JPH0194385A - Reduction of vector font - Google Patents

Abstract

PURPOSE: To attain smooth and quality printing by selecting/erasing points arranged in the vertical direction by rounding processing and preventing the side part of a character line from being projected due to the rounding processing. CONSTITUTION: When plural points arranged in an oblique direction before rounding processing are arranged in the horizontal direction by the rounding processing, one of points arranged in the horizontal direction is selected and other points are erased to reduce the thickness of a character line. After completing the reducing processing of all coordinate data stored in a memory M1, coordinate data in a memory M3 are transformed to an outline pattern and then the inside of the outline is painted out. Consequently a character line is prevented from being projected due to reducing processing, so that a smooth and quality vector font is reproduced and utilized for printing or the like.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a vector font reduction method used in fields such as electronic publishing and word processing.

``Prior Art'' For example, when printing calligraphy characters in a word processor, the calligraphy characters are read in advance with a scanner, camera, etc., and converted into binary data of, for example, 256×256 dots. Next, the outline of the character is extracted from this binary data to generate outline capture data and stored in the dictionary memory. And when printing,
First, contour interpolation data is read from the dictionary memory, and then coordinate data indicating the coordinates of the main outline points of 256x256 dots is obtained from the read data. Next, this coordinate data is reduced. That is, for example, when printing is performed at 48 x 48 dots/1 character degree, each coordinate data is multiplied by 48/256, and then the decimal part of this multiplication result is rounded off.Next, as described above, The coordinate data obtained by the reduction process is converted into an outline pattern.This results in a pattern in which the dots on the outline of the character are "l" and the other dots are "0".Figure 10 shows this outline. An example of a pattern is shown. In this figure,
The parts other than the number "l" are "0".

Next, change the inside of the character in the above outline pattern to “
1". In this way, a dot pattern (48 x 48 dots) of calligraphy characters is obtained.

"Problem to be solved by the invention" By the way, in the conventional reduction processing described above, 256
In the X256 pattern (when played back at 1x), multiple points running diagonally may line up horizontally in a straight line as a result of the reduction and rounding process, as shown with the symbol P in Figure 10. be. In such a case, there is a problem in that the character lines protrude from the sides, causing the character lines to lose their smoothness.

An object of the present invention is to provide a vector font reduction method that prevents the above-mentioned protrusion of the sides of character lines and enables smooth, high-quality printing.

"Means for Solving the Problem" This invention solves the problem when, as a result of the rounding process, a plurality of points that were lined up diagonally before the rounding process are lined up horizontally.
The method is characterized in that one of the points arranged in the horizontal direction is selected and the other points are erased so that the thickness of the character line becomes thinner.

"Operation" According to the present invention, points arranged in the horizontal direction as a result of the rounding process are selected and deleted as described above, so that it is possible to prevent the side parts of character lines from protruding due to the rounding process.

"Embodiment" Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a flowchart showing a vector font reduction method according to an embodiment of the present invention.

First, in step Sal, function font data of a character to be printed is read from the dictionary memory.

Next, in step Sa2, the coordinate data of the main outline points of the 256x256 dot outline pattern (that is, the pattern reproduced at 1 times the compression) is calculated from the read function font data and stored in the storage means. FIG. 2(a) shows a state in which the coordinate data of the kana character "i" whose outline is composed of two loops as shown in FIG. 2(b) is stored in the memory M1. In this memory Ml, coordinate data of main points on the outline are stored in clockwise order from the loop start point for each loop, and a loop end mark is added at the end. The coordinate axes on which the above coordinate data is based are as shown in FIG. 2 (b).

Furthermore, in step Sa2, as shown in FIG.
The total number of coordinate data in the memory M1 is written into the data counter DC set in the memory M2. Furthermore, in this step Sa2, the addresses of the memory Ml (ADa, A in FIG. 2(a)
Db.

ADc) is written. In this example, the total number of loops is two, but three addresses are stored in the loop start point stack.

The first and second addresses are the addresses of the first and second loop start points, respectively, while the third address stores the address next to the loop end mark of the second loop.

Next, in step Sa3 of FIG. 1, the loop start flag LSF (FIG. 3) is set to "t", and then in step Sa4, the loop counter LC (third
rlJ is set in (Figure). When this initial setting is completed, the process advances to step Sa5. In step Sa5, it is determined whether the data in the data counter DC is "0". Then, if the judgment result is "NOJ," step Sa
Proceed to step 6. In step Sa6, the first coordinate data (Xl.

