JPH06332431A - Character generator - Google Patents

Abstract

PURPOSE:To provide a character generator capable of improving the quality of a character by arranging the horizontal and vertical lines of the character at the optimum line width corresponding to generated character and character size. CONSTITUTION:The coordinate value of a contour point representing the contour of the character, correction information including the correction line width(line width in fundamental size of character) of the line of the character provided with the coordinate value, and the kind of coordinate transformation are received. When it is instructed of correction by the correction information, the coordinate value of a reference line is estimated (processing 901, 902, 904), and the change of line width by the coordinate transformation is added on the correction line width (including fraction calculation) (processing 903, 905-907). The coordinate value after the coordinate transformation is calculated based on the estimated coordinate value of the reference line, the line width on which the change is added, and the kind of the coordinate transformation (processing 908-913).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character generator which generates character data by converting coordinate data indicating the shape of a character contour line into raster format data after coordinate conversion such as affine transformation. The present invention relates to a device, and more particularly, to a character generation method and device suitable for improving character quality.

[0002]

2. Description of the Related Art Conventionally, in order to generate a high quality character, font data in which contour information of a character is expressed by coordinate points of polygon vertices is stored in a font ROM, and the font data is extracted from this font ROM. Then, coordinate conversion such as affine transformation is applied to the required character size, and the result is output in raster format to an output device such as a laser printer or a display. In this method, the thickness of the character, that is, the line width, is proportional to the reduction and enlargement ratio of the character. Therefore, when the character is reduced, the character is displayed crushed, and when the character is enlarged, the display is The letters that were written became poor and were difficult to see.

In order to cope with such a problem, for example, as disclosed in Japanese Patent Laid-Open No. 3-210598, optimum line width values of horizontal lines and vertical lines are defined according to the generated character size, and the line width table is set. In the horizon or
When performing coordinate transformation such as affine transformation on a vertical line, the coordinate of one of the two lines forming the horizontal line or the vertical line is the coordinate value, and the coordinate of the other line is the above-mentioned one. The line width corresponding to the character size is realized by specifying the thickness from the coordinate value of the line after the coordinate conversion and selecting the value of the above line width table corresponding to the generated character size. Was.

[0004]

According to the above-mentioned prior art, since the line width is uniquely determined by the character size, the correction of characters in which a plurality of different line widths exist in the horizontal line and the vertical line in a single character. There was a problem that I could not do.

Further, it is necessary to store the line widths of all the generated character sizes in a table. Further, when it is desired to make the horizontal line width and the vertical line width different for each typeface, the different linewidths for each typeface are stored in the table. Need to remember. For this reason, there is a problem that a large number of line width tables are required and a large storage capacity of the tables is required.

Further, there is a problem that it is impossible to correct a character that has been rotated or skewed.

In view of the above problems, the present invention can improve the character quality by correcting the line width of a character having a plurality of different line widths on a horizontal line and a vertical line in a single character. It is an object of the present invention to provide a character generation device that can perform.

Another object of the present invention is to provide a character generator capable of improving character quality without requiring a large storage capacity of a table.

Further, even if shearing in the direction of one axis is performed by rotating in 90-degree units or obliquely deforming, by adjusting all horizontal lines and all vertical lines in the generated character to the optimum line widths, An object of the present invention is to provide a character generator capable of improving character quality.

[0010]

According to the present invention, in order to solve the above-mentioned problems, coordinate data of a contour point of a character composed of a character forming line having a thickness in a predetermined character size is received. A coordinate conversion means for performing designated coordinate conversion on the coordinate data, and a raster conversion means for receiving the coordinate data converted by the coordinate conversion means and converting the coordinate data into character data represented in a raster format. In the character generation device, at least a part of the coordinate data may include a unit for storing correction information regarding correction of the coordinate data at the time of the coordinate conversion.

The correction information includes the direction of correction to be performed on the coordinate data to which the correction information is added, and the thickness information of the character forming line including the coordinate data in the predetermined character size. be able to. In addition, the correction information is a contour point on one of the upper and lower sides of the horizontal character component line, or a contour on one of the right and left line segments of the vertical character component line. Can only be given to points.

Further, the coordinate conversion means, in the coordinate conversion, with respect to the coordinate data to which the correction information is added, based on the coordinate data and the thickness information, above or below the character forming line. Of the other, the right side or the left side of the other line segment is obtained as the reference coordinate value,
The specified coordinate conversion is performed on the obtained reference coordinate values, and the coordinate conversion is reflected in the thickness information. The coordinate converted reference coordinate values, the reflected thickness information, and It is possible to provide a calculating means for calculating coordinate values of the contour points after coordinate conversion based on the designated type of coordinate conversion.

Further, a plurality of line width correction registers for storing the line width in the basic size of the character are provided, and the font ROM
Of the coordinate data input from, the correction information of the line width correction target line segment (horizontal line and vertical line) indicates which line width correction register is used to correct the line width. Correction number data, which is a value, can also be added.

Further, it is possible to provide a line width fraction processing register used when the calculation means performs a fraction processing of the data after the decimal point of the line width value after the affine transformation.

Further, rotation, mirror rotation, and oblique deformation (shear) in units of 90 degrees caused by coordinate transformation such as affine transformation.
It is also possible to provide a line width correction mode register for storing character deformation information such as, and to calculate the coordinate conversion type from the register.

[0016]

In the coordinate conversion, the coordinate conversion means, on the basis of the coordinate data and the thickness information, with respect to the coordinate data to which the correction information is added, the other of the upper side or the lower side of the character forming line, the right side. Alternatively, the coordinate value of the other line segment on the left side is obtained as the reference coordinate value, and the obtained reference coordinate value is subjected to the specified coordinate conversion, and the coordinate conversion is reflected in the thickness information to be coordinate converted. The coordinate value after the coordinate conversion of the contour point is calculated based on the reference coordinate value, the thickness information after reflection, and the designated type of coordinate conversion.

It is preferable to store different line width data values in the respective line width correction registers, and the correction number data added to the coordinate data of the horizontal line and the vertical line which are desired to have the same line width store the same line width value. Since the value for selecting the line width correction register is entered, the line width is corrected so that the most suitable line width value is obtained when the coordinates of each horizontal line and vertical line in the character are converted.

