relationship between the current values xl and xr of the x-coordinates of the left and right ridges at the time of painting to the height of V2 and the x-coordinate value x2 of V2 is checked in step S216 in FIG. If x2 <x1, the polygon is placed in the state as shown in FIG. 13 (a), and therefore the processing of step S217 and thereafter is executed. In step S217, the drawing process of FIG. 12 is called again without resetting the parameters for drawing A2. As a result, the area A2 following the area A1 is filled. When the first exit vertex V3 is reached, the left edge line exits and the drawing process ends. In the next step S218, the current value (xr, hr) of the right ridge is temporarily saved in a variable (xs, hs). Steps S219 to S22
1 draws the area A3 starting from the second approach vertex V2.
When the right ridge line exits and returns at the vertex V3, the next step S
At 222, the left edge line is used as it is, and only the right edge line is restored to the value (xs, hs). And step S2
The area A4 is drawn at 23. Finally, the left and right ridge lines simultaneously exit at the second exit vertex V4, and the drawing of the polygon is completed.

On the other hand, in step S216, x1 <x2 <x
In the case of r, as shown in FIG. 13B,
The space between the right and left ridgelines starting from the second approach vertex V2 is not filled. That is, in steps S224 to S226, only the right ridge line is replaced with the right ridge line starting from V2 and the area B2 is drawn. That is, the right edge is saved in step S224. Then, in step S225, the right edge of the second entry vertex is extracted. The right edge in this case is the right edge when the vertex V2 is seen from inside the polygon. Then, in step S226, the area B2 is drawn. afterwards,
In steps S227 to 230, the area B3 is drawn. That is, in step S227, the y coordinate y of the second approach vertex V2.
Take out 2. Then, in step S228, the left edge of the second approach vertex V2 is extracted. The left edge in this case is
It is an edge on the left side when the vertex V2 is seen from the inside of the polygon.
Then, in step S229, the right edge saved in step S224 is restored. By this, step S
At 230, the area B3 is drawn.

If xr <x2 in step S216, the process is as shown in FIG. 13C, and therefore a polygon is drawn by the same processing procedure as that shown in FIG. 13A (steps S231-). S237). In this case, step S232
The left edge is retracted, and the left edge is restored in step S236. When a horizontal line segment is included in the sides of the polygon as shown in FIG. 7C, control is performed so as to be skipped when the left and right edge line data is created in step S204 or when the edge is updated in FIG.

(Second Invention) FIGS. 14 to 21 are explanatory views of the outline character filling procedure according to the second invention. First, if the contour line of a character is formed by a plurality of line segments and the contour lines themselves do not intersect, the y coordinate of the vertex on the contour line with respect to the reference coordinate y axis introduced to the character is minimal. The entry vertices of y so that the y coordinates are in ascending order, and if the y coordinates are equal, the y axis is rotated counterclockwise 90
An approach vertex list is created in which the x-coordinates about the reference coordinate x-axis rotated by degrees are in descending order. Then, the left ridge line and the right ridge line are defined as a series of line segments from the respective entry vertices along the contour line to the leftward and rightward directions of the character internal region, respectively, to the exit vertices where the y coordinate is maximum. obtain. After that, painting is performed according to the following procedure.

First, as shown in FIG. 14, the area sandwiched by the left edge line and the right edge line starting from an unused entry vertex in the entry vertex list is filled in units of scanning lines, and the scanning line reaches the next entry vertex. At this time, the following filling is performed according to the position of the entry vertex. When the next entry vertex appears outside the currently filled area as shown in FIG. 15, either the area currently being filled is continuously filled or the newly created area is filled. Or select. For example, when the next entry vertex appears on the right outer side of the area currently being filled, this is ignored and the area currently being filled is filled as shown by the dotted line in FIG.

On the other hand, as shown in FIG. 16, when the next entry vertex appears inside the currently filled area, there are two areas sandwiched by the left edge line and the right edge line. At this time, the left and right edge data of one combination is stored in the stack memory. For example, in the example of FIG. 16, the right and left edge data of the right side area is stored in the first stack memory, and the left side area is filled as indicated by the dotted line. Also, as shown in FIG. 17, when the next entry vertex appears outside the left side of the area currently being filled,
The left and right edge data of the area currently being filled is stored in the first stack memory. Then, as shown by the dotted line, the area starting from the next entry vertex is filled. When the filled area is divided into two as described above, the left side area is preferentially filled.