Yl) is read out. Next, proceeding to step Sa7,
Data counter DC is decremented.

Next, in step Sa8, the loop start flag LS
It is judged whether F is “l” or not, and this judgment result is r
If YES, the process advances to step Sa9.

In step Sa9, based on the start address of the next loop (in this case, address ADb) stored in the start point stack SP in memory M2, the point immediately before the end point of the loop currently being processed is Coordinate data (in this case, X 78. Y ta) is read out. Note that the coordinate data of the end point of the loop (X to, Y 78) is the same as the coordinate data of the start point of the loop (X,, Yυ). The data is called final coordinate data.Next, step 5alO
Proceeding to step Sat 1, the direction attribute of the start point is detected based on the coordinate data of the start point and the final coordinate data described above, and then proceeding to step Sat 1,
A reduction calculation is performed on the coordinate data of the start point,
Next, proceeding to step 5a12, step Sal 1
Rounding of the calculation results is performed. Note that the processing in steps 5alO to 5a12 will be described in detail later. Next, when proceeding to step 5a13, the step S
The reduced coordinate data (XI', Ylo) obtained by the processing of al O, Sal 1 is stored in the memory M3 (Fig. 2 (c)
), and then proceeding to step 5a14,
Loop start flag LSF is cleared. and,
Return to step Sa5.

In step Sa5, it is determined again whether the data in the data counter CD is "0" or not, and then in step Sa6, the second coordinate data (Xm, Yt) is stored in the memory M.
2. Next, proceeding to step Sa7,
The data in the data counter CD is decremented, and the process advances to step Sa8. In this case, step Sa
The result of the determination in step 8 is "NO" (see step 5a14), and the process proceeds to step 5a15. In step 5a15,
In step Sa6, it is determined whether or not the read data is a loop end mark. If the result of this determination is "NO", the process advances to step Sal6. In step 5a16, it is determined whether the processed flag SZF (FIG. 3) is I''. This processed flag SZF is cleared in the initial state, and therefore, at this point, the judgment in step 5a16 is The result is "NO" and the process advances to step 5a17. In step 5a17, the direction attribute of the point currently being processed is detected, and then in step 5a1
8, the direction attribute of the next point to be processed is detected. The detection of this direction attribute will be described in detail below.

First, the direction attribute is the direction of the point currently being processed as seen from the previously processed point, and in this embodiment, the directions <1> to 8> shown in FIG. 4 are determined. Detection of this direction attribute is performed as follows.

That is, the amount of displacement in the X direction (Xm If the amount of displacement in the direction is 0, the direction attribute is <1>; if the amount of displacement in the X direction is negative and the amount of displacement in the Y direction is 0, the direction attribute is <2>; if the amount of displacement in the If the displacement in the X direction is O and the displacement in the Y direction is positive, the attribute is <4>. If the displacement in the X direction is positive and the displacement in the Y direction is negative, the direction attribute is <3>. is <5>,
If the amount of displacement in the X direction is negative and the amount of displacement in the Y direction is positive, the direction attribute is <6>, and if the amount of displacement in the X direction is negative and the amount of Y direction displacement is negative, the direction attribute is <7>, the amount of X direction displacement. is positive and the amount of displacement in the Y direction is positive, the direction attribute becomes <8>.

Once the direction attributes of the point currently being processed and the point to be processed next are determined as described above, the process proceeds to step 5a19. In step 5a19, it is determined whether the point currently being processed is an acute angle portion or an obtuse angle portion based on the directional attributes detected in steps 5a17 and 5a18 described above. In other words, Figure 5 (a)
As shown in ~(d), the direction attribute of the current point and the direction attribute of the primary point are <5>, <6>, and <7>, respectively.
, <8>, <6>, <5>, <8>, <7>
In the case of , the previous point, current point, and next point form an acute angle, and on the other hand, as shown in Figure 5 (E) to (H), the direction attribute of the current point and the direction attribute of the primary point each , <5>, <8>, <6>, <7>, <7>
, <5>, <8>, ,',6>, then 1
The previous point, current point, and next point form an obtuse angle. Therefore, if the direction attributes of the current point and the next point are any of those shown in FIG.
Y E S J, and if it is neither of the cases shown in FIG. 5, the judgment result will be rNOJ.