Further, by using the line width correction register and the line width fraction processing register in the calculation for obtaining the line width of the generated character, the ratio between the basic size of the character and the size of the generated character, and the line in the basic size. Even if the ratio of the width value to the most suitable line width value for the generated character size is different, you can select the fraction processing method after the decimal point of the generated line width of the character, so it is optimal for each generated character size. Gets the line width value character.

Further, the coordinate conversion method for calculating the line width is switched depending on the line width correction mode register value.
A line width correction process corresponding to a deformed state of a generated character is realized.

Of the start point coordinates and end point coordinates of the line segment to be corrected, only the start point coordinates are corrected, and the end point coordinates are processed by inheriting the coordinate values corrected by the start point coordinates, thus reducing the time required for character generation.

[0021]

Embodiments of the present invention will be described below with reference to the drawings.

First, an embodiment of an information processing apparatus used to carry out the present invention will be described with reference to FIG.

As shown in the figure, this information processing apparatus has a font RO for storing font data in which contour information of a character is represented by coordinate points (coordinate data) of the vertices of a polygon.
M (read only memory) 101, character generator 10
2, an input device 103 such as a keyboard device, a RAM (random access memory) 104 for storing programs for operating the system, a central processing unit (hereinafter, C)
A PU) 105, an output device 107 such as a display device or a printer device, and a page memory 106 for storing character data in raster format generated by the character generation device 102.
It consists of.

The CPU 105 is characterized by various parameters (detailed later) necessary for correcting the size and line width of a character to be generated according to the information from the RAM 104 and the input device 103. After setting the various registers in the character generator 102, the font data is read from the font ROM 101 and output to the character generator 102. The character generation device 102, which is a feature of the present invention, converts the coordinate data of the input font data into raster format character data according to the set register value. The raster-format character data created by the character generator 102 is read by the CPU 105 and temporarily stored in the page memory 106. The character data stored in the page memory 106 is output to the output device 1 by the CPU 105.
It is supplied to 07, and characters and the like are printed and displayed.

Next, the font data input from the font ROM 101 by the character generator 102 will be described with reference to the drawings. 2 is a data configuration example of the font data stored in the font ROM 101 of the characters shown in FIG. 1, FIGS. 3 and 4 are configuration examples of the coordinate data shown in FIG. 2, and FIG. It is explanatory drawing which shows the addition position of the correction number mentioned later.

In the following embodiments, the coordinate system of the lower left origin is used as the coordinate system. Unless otherwise specified, 16-bit data is used, the most significant bit is the sign bit, and the lower bit is the lower bit. 4 bits are fixed point data with a decimal part.

As shown in FIG. 2, the font ROM 10
Font data input from 1 is roughly divided into flag data (201, 206) and coordinate data (202, 2).
03, 204, 205). The flag data is one word in this embodiment, and the type of the flag data is the loop start flag 201 existing immediately before the head of one character data or the head coordinate of one closed loop.
In addition to the end flag 206 indicating the end of the data of one character, although not shown in FIG. 2, a curve flag indicating that the coordinate value on the curve is input in the outline of the character, a straight line There is a straight line flag that indicates inputting coordinate values, and negative numbers are preferably used for these flag data.

Basically, the coordinate data is, as shown in FIG. 3, the coordinate values of the X coordinate and the Y coordinate, and the correction number data used for correcting the line width (whether or not the correction is performed depending on this data, When the correction is performed, the X coordinate or the Y coordinate is corrected. When the correction is performed, the line width can be known.
The outer loop coordinate points (coordinate points P0 to P7) representing the character contour are stored counterclockwise, and the inner loop coordinate points (coordinate points P10 to P17) are sequentially stored clockwise. The correction number data is added only to the right side of the vertical line and the upper side of the horizontal line. In FIG. 6, the coordinate point P
Of the 0 to coordinate points P17, correction number data indicating that the line width of the vertical line is to be corrected is added to the coordinate points surrounded by circles, and the character generator corrects the X coordinate value of this coordinate point. Further, correction number data indicating that the line width of the horizontal line is corrected is added to the coordinate point surrounded by Δ, and the character generator corrects the Y coordinate value of this coordinate point.

For example, in the line segment connecting the coordinate points P13 and P14 and the line segment connecting P4 and P5, which are two line segments forming a horizontal line whose line width is to be corrected, the coordinate point P13,
The correction process is not performed on the line segment connecting P14 (such a line segment is referred to as a parallel reference line), and the coordinate points P4, P
The line width is corrected by correcting the Y coordinate of the line segment connecting the five.

Further, among the line segments connecting the coordinate points P4 and P5, only the coordinate point P4 has the horizontal line correction number data added to the coordinate data, and the coordinate point P4 has the Y coordinate value. The correction process is performed, but the coordinate value of the Y coordinate is not corrected for the coordinate point P5. As the coordinate value of the Y coordinate of the coordinate point P5, the coordinate value of the Y coordinate of the result of the correction processing performed on the coordinate point P4 is given (this is called inheritance).
After that, the process of inheriting the coordinate value of the result of the correction process performed for the coordinate point immediately before in this font data is
This is called the inheritance process of the previous point. Focusing on the processing performed by the character generation device for one coordinate point, since the correction data of the horizontal line is added to the coordinate data of the coordinate point P6, the coordinate value of the Y coordinate is corrected and the coordinate value of the X coordinate is corrected. Value is coordinate point P
Since it is the same coordinate point as 5, the coordinate value of the X coordinate corrected at the coordinate point P5 may be inherited.

As described above, the coordinate data is configured so that the coordinate points of the outer loop representing the character contour are sequentially input counterclockwise and the coordinate points of the inner loop are clockwise and sequentially input to the character generator. If the above correction number data is added only to the right side of the vertical line of the character and above the horizontal line, only the coordinate value of the start point coordinate of the start point coordinate and end point coordinate of the line to be corrected is corrected, and the end point coordinate is corrected. As for the coordinate value of, the coordinate value corrected by the start point coordinate may be inherited. This is done by storing in a flag whether or not the correction process was performed at the previous point,
This flag may be referred to when the processing of the current point is performed, and the inheritance processing of the previous point may be performed when the correction processing is performed at the previous point. According to this structure, even when the correction number data is one word as shown in FIG. 3 (only one of the X and Y coordinates can be specified), the X and Y coordinates can be corrected.