On the other hand, when the scanning line reaches the exit vertex and the right ridge exits as shown in FIG. 18, the left ridge data of the area currently being filled is stored in the second stack memory, Restore the left edge and right edge stored in the first stack memory. And
As indicated by the dotted line, the area sandwiched between the restored left and right ridges is filled in scan line units.

As shown in FIG. 19, when the left ridgeline leaves, the left ridgeline data stored in the second stack memory is restored and the area indicated by the dotted line is filled. On the other hand, as shown in FIG. 20, the right ridge line exits and the first
When the stack memory of is empty, as shown by the dotted line, filling of the area starting from an unused entry vertex is started.
Then, as shown in FIG. 21, both the left and right ridgelines leave,
When there are no unused entry vertices, the filling is finished.

22 and 23 show a flowchart of scan conversion for one character according to the second invention. Also, FIG.
Indicates the result of filling the outline character “O” according to this flowchart. And in FIG. 25,
After rotating the outline character of "love", the result of painting according to this flowchart is shown. In FIG. 24 and FIG. 25, the internal area of the character is painted using five types of paint patterns, but these paint patterns ~ pass through any route of ~ on the processing paths of the flowcharts of FIGS. 24 and 25. Shows what you did.

In step S301 of FIG. 22, an entry vertex list is created. In the example of FIG. 24, there are two entry vertices. Then, in step S302, the R stack (first stack memory) and the L stack (second stack memory) are emptied. Then, in step S303, the entry vertex list is viewed from the beginning, and in step S304, it is determined whether there is an unused entry vertex. If there is an unused entry vertex, the first entry vertex is set in each scanning parameter in step S305. And step S
At 306, the next entry vertex is set to y next.

As a result, steps S307 to S310 are performed.
The area of FIG. 24 is filled with. When the next entry vertex is filled, step S30
After exiting the determination of step 7, steps S311 and S31 of FIG.
At 4, it is determined whether the vehicle is to the left or to the right. In the example of FIG. 24, the part to be filled is divided into the left and right, not the left approach or the right approach. Therefore, step S315
Is executed and the parameter of the right ridge line currently being drawn is R
It is stored in the stack (first stack memory). Further, in step S316, the parameter on the left side of the entry vertex is stored in the R stack, and the parameter on the right side of the area currently being drawn is changed to the parameter on the right side of the entry vertex. Then, in step S317, the next entry vertex is set to ynext. In this case, since there is no next entry vertex, the information to that effect is set.

As a result, steps S307 to S310 are performed.
The area in the figure is filled with. When the exit vertex is filled, the determination in step S310 is skipped, and in steps S318 and S320 of FIG. 23, it is determined whether only the left exit or the right exit. In the example of FIG. 24, only the right exit is performed, so step S321 is executed, and the parameter on the left side of the area currently being drawn is the L stack (second
Stack memory). And step S
At 322, it is determined whether the R stack is empty or not. In the example of FIG. 24, the R stack is not empty, so the drawing parameters on the left and right are extracted from the R stack at step S323. Then, in step S324, the next entry vertex is set to y next,
In this case, since there is no next entry vertex, the information to that effect is set.

As a result, steps S307 to S310 are performed.
The area of FIG. 24 is filled with. When the exit vertex is filled, the determination in step S310 is skipped, and in steps S318 and S320 of FIG. 23, it is determined whether only the left exit or the right exit. In the example of FIG. 24, only the left side exits, so step S319 is executed and the contents of the L stack are set in the drawing parameter on the left side.

As a result, steps S307 to S310 are performed.
The area in the figure is filled with. When the exit vertex is filled, the determination in step S310 is skipped, and in steps S318 and S320 of FIG. 23, it is determined whether only the left exit or the right exit. In the example of FIG. 24, since both left and right exits, step S322 is executed and it is determined whether the R stack is empty. At this time,
Since the R stack is empty, step S in FIG.
Return to 303. And since there are no unused entry vertices,
The drawing process ends after passing the determination in step S304.

Next, the case of painting the character "love" shown in FIG. 25 will be described. First, in step S301, a list of entry vertices for the character "love" is created, and in step S302, the R stack and the L stack are emptied. Then, through step S304, step S305
The first entry vertex P1 is set as a drawing parameter. At that time, in step S306, the next entry vertex P2 is y.
Set to next.