If the determination result in step Sal 9 is rNOJ, the process advances to step 5a20. In step 5a20, a coordinate reduction calculation of the current point is performed. That is, for example, when printing is performed using "48x48 dots/character", the coordinate data of the point currently being processed is multiplied by 48/256. Next, proceeding to step 5a21, step 5a20
A first rounding process is performed to round off the calculation result below the decimal point.

This rounding process is performed based on the direction attribute detected in step 5a17 described above and the table shown in FIG. In other words, if the direction attribute of the point currently being processed is, for example, <1>, the first decimal place of the reduced X coordinate will be 8 to 9, and the first decimal place of the reduced Y coordinate will be One person is thrown away, two people are thrown away. Similarly, if the direction attribute is <7>, for example, the X coordinate after reduction is 1 to 2 people, Y
Coordinates are given by 1 and 2 people. Furthermore, in the table of Figure 6,
In reality, a boundary value for rounding up/down is set. Further, the processing in steps Sal 1 and 5a12 described above is also similar to the processing in steps Sa20 and Si21 described above. Further, instead of the first rounding process described above, rounding may be performed.

Next, in step 5a22, it is determined whether the Y coordinate of the point previously reduced is the same as the Y coordinate of the point currently reduced. If the result of this judgment is rNOJ, that is, if the previously processed point after reduction and the point after current processing is not on the horizontal straight line, the process advances to step 5a23. In step 5a23, the reduced coordinate data (Xs', Yt'
) is written in the memory M3 (FIG. 2(c)), and the process returns to step Sa5. On the other hand, if the determination result in step 5a22 is rY E S J, that is, if the previously processed point and the point currently processed are on a horizontal straight line, the process advances to step 5a24. In step 5a24, one of the coordinate data of the previously processed point and the coordinate data of the point for which the current reduction process has been completed is selected as the valid coordinate.

This selection is made based on the table shown in FIG. This table will be explained below.

In this table, the coordinates of the previously processed point are a(ax, by), and the coordinates of the point currently processed are b(bx, by). Also, ax=bx+
1 means that the X coordinate of the previous point is 1 from the X coordinate of the current point.
This is the case where the point is to the right, and ax=bx-1 is the case where the X coordinate of the previous point is one point to the left of the X coordinate of the current point. be. The numerical value at the top indicates the direction attribute of the current point.

As is clear from this table, if the direction attribute of the current point is <1 > or <2 >, it is a point that was originally lined up horizontally (even before reduction), and therefore a (of the previous point) coordinate data), b (other than current information)
In the case of , both a and b are valid. Also, “ax=bx+
In the case of IJ or rax=bx-1 and the direction attribute of the current point is <3> to <8>, one of the directions shown in the table is selected.

When the above-mentioned selection is completed, the process proceeds to step 5a25.

In this step 5a25, the effective coordinate data selected in step 5a24 is stored in the memory M3.

That is, first, if both a and b are valid, coordinate data b is stored in memory M3 since coordinate data a is already stored in memory M3 at this point. Further, if coordinate data a is selected, no processing is performed because the coordinate data a has already been stored. Further, when coordinate data b is selected, coordinate data a already written in the memory M3 is erased, and coordinate data b is written in its place. And step Sa5
Return to

On the other hand, the determination result of step 5a19 is rY E S J
In this case, the process advances to step 5a26. The reduction calculation in step Sa2 is performed. Next, when the process proceeds to step 5a27, a second rounding process is performed on the calculation result of step 5a26. This second rounding process is performed in step S
al 7. This is done based on the direction attribute detected in Sal 8 and the table in FIG. That is,
For example, if the direction attribute of the current point is 5〉 and the direction attribute of the next point is 〈6〉, round the X coordinate of the current point, round the Y coordinate to the nearest whole number, and round off the X coordinate of the next point. 718 people, round the Y coordinate by 112 people. Also, for example, the direction attribute of the current point 7〉,
If the direction attribute of the next point is <5>, round the X coordinate of the current point, round the Y coordinate by 213 people, and round the X coordinate of the next point to 213 people, and the Y coordinate of the next point by 6 Rounded up by seven people. Next, when the process proceeds to step 5a28, the result of the rounding process for the current point is written to the memory M3, and the result of the rounding process for the next point is temporarily stored in the memory M2. Next, when the process advances to step 5a29, "1" is written in the processed flag SZF. This flag SZF "!" indicates that the reduction processing of the coordinate data of the next point has already been completed. and,
Return to step Sa5. In addition, the table in Figure 8 has the following information:
In reality, a boundary value for rounding up/down is set.