The coordinate data is configured so that the coordinate points of the outer loop representing the character contour are sequentially input to the character generator in the clockwise direction, and the coordinate points of the inner loop are counterclockwise. The same correction can be performed by adding the correction number data only to the left side of the vertical line of the character and below the horizontal line.

FIG. 4 shows an example of the coordinate data structure when the start coordinates of the loop are the end coordinates of the line segment to be corrected.
In addition to the coordinate data configuration example shown in (1), the correction number data of the coordinate data of the previous point is added to form a 4-word configuration. For example, in FIG. 6, when the coordinate point P6 is the start coordinate of the loop, the correction number data of the coordinate data of the coordinate point P17 is also added to the coordinate data of the coordinate point P6. When there are two correction numbers in this way, the coordinate values of both the X coordinate and the Y coordinate are corrected (because it is the start point, inheritance cannot be performed).

In this embodiment, the correction number data is added immediately before the coordinate value for correcting the line width as shown in FIG. 3, and only the lower 4 bits are used and the upper 12 bits are "0". Goes in. Further, when there are two correction number data as shown in FIG. 4, the first correction number data is used as the correction number data of the previous point, and the correction number data is set by setting the fifth bit from the lower order to “1”. It is determined whether there is one and two correction number data.

The processing of the character generator characteristic of the present invention will be described in detail below.

FIG. 5 is an example of a schematic flow chart of the processing performed by the character generating device 102. An outline of the operation of the character generating device 102 will be described below with reference to FIGS. 1, 2, 3 and 5. .

First, the font data is stored in the font ROM 1
One word is read from 01 (process 501), and the following branch processes (1) to (3) are performed according to the read data (branch processes 502 and 503).

(1) If the read data is the loop start flag 201, the process returns to step 501, and one word is read again from the font ROM.

(2) If the read data is the end flag 206, the process proceeds to step 508.

(3) If the read data is neither the loop start flag 201 nor the end flag 206,
The following processing is performed.

(A) The coordinate data for one coordinate (3 words in FIG. 3, 4 words in FIG. 4) is used as the font R.
Read from OM (process 504).

(B) The read coordinate data is subjected to coordinate conversion such as enlargement, reduction, oblique deformation (shear), mirror rotation, and line width correction characteristic of the present invention (process 505).

(C) Curve interpolation processing for generating coordinate values for dividing the coordinates into short straight lines that look like curves, for example, cubic Bezier curve interpolation (however, input coordinate data is on a curve This is executed only when the coordinate value is reached (when the curve flag is input) (process 506).

(D) Convert the input coordinate data into raster format data of a contour image in which lines are connected, and process 5
Return to 01. (Process 507) The process 508 performed when the end flag 206 is input is a process of filling the area surrounded by the line generated in the line generation process 507. In the present embodiment, by performing the series of processes described above, the input coordinate data can be converted into raster-type character data.

Next, the coordinate conversion processing 505 which is characteristic of the present invention.
FIG. 17 shows a list of registers that hold various parameters required for the above, and the contents will be described, and the enlargement, reduction, oblique deformation (shear), mirror rotation, and the like performed in the coordinate conversion processing 505 will be performed.
An outline of coordinate conversion processing such as line width correction characteristic of the present invention will be described.

Affine register A (hereinafter referred to as AMA), affine register B (hereinafter referred to as AMB), affine register C (hereinafter referred to as AMC), affine register D (hereinafter referred to as AMD), affine shift register An X (hereinafter referred to as ACSX) and an affine shift register Y (hereinafter referred to as ACSY) are registers that hold parameters used when performing affine transformation processing on input coordinate data. The affine transformation process is a process for performing coordinate transformation such as enlargement, reduction, rotation, oblique deformation (shear), and mirror rotation, and the coordinate data after the coordinate transformation of the coordinate data (XI, YI) is (XO, YO). Then, the formula (1
1) and the formula (12).

XO = AMA × XI + AMB × YI + ACSX (11) YO = AMC × XI + AMD × YI + ACSY (12) Among these parameters, AMA, AMB, AMC,
AMD can be calculated by equation (13), equation (14), equation (15), and equation (16) according to the conversion to be performed.
Values for X and ACSY are set for parallel movement of a region protruding from the negative region to a region that can be expressed by a positive value due to rotation, oblique deformation, or the like. In this embodiment, AMA, AMB, AMC, A
In the MD, the most significant bit is a sign bit and the lower 10 bits are data of a decimal part. Therefore, AM in equation (11)
The multiplication result using AMA, AMB, AMC, AMD such as A × XI needs to be shifted to the right by 10 bits. Then, these parameters are set in advance by the CPU 105 before the character generator 102 reads the font data from the font ROM 101, similarly to the LWR 6 and the like.

AMA = a (cosθ-sinθtanψ) (13) AMB = d (cosθtanφ-sinθtanφtanψ-sinθ) (14) AMC = a (sinθ + tanψcosθ) (15) AMD = d (sinθtanφ + cosθtanφtanψcosθ) (16) a: Expansion ratio in X direction d: Expansion ratio in Y direction θ: Rotation angle (clockwise is positive) φ: Italic angle 1 (Italic with respect to the Y axis, counterclockwise is positive) ψ: Italic angle 2 ( The italic type with respect to the X-axis and the clockwise direction are positive.) The line width correction mode register (hereinafter, referred to as WMR) is a parameter of the character deformation information obtained by the affine transformation, which is necessary for determining the direction for correcting the line width. Is a register for holding the
It is set whether or not rotation and mirror rotation in units of degrees and the presence or absence of shear caused by oblique deformation. Rotations other than 90 degree units
Set with shear. WMR configuration (bit allocation)
An example is shown in FIG. The WMR value can also be derived from the affine register value. For example, AMB and A
If "0" is set for MC and a positive number is set for AMA and AMD, it can be seen that the characters after affine transformation are only scaled and not rotated or sheared.
If "0" is set for A and AMD, a negative number is set for AMB, and a positive number is set for AMC, the characters after affine transformation are only scaled and rotated by 90 degrees, and are sheared. I know there isn't.