As a result, steps S307 to S310 are performed.
At, the pattern is filled from the first entry vertex P1. When reaching the next approaching vertex P2, the process exits step S307, and steps S311 and S3.
Although the determination at 314 is made, this approach vertex P2 is ignored because it is a right approach, and the next approach vertex P3 is set to ynext in step S317. Then, returning to step S307 of FIG. 22, the pattern filling is continued in steps S307 to S310. Even if the approaching vertex P3 is reached, the approaching vertex P3 is ignored like the approaching vertex P2 because it is a rightward approach. Therefore, pattern filling is performed up to the entry vertex P4.

At the entrance vertex P4, step S307
When exiting and the determinations of steps S311 and S314 of FIG. 23 are made, since the entry vertex P4 is neither leftward entry nor rightward entry, step S315 is executed and the right edge data is stored in the R stack. Then, in step S316, the ridge line data on the left side of the entry vertex P4 is stored in the R stack, and the ridge line data on the right side is set to the drawing parameter on the right side. At that time, the next approach vertex P5 is set to ynext in step S317.

As a result, steps S307 through S307 of FIG.
In S310, pattern filling is started. The approach vertices P5 and P6 are ignored because they are right approaches. Therefore, pattern filling is performed up to the entry vertex P7. Since the entry vertex P7 is neither leftward entry nor rightward entry, the same processing as in the entry vertex P4 is performed in steps S315 to S317. Similarly, since the approach vertices P8 to P11 are rightward approaches, they are ignored and the pattern is filled up to the approach vertex P12. Since the entry vertex P12 is neither leftward entry nor rightward entry, the same processing as in the entry vertices P4 and P7 is performed in steps S315 to S317. Then, at the exit vertex T1, the determination in step S310 is exited, it is determined to be the central exit vertex in the determinations in steps S318 and S320, and the determination in step S322 is performed.

In the determination of step S322, since the R stack is not empty, step S323 is executed and the contents of the R stack are set in the drawing parameter. that time,
In step S324, the next entry vertex P13 is set. Thereby, in steps S307 to S310, the region R
The pattern 3 is filled with the pattern. Then, the filling is finished at the exit vertex T2, and the left ridge line data at that time is stored in the L stack at step S321. Then, in step S323, the next contents of the R stack are set in the left and right edge line data. As a result, the area R4 is filled with the pattern. Similarly, the area R5 is filled with a pattern. After this, the R stack becomes empty, and the process exits step S322 and returns to step S30 of FIG.
Return to 3.

By repeating the above processing, the character "love" is painted as shown in FIG. The result is as follows. The area drawing from the first unused entry vertex of the entry vertex list starting from step S303 is filled with a pattern. This filling is repeated 7 times. When the entry vertex enters the drawing area and is divided into the left and right areas, the area on the left side (small x coordinate) is filled with a pattern. In that case, the area on the right side is step S31.
5, stored in the R stack in S316.

End or suspend drawing of the left area,
In step S323, when the R stack is taken out and the drawing of the area on the right side is restarted, it is filled with the pattern.
The drawing of the left side area is interrupted when only the right side edge line leaves, and the remaining left side edge line is stored in the L stack in step S321. When the left and right areas separated by the entry vertex merge, they are filled with a pattern. This is the case where only the left ridgeline of the right side area has exited, and in step S319, L
The left edge of the stack is fetched. When the entry vertex enters the left side outside the drawing area, it is filled with a pattern. In the character “love” shown in FIG. 25, three areas are filled with a pattern.

When the entry vertex enters the right side outside the drawing area, the entry vertex is ignored and the drawing area is not changed. In the example of FIG. 25, while the left side portion of the character is first drawn, many of the entry vertices on the right side do not satisfy the above condition and thus remain unused. In the flowcharts of FIGS. 22 and 23, the processing for this unused approach vertex is continued until there is no such unused approach vertex. Then, the entry of the next unused entry vertex in the entry vertex list is detected by its y coordinate y next. Steps S307 to S3 are being performed while drawing an area with neither entry nor exit.
Go around 10 loops. Steps S307 to S3 of FIG.
The processing content 10 is almost the same as that shown in FIG. Therefore,
Like the drawing process of the first invention, the drawing process of the second invention has a very high processing speed as compared with the first conventional method.