Next steps Sa5, Sa6, Sa? , Sa8.
5a15 is executed sequentially, and the process advances to step 5a16.

Then, the judgment result of this step Sal 6 is rYES.
”, the process advances to step 5a30. Step 5a3
At 0, the coordinate data temporarily stored in the memory M2 and subjected to the reduction and second rounding processing is stored in the memory M3. Then, the process returns to step Sa5.

Thereafter, each coordinate data of the first loop is sequentially read out and reduced in the same manner. Then, when the loop end mark of the first loop is read out from the memory Ml, the determination result in step Sal 5 becomes "YES" and the process proceeds to step 5a31. In step 5a31, the end point coordinate data (X 7s', Y
7+1'') is rewritten as the coordinate data (xl'', y+') of the start point after the reduction process. Note that this process is performed because the coordinate data of the end point must always match the coordinate data of the start point. Next, when the process proceeds to step 5a32, the loop start flag LSF is set to 'l''.The process then proceeds to step 5a33, where the data in the loop counter LC is incremented to ``2''. Then, the process returns to step Sa5, and thereafter, the same reduction process as above is performed for the second loop.

Next, when the reduction process is completed for all the coordinate data in the memory Ml, the determination result in step Sa5 is rY E
The result is SJ, and the process proceeds to step 5a34.

In step 5a34, the coordinate data in the memory M3 is converted into an outline pattern, and then the inside of the outline is filled in. In this way, a reduced pattern in which the inside of the character is "l" and the outside is "0" is obtained.

The above is a vector font reduction method according to an embodiment of the present invention. Next, the effect of the effective coordinate determination process (step 5a24; see FIG. 7), which is the key point of this invention, will be explained. Now, let the solid line in Figure 9 be the outline of the character, and point PI.

Let P2 be the point after the reduction process and before the second rounding process, and the points Plc and P2c are each the point P1
, P 2 after rounding processing. Here, points Pie and P2c are lined up on a horizontal straight line. In this case, the X coordinate aX of the previous point P1 and the X coordinate bx of the current point P2 have a relationship of ax=bx+1, and
The direction attribute of the current point P2 is <6>. Therefore, the coordinate data b of the current point is selected based on the table shown in FIG. That is, point P2c in FIG. 9 is selected and point Pie is deleted. As a result, the point PLa that protrudes from the character line is erased.

"Effects of the Invention" As explained above, according to the present invention, it is possible to prevent character lines from jumping out due to reduction processing, and as a result, smooth and high-quality vector fonts can be reproduced.
It can be used for printing, etc.

[Brief explanation of the drawing]

FIG. 1 is a flowchart showing the process of a vector font reduction method according to an embodiment of the present invention, and FIG.
2 is a diagram showing the memory M1 in which the coordinate data before reduction is stored, FIG. 2 (b) is a diagram showing an example of the main outline points of characters, and FIG. A diagram showing the memory M3, FIG. 3 is a temporary storage memory M
2, FIG. 4 is a diagram for explaining direction attributes, FIG. 5 is a diagram showing combinations of direction attributes, and FIG. 6 is a diagram showing a table in which boundary values for the first rounding process are set. , FIG. 7 is a diagram showing a table used in the effective coordinate determination process, FIG. 8 is a diagram showing a table in which boundary values for the second rounding process are set, and FIG. 9 is a diagram showing a detailed explanation of the present invention. Figure 1.theta. is a diagram showing an example of a character outline pattern. Ml~M3...Memory.

Claims (1)

[Claims] In a vector font reduction method that reduces the vector points (coordinates) that make up the outline of a character at a specified reduction rate, and rounds off the numbers below the decimal point that occur in this reduction operation to obtain the reduced coordinates. , as a result of the rounding process, when multiple points that were lined up diagonally before the rounding process are lined up horizontally, one of the points lined up horizontally so that the thickness of the character line becomes thinner. A vector font reduction method characterized by selecting one point and erasing the other points.