The line width fraction processing register (hereinafter, referred to as WCR) is a register for holding a parameter related to the fraction processing when calculating the line width, and the value of the parameter stored in this register in the calculated line width. Is added, and the fractional part (lower 4 bits) is set to zero to perform fraction processing. For example, when WCR is set to "0", the fractional part is truncated, and when "8" is set, the fractional part is set to 0.
If the rounding to four rounds 0.5 is performed and "4" is set, the rounding to 0.74 rounds the fractional part is performed (note that it is addition in hexadecimal notation).

The line width correction register 6 (hereinafter referred to as LWR6) to the line width correction register 15 (hereinafter referred to as LWR15) are registers for holding the line width in the basic size. This WMR and WCR, LWR6 ~ LWR
The parameters set in 15 are the same as in the AMA, etc.
It is set in advance by the CPU 105 before the character generator 102 reads the font data from the font ROM 101.

The X coordinate input register (hereinafter referred to as XI), the Y coordinate input register, and (hereinafter referred to as YI) are registers for holding the coordinate values to be converted in the coordinate conversion processing 505, and X Coordinate output register (hereafter
XO), a Y coordinate output register, and (hereinafter, referred to as YO) are registers for holding the coordinate values obtained as a result of the coordinate conversion processing 505. The line width correction result save register (hereinafter referred to as TEMP) is a register for saving the correction result to be inherited when the previous point inheritance process is performed. That is, each time the correction processing is performed, the correction result is saved in TEMP.

The applicable range of the line width correction in this embodiment extends to the case where the coordinate conversion is enlargement, reduction, rotation by 90 degrees, mirror rotation, and uniaxial shearing. Of these, the coordinate data (XI, Y
When the coordinate data after the coordinate conversion of I) is (XO, YO) and the value of LWR corresponding to the correction number data is W, XO,
YO can be calculated by the equation (17) and the equation (18),
This is the most characteristic point of the line width correction method according to the present invention. Equations (17) and (18) are equations when the upper left origin coordinate system is used as the coordinate system. For example, when the lower left origin coordinate system is used, equation (18) is 18 ')
Becomes Also, when the relationship between the reference line and the correction target line of the two lines forming the horizontal line or the vertical line is reversed, the calculation formula changes as in the case of changing the coordinate system.

XO = (AMA × (XI−W)) + (AMA × W−1) (17) YO = (AMD × (YI + W)) − (AMD × W−1) (18) YO = ( AMD * (YI-W)) + (AMD * W-1) ... (18 ') Here, Formula (17) is demonstrated easily. The idea of the line width correction method in the present invention is basically "to move the line to be corrected so as to have an optimum line width with respect to the reference line". Therefore, in equation (17), first, the coordinate value of the reference line is obtained by subtracting the line width W in the basic size from the coordinate value XI of the correction target line, and the reference value after affine transformation is obtained by multiplying this by the affine parameter AMA. The coordinate value of the line is calculated. By adding the line width after the affine transformation to the coordinate value of this reference line, the coordinate value after the affine transformation of the correction target line can be obtained. The line width after affine transformation can be basically obtained by multiplying the line width W (line width before affine transformation) in the basic size by the affine parameter AMA. However, the line width becomes 1 dot thicker when it is actually output to a printer, so (AM
A × W-1) is added as the line width.

The reason why the line width becomes thicker by one dot will be described with reference to FIG. For example, a horizontal line composed of a line segment whose Y coordinate is “2” and a line segment whose Y coordinate is “5” is
It should be 3 dots originally. However, when it is printed, it becomes 4 dots as shown in the figure, which is 1 dot thicker than the original line width.

The calculation results of (AMA × W-1) and (AMD × W-1) of the second term indicating the line width of the character to be output are rounded according to the value of WCR ((AMA × W-
If 1) and (AMD × W-1) are added to the value of WCR), higher quality characters can be generated.

Next, FIG. 18 shows an example of a correspondence table of the correction number data values, the line width correction contents and the processing in the coordinate conversion processing 505, and the correction number data will be described. FIG.
As shown in, when the correction number data value is "0" to "5", it means that the coordinate data is not a line width correction target, and when it is "6" to "9", the coordinate data is Means that it is the start point data of the vertical line of the line segment to be corrected, and "1
When it is from 0 "to" 15 ", it means that the coordinate data is the start point data of the horizontal line of the correction target line segment, and the value stored in the line width correction register corresponding to the correction number data value is As the line width in the basic size, the line width after affine transformation of vertical lines and horizontal lines is corrected.
When the value of the correction number data is "6", the value of LWR6 is set as the line width W of the vertical line in the basic size, and the correction process of the coordinate value of the X coordinate of the input coordinate data is performed. Similarly, when the correction number data value is "10", the value of LWR10 is set as the line width W of the horizontal line in the basic size, and the correction process of the coordinate value of the Y coordinate of the input coordinate data is performed.

As described above, in the present invention, there are four registers (LWR6 to LWR9) for holding the line width of the vertical line,
There are six registers that hold the line width of the horizontal line (LWR10-10
LWR15) Yes. Therefore, it is possible to generate a high quality character even for a character having a plurality of line widths (mixed) in one character. Further, in the above-described embodiment, four registers are provided as the registers for holding the line width of the vertical lines and six registers are provided as the registers for holding the line width of the horizontal lines. May be. In addition, since many Kanji characters have more horizontal lines than vertical lines, it is possible to set more line widths for horizontal lines, but more line widths for vertical lines. The number may be set, or the same number may be set.

FIG. 19 is a list of flags necessary for carrying out the coordinate conversion processing 505. The correction number number flag (hereinafter referred to as FC) is a 1-bit flag. In the process 504, "0" is set when only one correction number data is added to the input coordinate data, When two correction number data are added to the data, "1" is set.

Current point correction flag (hereinafter referred to as WCF)
The correction flag of the previous point (hereinafter referred to as BCF) is a 2-bit flag used when the inheritance process of the previous point is performed, and the correction process of the line width of the X coordinate is performed on the input coordinate data. When the correction is performed, the value is set to "3", when the correction processing of the line width of the Y coordinate is performed, the value is set to "2", and when the correction processing is not performed, the value is set to "0" or "1".