[Performance Comparison with Conventional Method] FIG.
22.5 outline characters of "love" of 0x200 pixels
16 shows the display results of 16 rotations. Also, FIG.
7 is a first conventional method, a second conventional method, and a first method of the present invention.
The bar graph shows the performance measurement results of the four methods of the invention and the second invention. As can be seen from the graph in FIG. 27, it takes 5.4 ms to draw the character of “love” that does not rotate with the conventional first method.
If ec is applied and the coordinates are converted and rotated 45 degrees, 8.4
It takes msec. On the other hand, in the first invention, it is 1.9 msec when not rotating and 2.5 msec when rotating 45 degrees,
About 3 times faster. At other rotation angles, the speed is similarly high.

The second invention also exhibits the same performance as the first invention. This is because the first method of the related art spends most of the time for processing the AEL, and indirectly calls edge data using the AEL is repeated many times in a small processing unit of horizontal line segment drawing. This is because in the invention and the second invention, drawing can be executed in area units using a pair of edge line data that has been called once. With the second conventional method, high-speed filling can be performed due to the same property, but the drawing speed is maximum (1.2 msec) only for upright characters without rotation, and when rotated, it is faster than the first conventional method. On the contrary, the processing becomes slow.

In the above embodiment, for example,
In the embodiment of the second invention, when two filled areas appear, the left area is preferentially filled, but the present invention is not limited to this, and the right side is preferentially filled. It goes without saying that it is okay.

[0052]

As described above, according to the outline character drawing method of the first aspect of the present invention, the internal area of the character is divided into polygons having a predetermined number or less of entry vertices. By filling, high-speed filling can be performed even if coordinate conversion such as rotation of characters is performed. Further, not only a convex polygon having one entry vertex but also a concave polygon having a predetermined number of entry vertices is allowed, so that the polygon can be easily divided into a small number of simple polygons. Furthermore, the result of polygon division is created and stored in advance as intermediate font data for the characters that are required to be drawn, so the process of the dividing process is not included in the drawing process, and it is possible to perform the optimum division for drawing. Is. And these polygons can be filled fast,
This property can be maintained even when the polygon is rotated. On the other hand, according to the outline drawing method of the second aspect of the present invention, the internal area of the character is divided by the scanning line passing through the entry vertex and the exit vertex, and each segmented area is drawn at high speed. Even in such a case, high-speed filling can be performed without dividing the polygon.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of an example of polygon division of a character according to the first invention.

FIG. 2 is an explanatory diagram of a character filling method according to a first conventional method.

FIG. 3 is a flowchart (part 1) of a scan conversion procedure according to a first conventional method.

FIG. 4 is a flowchart (No. 2) of the scan conversion procedure according to the first conventional method.

FIG. 5 is a configuration diagram of an active edge list according to a conventional method.

FIG. 6 is an explanatory diagram of a character filling method according to a second conventional method.

FIG. 7 is a flowchart of a scan conversion procedure according to a second conventional method.

FIG. 8 is an explanatory diagram of a method of dividing a concave polygon.

FIG. 9 is an explanatory diagram of coordinate conversion of a concave polygon.

FIG. 10 is a flowchart (No. 1) of the scan conversion procedure according to the first invention.

FIG. 11 is a flowchart (No. 2) of the scan conversion procedure according to the first invention.

FIG. 12 is a flowchart (No. 3) of the scan conversion procedure according to the first invention.

FIG. 13 is an explanatory diagram of a method of drawing a concave polygon.

FIG. 14 is an explanatory diagram of an outline character filling procedure (No. 1) according to the second invention.

FIG. 15 is an explanatory diagram of an outline character filling procedure (No. 2) according to the second invention.

FIG. 16 is an explanatory diagram of an outline character filling procedure (3) according to the second invention.

FIG. 17 is an explanatory diagram of an outline character filling procedure (No. 4) according to the second invention.

FIG. 18 is an explanatory diagram of an outline character filling procedure (5) according to the second invention.

FIG. 19 is an explanatory diagram of an outline character filling procedure (6) according to the second invention.

FIG. 20 is an explanatory diagram of an outline character filling procedure (7) according to the second invention.