The values set in the WCR are shown in FIG.
A detailed description will be given using 0. FIG. 20 is an example showing the optimum line width for each generated character size. For example, a character of 1024 dots is stored in a font ROM as a reference size, and the horizontal width of this character is 2
Assuming that there are two types of line width of 5 dots and vertical line width of 60 dots or 50 dots, this character is simply affine transformed to generate a 56 dot character (reduction).
Then, the line width of the horizontal line is 1.37 dots (56 × 2
5 ÷ 1024), and the line width of the vertical line is 60 dots: 3.
28 dots (56 × 60 ÷ 1024), 50 dots:
Since it is 2.73 dots (56 × 50 ÷ 1024),
When the rounding processing or the rounding down processing after the decimal point is performed, the optimum line width (horizontal line 2 dots, vertical line 4 to 3 dots) shown in FIG. 20 is not obtained when the size is 56 dots. At this time, in order to realize the line width shown in FIG. That is, "Ch" (hexadecimal C, that is, decimal 13) is set in the WCR, and the WCR value is added to the line width value after the affine transformation.
As described above, by using WCR, it is possible to generate a character having an optimum line width according to the generated character size. Since the WCR can be used for fine adjustment of the line width, by setting the WCR to a value larger than the value used for the fractional processing below the decimal point according to the characteristics of the printer, the thickening of the line width and the decimal point can be performed. The line width may be narrowed by setting a value smaller than the value used in the following fraction processing, and "Fh" (F in hexadecimal, that is, 1 in decimal).
5) A larger value or a negative number may be set. The WCR may be set to different values depending on the character size, or may be set to the same value for all character sizes or a plurality of character sizes.

The processing performed in the coordinate conversion processing 505 will be described in more detail below with reference to the drawings. FIG. 7 is an example of a flowchart of the processing performed in the coordinate conversion processing 505.

First, the AMA, AMB, AMC, AM
Using the affine parameters stored in D, ACSX and ACSY, the affine transformation process is performed on the coordinate values stored in XI and YI, and the transformation result is stored in XO and YO (process 701).

In the branch process 702, when FC is "1", the line width correction process for the correction number of the previous point is performed. Therefore, the process moves to the branch process 703, and when FC is "0", the line for the correction number of the current point is performed. In order to perform the width correction processing, the processing shifts to the branch processing 707. FC becomes “1” when the coordinate data structure is 4 words as shown in FIG. 4, and when it is 4 words, the start coordinate of the loop is the end coordinate of the correction target line segment as described above. If it is.

In the branching process 703, when the correction number data of the previous point is "6" or more, the line width correction process is performed according to the correction number data of the previous point, and the coordinate value of the processing result is stored in TEMP ( When the processing 704) is less than “6”, the input coordinate data is not a correction target, so the above processing is not performed. Then, by clearing WCF and FC to "0" (process 705), the line width correction process for the correction number data of the previous point is ended, and the process returns to the branch process 702.

When the correction number data is "6" or more, the branching process 707 performs the line width correction process according to the correction number data of the current point and sets WCF according to the correction process (process 708). ), The process moves to the branching process 709, but when it is less than “6”, the input coordinate data is not the correction target, and the coordinate conversion process is ended.

The branching process 709 moves to the branching process 712 when the BCF is less than "2", but when the BCF is "2" or more, the line width is corrected at the previous point, so the coordinates stored in the TEMP. The inheritance process of the previous point is performed using the value (process 710). Since the inheritance process of the previous point is performed, BCF is cleared to "0" (process 711), and the process proceeds to branch process 712.

In the branch processing 712, the coordinate conversion processing is ended when WCF is less than "2", but when WCF is "2" or more,
The coordinate value after the line width correction calculated in the process 708 is TEMP.
(Processing 713), the value of WCF is transferred to BCF, WCF is cleared to “0” (Processing 714), and then the coordinate conversion processing is ended.

As described above, in the present embodiment, of the correction target line segment whose line width is to be corrected, only the start point coordinates are corrected and the end point coordinates are simply inherited from the previous point (loop start). Only when the coordinates are the coordinates of the end point of the correction target line segment, the inheritance processing of the previous point is not performed), so that the character can be generated at high speed. Further, the correction processing may be performed on both the coordinates of the start point and the end point of the correction target line segment for correcting the line width, and the character quality remains unchanged. In this case, WCF, BC
F, FC, TEMP are not required, nor is the inheritance process of the previous point required.

Next, the line width correction processing 704 and the line width correction processing 708, which are the most characteristic features of the present invention, will be described in more detail with reference to the drawings. FIG. 8 is a schematic flowchart of the line width correction processing 704 and the line width correction processing 708, which are the most characteristic features of the present invention, FIG. 9 is a detailed flowchart of the X coordinate line width correction processing, and FIG. 7 is a detailed flowchart of a coordinate line width correction process.

The line width correction processing 704 uses the correction number data when correcting the line width, while the line width correction processing 708.
Performs exactly the same processing except that the correction number data of the previous point is used, and therefore only the embodiment of the line width correction processing 704 will be described here.

The following embodiment will be described using the upper left origin coordinate system.

First, in the branch processing 801, it is determined whether the line width of the horizontal line or the vertical line is corrected depending on whether the correction number data added to the coordinate data is "10" or more. , If it is less than “10”, the vertical line, that is, the X coordinate is the correction target, and the process moves to the branch processing 802. If it is “10” or more, the horizontal line, that is, the Y coordinate is the correction target, the branch process 80.
Move to 3. In the branching process 802, the value of WMR is referred to, and if the Y-axis shear is not generated by the affine transformation process, the correction process of the X coordinate to be corrected (see FIG.
Is performed (process 804) and the line width correction process ends, but if Y-axis shearing is occurring, the line width correction process is not executed and the line width correction process ends. Similarly, in the branching process 803, the value of WMR is referred to, and if the X-axis shear is not generated by the affine transformation process,
The Y-coordinate correction process (details will be described with reference to FIG. 10) that is the correction target is executed (process 805), and the line width correction process ends, but if the X-axis shearing is generated, the line width correction is performed. The line width correction process is terminated without performing the process.

The correction processing of the X coordinate in the processing 804 will be described with reference to FIG. First, in process 901, the X coordinate of the reference line that is the reference of the line whose line width is to be corrected is calculated. The X coordinate of the reference line can be calculated by subtracting the line width value (W) stored in the line width correction register corresponding to the correction number data from the input coordinate (XI). Next, according to the following condition 1, processing 902, 903 or processing 904,
Move to 905.

Condition 1: Rotation of 90 degrees or 270 degrees is performed by the affine transformation process (determined by referring to the value of WMR).

Processes 902 and 903 and processes 904 and 90
In both 5, the affine transformation processing of the X coordinate of the reference line calculated in the processing 901 and the line width value (W) used in the processing 901 is performed, but the affine parameters used in the affine transformation processing are different. Although AMA and ACSX are used as affine parameters in processing 902 and 903, processing 904 and
In 905, AMC and ACSY as affine parameters
To use. The affine transformation processing result of the coordinates of the reference line in the processing 902 can be expressed by equation (19) as XB ′ = AMA × XB + ACSX (19), where XB is the coordinates of the reference line and XB ′ is the affine transformation processing result.

The coordinates of the affine transformation process of the reference line in the process 904 are as follows: AMA in equation (19) is AMC, ACSX
Should be replaced with ACSY.

The line width affine transformation processing result in the processing 903 is W ′ = | AMA × W | + WCR (20) where W is the line width corresponding to the correction number data and W ′ is the affine transformation processing result. Can be expressed by equation (20).

As the result of the affine transformation processing of the line width in the processing 905, AMA in equation (20) may be replaced with AMC.

Then, the decimal point of the line width W'calculated in the process 903 or the process 904 is set to "0" (process 90).
6) It is determined whether or not the line width W ′ is 1 or less (condition 2). If it is not 1 or less, the line width W ′ is decremented by “1”. Here, the reason why "1" is not decremented when it is 1 or less is that the line width W'is not a negative number. The coordinate XB ′ and the line width W ′ of the reference line obtained by the above calculation are added (process 908) or subtracted (process 90) under the following condition 3.
9) Do.

Condition 3: 180 by affine transformation processing
Either rotation or Y-axis mirror rotation is performed (WMR
Judgment by referring to the value of).

Furthermore, it is determined whether a carry or a negative number has occurred in the calculation result of the process 908 or the process 909 (condition 4). If it is determined that a carry or a negative number has occurred, , The operation result is forcibly set to “0” (Process 9
10).

Then, under the following condition 5, the calculation result up to the above is stored in XO (process 911) and stored in WCF.
3 "is put (process 912), or the calculation result up to the above is stored in YO (process 913) and" 2 "is put into WCF (process 914).

Condition 5: Rotation of 90 degrees or 270 degrees by affine transformation processing (determined by referring to the value of WMR).

Here, the reason why the type of affine transformation process is determined by the conditions 1, 3 and 5 as described above and the line width correction process is changed according to this type will be described with reference to FIG.
3 will be used for the explanation.

FIG. 13 illustrates the difference in the line width correction processing between the case where the rotation is not performed by the affine transformation processing and the case where the rotation is performed by 90 degrees in order to explain the above reason. FIG. Here, a character that models a Gothic "L" is described as an example. In the figure, the solid line indicates the reference line and the broken line indicates the correction target line. As shown, the horizontal line (Y
In the correction process of (coordinates), the correction target line is moved in the negative direction with respect to the reference line, and the affine parameter used is AM.
It is D. Further, in the correction processing of the vertical line (X coordinate), the correction target line is moved in the plus direction with respect to the reference line, and the affine parameter used is AMA. On the other hand, 90
In the case of performing the degree rotation, since the horizontal line becomes the vertical line and the vertical line becomes the horizontal line, the moving direction of the correction target line, the affine parameter to be used, and the like are different as illustrated.

As described above, in the line width correction processing, it is necessary to change the processing according to the deformation state in the affine transformation (rotation in units of 90 degrees, mirror coordinate axis, combination thereof). In the present embodiment, the case where the upper left origin coordinate system is used as shown in the figure is explained. If the coordinate system is different, the moving direction of the correction target line from the reference line (addition of line width /
Subtraction) is not the same as this embodiment.

FIG. 10 shows a detailed flowchart of the Y coordinate correction processing in the processing 805. As shown in FIG. 10, the Y coordinate correction processing basically has the same processing content as the X coordinate correction processing described with reference to FIG. 9, and only the following four items are different. (1) Calculation process of coordinates of reference line X-axis correction process: calculated by XI + W Y-correction process: calculated by YI-W (2) Affine parameter used to calculate coordinates YB and line width W of the reference line X coordinate correction processing: AMA, AMC, ACSX, ACS
YY coordinate correction processing: AMD, AMB, ACSY, ACS
X (3) Branching condition 3 X coordinate correction process: 180 by affine transformation process
Either Y-axis rotation or Y-axis rotation is performed. Y coordinate correction processing: One or three of three conditions of X-axis rotation, 180-degree rotation, 90-degree or 270-degree rotation by affine transformation processing. (4) Branch destination under branching condition 5 The X coordinate correction process and the Y coordinate correction process are reversed.

Therefore, the description of the same processing will be omitted.

As described above, according to the present invention, the addition and subtraction of the calculated line width data with respect to the reference line is switched according to the information such as the rotation and shearing of the character in 90 degree units by the affine transformation. Even when rotating by 90 degrees or shearing in the uniaxial direction, it is possible to generate high-quality characters with uniform line widths of horizontal lines and vertical lines.

That is, when a character of "day" having a basic size of 256 dots is reduced to 16 dots, it is output as shown in FIG. 21B without correction.
When the correction is performed, the output is of high quality as shown in FIG. Even when a character is rotated or uniaxially sheared, such a high quality character can be output.

Further, in the above-described embodiments, only the embodiment for adjusting the line width of the character has been described. However, the correction number data and the line width correction register include data indicating the dot width of the space between the character lines. Is set and the position of the correction number data added to the font data stored in the font ROM is changed, the width of the white portion of the space between the character lines can be adjusted with the same processing contents as the present invention. Thus, the white portion of the space between the lines is narrower than the line width of the character, such as emphasized characters having a thick line width.