FIG. 21 is an explanatory diagram of an outline character filling procedure (8) according to the second invention.

FIG. 22 is a flowchart (1) of a scan conversion procedure according to the second invention.

FIG. 23 is a flowchart (No. 2) of the scan conversion procedure according to the second invention.

FIG. 24 is an explanatory diagram of a character filling result (No. 1) according to the second invention.

FIG. 25 is an explanatory diagram of a character filling result (No. 2) according to the second invention.

FIG. 26 is an explanatory diagram of a display result of rotated characters.

FIG. 27 is a performance comparison diagram between the first invention and the second invention and the conventional method.

[Explanation of symbols]

 +1 to +13 polygon

Claims (3)

[Claims] 1. When a contour line of a character is formed by a set of a plurality of line segments and an inner region surrounded by the plurality of line segments is divided into a plurality of polygons, the inner region of each polygon is The logical sum is for the entire internal area of the character, and the number of entry vertices at which the position coordinates on the coordinate axes in which all the polygonal shapes are set in arbitrary directions are the minimum is less than or equal to a predetermined number. The polygons are selected, line segment data of the polygons are prepared in advance as intermediate font data for the characters, the polygons that are the intermediate font data are read out, the coordinates are converted, and the coordinates are converted. An outline character drawing method characterized by recognizing an intersection between each polygon and a scanning line and filling the space between the intersections. 2. All entry vertices in which the y-coordinate of the vertex on the contour line with respect to the reference coordinate y-axis is minimum when the contour line of the character is formed by a plurality of line segments and the contour lines themselves do not intersect. To obtain an entry vertex list in which the y-coordinates are in ascending order, and if the y-coordinates are equal, the x-coordinates with respect to the reference coordinate x-axis rotated 90 degrees in the counterclockwise direction are in descending order, A left ridge line and a right ridge line are defined as a series of line segments from the respective entry vertices to the exit vertex where the y coordinate is maximized in a direction looking to the left and a direction looking to the right of the inner region of the character along the contour line. Then, the area sandwiched between the left edge line and the right edge line starting from an unused entry vertex in the entry vertex list is filled in for each scan line, and when the scan line reaches the entry vertex, another left edge line and a right edge line are created. When a region between ridge lines occurs, the left and right ridge line data of the current left ridge line and right ridge line and another left ridge line and right ridge line combination is stored in the stack memory, and the left ridge line and right side ridge line of the other combination are stored. When the left ridge line or the right ridge line exits by filling the area sandwiched by the ridge lines in units of scanning lines and the scan line reaches the exit vertex, the left ridge line and the right ridge line stored in the stack memory are restored. The outline character drawing method is characterized in that the area sandwiched between the restored left edge line and right edge line is filled in units of scanning lines. 3. The area between the left ridge line and the right ridge line starting from an unused entry vertices in the entry vertices list is filled in units of scanning lines, and the left ridge line or the right ridge line sandwiching the region reaches the exit vertex, or Continuing to fill the area until the scan line reaches the next unused entry vertex in the entry vertex list, and when the scan line reaches the entry vertex, the x coordinate of the entry vertex is the x coordinate of the left and right edges. If it is larger than the current value of, the area is continued to be filled without using the approach vertex, and if the approach vertex is between the left edge line and the right edge line, another left edge line starting from the approach vertex And storing the right edge line as a pair in the first stack memory, filling a region between the original left edge line and another right edge line starting from the entry vertex, and setting the x coordinate of the entry vertex. Is smaller than the current value of the x coordinate of the left edge line, the pair of original left edge line and right edge line is stored in the first stack memory, and between the other left edge line and right side edge line starting from the entry vertex. When the left edge line or the right edge line reaches the exit vertex and only the right edge line reaches the exit vertex, the left edge line is stored in the second stack memory and the first stack is filled. If the latest left-side edge and right-side edge pair stored in the memory is recovered, the y coordinate is reset, and filling is resumed, and if only the left-side edge reaches the exit vertex, the second
Recovers the latest left edge stored in the stack memory of, resumes the fill with the current right edge, and when the right edge reaches the exit vertex, if the first stack memory is empty, 3. The outline character drawing method according to claim 2, wherein filling of the area from the unused entry vertex having the smallest y coordinate is resumed, and if there is no unused entry vertex, the filling is finished.