FIG. 11 shows an embodiment in which the coordinate conversion processing characteristic of the present invention described above is implemented as an LSI. Reference numeral 111 is a data path for executing a 16-bit multiply-add operation at high speed, and 112 is a control unit for controlling the data path 111. Reference numeral 116 is a control signal group from the control unit 112, and 113 is a flag group storage unit for holding the flags shown in FIG. 1
Reference numeral 17 is the WMR, 114 is a data bus for transmitting data input from the font ROM and parameters to be set in the register, and 115 is a curve of the coordinate data converted by the coordinate conversion processing in the data path 111. It is a data bus for passing to a processing unit (not shown) that performs interpolation processing.

The data path 111 is composed of a register file for storing all the registers other than WMR among the registers shown in FIG. 17, a multiplier, a shifter, an ALU, a bus and the like. In this embodiment, the control unit 11
In accordance with the values of the WMR 117 and the flag group 113, 2 places a required parameter from a register file in the data path 111 on the bus in the data path 111, and multiplies by using a multiplier in the data path 111. The data path 111 is controlled. By controlling the data path 111 as described above,
The coordinate conversion processing described with reference to FIGS. 8, 9 and 10 can be realized. The control unit 112 can be easily realized by using a microprogram, a PLA, a state machine, etc., and there is no problem in the realization.

In addition, depending on the processing contents, if a problem such as a slow processing speed occurs when the processing is executed by a program built in the state machine or the like and the processing is carried out by using a state machine or the like, as shown in FIGS. It is not necessary to perform all of the coordinate conversion processing described with reference to FIGS. 9 and 10 by the control unit 112 configured by a state machine or the like.

As an example, in the coordinate conversion process, a process of generating a control signal for loading a line width value corresponding to the correction number data from a register file in the data path 111 to a bus in the data path 111 is a state machine or the like. not,
FIG. 16 shows a block diagram of the control unit 112 when the hardware (LWR control unit) is provided and implemented. Further, FIG. 11 shows a block diagram of an embodiment of the LWR control unit 160 of FIG. 16, and the operation will be described.

In FIG. 11, reference numeral 121 is a correction number holding unit for holding the correction number data, and 122 is a front point correction number holding unit for holding the correction number data of the previous point. 123 is a selector, and 124 is 4to1
6 is a decoder (a decoder that inputs 4-bit data and outputs 16-bit data), and 125 is a select signal of the selector 123. 126A is the data path 1
A control signal for loading the value of LWR6 from the register file in 11 onto the bus in the data path 111.
6B is a control signal for putting the value of LWR7 on the bus, 126C is a control signal for putting the value of LWR8 on the bus, 126D is a control signal for putting the value of LWR9 on the bus, and 126E is This is a control signal for loading the value of LWR10 on the bus, and 126F is LWR11.
Is a control signal for loading the value of LWR12 on the bus, and 126G is a control signal for loading the value of LWR12 on the bus.
126H is a control signal for putting the value of LWR13 on the bus, 126I is a control signal for putting the value of LWR14 on the bus, and 126J is a control signal for putting the value of LWR15 on the bus. 127 is the correction number data, 128 is the correction number data of the previous point, 12
Reference numeral 9 is a control signal output by the state machine 161 of FIG.

The correction number data 127 and the previous point correction number data 128 held in the correction number holding unit 121 and the previous point correction number holding unit 122 are input to the selector 123. The selector 123 selects one of the two input correction number data by the select signal 125 and outputs it to the decoder 124. This select signal 12
Reference numeral 5 is a correction number flag FC (refer to FIG. 19) held in the flag group 113. When the select signal 125 is "0", the correction number data 127 is selected, and when it is "1", the previous point The correction number data 128 is selected.

The decoder 124 decodes the input correction number data 127 or the previous correction number data 128, and when the state machine 161 outputs the control signal 129, the decoded signal corresponds to the correction number data. The line width value is output from the register file in the data path 111 to the data path 111 as control signals 126A to 126J for loading on the bus in the data path 111. For example, the data input to the decoder 124 is "0".
In the case of "5", the control signals 126A to 126J all output "Low", and in the case of "6", the control signal 126A.
Out of 126J, only the control signal 126A is "High"
Is output. In this way, a process other than the control unit 112 can be used to generate a control signal for causing the line width value corresponding to the correction number data to be loaded on the bus in the data path 111 from the register file in the data path 111. . In this embodiment, the state machine 161 can output the control signal 126A immediately after outputting the control signal 129, which is suitable for high-speed processing.

Further, in FIG. 11, the coordinate conversion processing is performed by the LSI.
Although the description has been given only to the embodiment in which the conversion is performed, the Bezier curve interpolation processing, the line generation processing, and the filling processing can be easily realized with the same configuration, and the character generation device can be easily integrated into an LSI. The data path 111 may be shared by a plurality of processes.

[0100]

As described above, according to the present invention,
Even for a character having a plurality of line widths in a horizontal line and a vertical line in a single character, it is possible to adjust the line widths of the horizontal line and the vertical line, and it is possible to generate a high quality character.

Further, since it is not necessary to have a line width storage means (register) for each typeface, there is an effect that the storage capacity required for line width correction can be small.

Furthermore, not only the enlargement / reduction but also the rotation of 90 degrees and the uniaxial shearing of characters can be used to adjust the horizontal and vertical lines, which produces high quality characters. is there.

[Brief description of drawings]

FIG. 1 is a block diagram of an information processing device.

FIG. 2 is an explanatory diagram of a font data configuration example.

FIG. 3 is an explanatory diagram of a coordinate data configuration example 1.

FIG. 4 is an explanatory diagram of a coordinate data configuration example 2;

FIG. 5 is a schematic flowchart of the operation of the character generation device.

FIG. 6 is an explanatory diagram of an example of character and correction number data addition positions.

FIG. 7 is a coordinate conversion processing flowchart.

FIG. 8 is a detailed flowchart of line width correction processing.

FIG. 9 is a detailed flowchart of the X-coordinate line width correction processing.

FIG. 10 is a detailed flowchart of a Y-coordinate line width correction process.

FIG. 11 is a block diagram of a coordinate conversion processing unit.

FIG. 12 is a block diagram of an LWR control unit.

FIG. 13 is an explanatory diagram showing a difference in line width correction processing depending on the type of affine transformation.

FIG. 14 is an explanatory diagram of a difference in line width image during printing.

FIG. 15 is an explanatory diagram of bit width correction mode register (WMR) bit allocation.

FIG. 16 is a block diagram of a control unit.

FIG. 17 is an explanatory diagram of a register list.

FIG. 18 is an explanatory diagram of a correction number data correspondence table.

FIG. 19 is an explanatory diagram of a flag list.

FIG. 20 is an explanatory diagram of an example of an optimum line width in character size.

FIG. 21 is an explanatory diagram of a corrected character shape.

[Explanation of symbols]

101 ... Font ROM, 102 ... Character generator, 10
3 ... Input device, 104 ... RAM, 105 ... CPU, 10
6 ... Page memory, 107 ... Output device.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Kazuhisa Nishimoto 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Inside Hitachi Imaging Information Systems (72) Inventor Shinro Wakisaka 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Hitachi, Ltd., Microelectronics Equipment Development Laboratory (72) Inventor Hiroaki Shirane, 292 Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Prefecture In-house Hitachi Imaging Information Systems (72) Inventor, Hiroshi Wada Totsuka, Yokohama-shi, Kanagawa 292, Yoshida-cho, Tokyo, within Hitachi, Ltd. Microelectronics Equipment Development Laboratory (72) Inventor, Yuko Sato, 292, Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Prefecture Within Hitachi, Ltd. Microelectronics Equipment Development Laboratory (72) ) Inventor Yoshifumi Takahashi Totsuka Ward, Yokohama City, Kanagawa Prefecture Tamachi 292 address Co., Ltd. Hitachi image information in the system (72) inventor Shigeo Hayashi Tokyo Kodaira Josuihon-cho, Chome No. 20 No. 1 Co., Ltd. Hitachi, semiconductor design and development within the center

Claims (14)

[Claims] 1. A character composed of character constituent lines having a thickness,
Raster format by receiving coordinate data of a contour point having a predetermined character size and performing coordinate conversion specified by the coordinate data, and coordinate data converted by the coordinate converting means. In a character generator provided with a raster conversion means for converting into character data represented by, at least a part of the coordinate data comprises correction information regarding correction of the coordinate data at the time of the coordinate conversion, and the correction information is The correction information includes a direction of correction to be performed on the coordinate data to which the correction information is added, and thickness information of a character forming line including the coordinate data in the predetermined character size, and the correction information is a horizontal line. The outline point of one of the upper or lower sides of the character construction line, or one of the right or left sides of the vertical character construction line The coordinate conversion means is added only to the contour points on the minute line, and the coordinate conversion means, in the coordinate conversion, with respect to the coordinate data to which the correction information is added, based on the coordinate data and the thickness information, the character component line. The coordinate value of the other of the upper side and the lower side of, and the other line segment of the right side or the left side is obtained as a reference coordinate value, and the specified coordinate conversion is performed on the obtained reference coordinate value. The coordinate conversion is reflected in the thickness information, based on the coordinate-converted reference coordinate value, the reflected thickness information, and the designated type of coordinate conversion,
A character generator characterized in that it calculates coordinate values of the contour points after coordinate conversion. 2. The correction mode information according to claim 1, further comprising: correction mode information indicating a type of the coordinate conversion, wherein the correction mode information is such that at least the type of the coordinate conversion is 90 degrees rotation, 180 degrees rotation, 270 degrees. Of the coordinate data after the coordinate conversion is calculated based on the type of coordinate conversion indicated by the correction mode information. Character generator characterized by. 3. The correction information according to claim 1, wherein the correction information is replaced with thickness information of the character formation line including the coordinate data in the predetermined character size, and the character formation line including the coordinate data and other information. A character generator characterized in that it has information on the size of the space between the character forming lines. 4. The character generation device according to claim 1, wherein the correction information includes a plurality of correction numbers indicating predetermined correction direction and thickness information. 5. The character generator according to claim 4, wherein the thickness information is given by a value stored in a register which is associated with the correction number in advance. 6. The character generator according to claim 4, wherein at least one of the correction numbers indicates that the coordinate data to which the correction number is assigned is not a correction target. 7. The character generator according to claim 1, wherein when the designated coordinate conversion is reflected in the thickness information, a predetermined value is added to perform fraction processing. . 8. The correction direction according to claim 1, wherein the correction direction is a vertical direction when the character constituting line before the coordinate conversion is a horizontal line, and a horizontal direction when the character constituting line is a vertical line. The character generation device is a direction, and the thickness information is selected and designated from a plurality of predetermined thicknesses with respect to the correction direction. 9. The character generator according to claim 8, wherein the number of pieces of thickness information predetermined with respect to the correction direction differs depending on the correction direction. 10. The vertical and horizontal correction information is added only to a contour point which is a start point of a loop on the contour of the character component line, and contour points other than the start point are added to the contour points. On the other hand, correction information in one of the vertical direction and the horizontal direction is added, and the coordinate conversion means sequentially receives the coordinate data one by one, and correction information that is not added in the vertical direction and the horizontal direction. For the character generator, the correction information of the contour point received immediately before is inherited and used. 11. The method according to claim 1, wherein when the designated coordinate conversion is enlargement conversion or reduction conversion, the thickness information is W, the enlargement ratio or reduction ratio is A, and the contour point before coordinate conversion. Coordinate data of (XI, Y
I), assuming that the coordinate data after coordinate conversion is (XO, YO), for coordinate data that is not corrected, XO = A × XI YO = A × YI, and for coordinate data that is corrected, XO = A × ( XI-W) + (A * W-1) YO = A * (YI + W) + (A * W-1). 12. When the coordinate data is coordinate data to be corrected in claim 11, XO = A × (XI−W) + (A × W + WCR-1) YO = A × (YI + W) + (A × W + WCR-1) to perform fractional processing. 13. The method according to claim 12, wherein A × W + WCR <
If 1, then XO = A × (XI−W) + (A × W + WCR) YO = A × (YI + W) + (A × W + WCR) to perform fraction processing. Generator. 14. The method according to claim 11, 12 or 13,
A character generator characterized in that when the value of XO or YO is less than 0, the value of XO or YO is set to 0.