JPH04340591A - Vector font output device - Google Patents

Abstract

PURPOSE:To automatically generate a specified kind of line end parts only by specifying a line end shape and to control the size of line end parts corre sponding to the thickness of a line where the line end parts are connected. CONSTITUTION:A base part library stored in a dot data conversion device 12 is stored with skeleton data and thickness data by character constituent parts and also stored with the kinds of line end shapes. The dot data conversion device 12 generates contour data on the constituent parts with the skeleton data and thickness data and also generates contour data corresponding to the kinds of the line end shapes. In this case the dot data conversion device 12 generates contour data having thickness corresponding to the thickness of the line end parts of the constituent parts. A printer 14 adds the contour data on the line end parts to the contour data on the constituent parts and outputs them.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vector font output device that creates and outputs a vector font from skeleton data of lines constituting a character and thickness data added to the skeleton.

0002

2. Description of the Related Art Conventionally, data processing devices that perform document processing (Japanese language processing) are equipped with vector font output devices that display and print out document data in high quality. In this vector font output device, for example, font data representing the outline of a character using a coordinate string or a function string is stored in a font memory in correspondence with each character. In this way, the font data is a coordinate string or a function string, and its shape is fixed, and as shown in FIG. 22, it must also have data for the scales of the horizontal lines, which are the line end shapes.

0003

[Problems to be Solved by the Invention] As described above, since it is necessary to have data on the shape of the line ends, the amount of data becomes large and it is difficult to change the font. Therefore, as shown in FIG. 23, a method can be considered to reduce the amount of data by dividing the constituent elements of a character into parts and sharing line ends such as horizontal line stops. In the case of this method, the shape of the characters can be changed by replacing the parts at the end of the line, as shown in FIG. However, the parts at the end of the line must still be prepared, and if the thickness of the line connected to it changes even slightly, another part must be prepared, resulting in inconveniences such as an increase in the amount of data. There was also. The problem of this invention is to automatically generate a line end component of the specified type by simply specifying a line end shape, and to adjust the size of the line end component according to the thickness of the line to which this line end component is connected. The aim is to be able to control the

0004

[Means for Solving the Problems] The means of the present invention are as follows. A first storage means 1 (see the functional block diagram in FIG. 1; the same applies hereinafter) stores skeleton data and thickness data for each character component. Here, the skeleton data is, for example, a straight line passing through the center of the character component, a Bezire curve, etc.
The thickness data represents the distance from the skeleton using an n-dimensional polynomial. The second storage means 2 stores the type of line end shape for each character component. Here, the types of line end shapes include brush marks, brush stops, and the like. The first contour data creation means 3 creates contour data of a component from the skeleton data and thickness data stored in the first storage means 1. When creating contour data corresponding to the type of line end shape stored in the second storage means 2, the second contour data creation means 4 generates data according to the thickness of the line end of the component. Create contour data with size. The output means 5 is a printing device or a display that adds contour data of the line end portion created by the second contour data creation means 4 to the contour data of the component created by the first contour data creation means 3 and outputs the result. equipment, etc.

[0005]

[Operation] The operation of the means of the present invention is as follows. 1st
The contour data creation means 3 creates contour data of the component from the skeleton data and thickness data stored in the first storage means 1. Then, the second contour data creation means 4
When creating outline data corresponding to the type of line end shape stored in the second storage means 2, create outline data having a size corresponding to the thickness of the line end of the component. do. When the contour data of the component and the contour data of the line end are created separately in this way, the output means 5 adds the contour data of the line end to the contour data of the component and outputs the result. Therefore, simply by specifying the line end shape, the specified type of line end part is automatically generated, and the size of the line end part is controlled according to the thickness of the line to which this line end part is connected. be able to.

[0006]

[Embodiment] An embodiment will be described below with reference to FIGS. 2 to 21. First, the data structure of character fonts etc. will be explained. Figure 2 shows how the character font "Kan" is constructed by assembling various parts made up of character components.This character font can be roughly divided into "hen", "tsukuri", " It can be a combination of the upper part ``Ukan'', which is the upper part of a radical such as ``Kanmuri'', and the character ``Gen'', which is another character font, as parts. Upper parts and character fonts are made up of multiple base parts, etc., combined according to the shape of the character.Base parts represent character constituent elements such as a single stroke of a line or a dot, and are basic parts that cannot be further decomposed. , itself constitutes one outline font, but "Ukan" (upper part) is composed of four base parts, and the character font "Gen" also has multiple base parts (
(not shown).

FIG. 3 is a diagram for explaining the data structure of a part. A base part consists of skeleton data, fleshing data, both end shape data, etc. Skeleton data represents a line segment passing through the center of a character component, and in this embodiment, the type (1) is one in which the skeleton is represented by a straight line. (2) The skeleton is represented by a Bézire curve. (3) There is a skeleton whose skeleton is represented by a part of an ellipse. Note that there are small circles such as symbols that do not have a skeleton, and these are also treated as part of the skeleton data. The fleshing data is the distance from the skeleton using a quadratic polynomial (Ax2+Bx+c)
The contour data is created by sequentially calculating the thickness from the skeleton based on the skeleton data and thickness data based on a quadratic polynomial. The shape data for both ends defines the shape of the line end, such as brush scales, circular ends, etc. Just by specifying the line end shape, the specified type of line end part is automatically generated, and this The size of the line end component is automatically controlled according to the thickness of the line to which the line end component is connected.

The data structure of a character font consists of the part number (or other character recognition number) of the upper part or base part, placement position, enlargement ratio, etc. For example, as shown in FIG. 1" can be constructed by reading out three times by changing the length and arrangement position of the skeleton. Note that the data structure of the higher-level parts also includes part numbers, placement positions, magnification ratios, etc., similar to the data structure of character fonts.

[0009] Figure 4 shows the skeleton of the base part, horizontally 1/2 times larger.
It is transformed to 1.4 times the height, and such magnification is defined in the character font and upper parts, but in this case, the thickness and shape of both ends of the base part do not change, and the height and width of the base part do not change. Only the shape will change. Furthermore, Figure 5
is a result in which the same magnification as in FIG. 4 described above is applied to the "diagonal sweep". Even in the case of diagonal lines like this, all you have to do is apply the magnification to the length and breadth of the skeleton. In this case, since the thickness is always defined as the distance in the direction perpendicular to the skeleton, a diagonal line that looks like it has been rotated in a pseudo manner is obtained.

FIG. 6 is a block diagram showing the overall configuration of a character font output device. The character font output device includes a parts placement device 11, a dot data conversion device 12, a bitmap print buffer 13, and a printer 14.
When the parts placement device 11 receives a character code to be printed along with a print command, it starts operating, sequentially extracts the base parts that make up the character, and sends the part number, placement position, and magnification data to the dot data conversion device 12. give. The dot data conversion device 12 calls the base part corresponding to this part number, changes the skeleton data by applying a magnification to the skeleton, generates contour data by referring to the fleshing data, etc., and stores it in the bitmap print buffer. 13 is performed for each base part at a designated placement position. Note that the contour data filling process is performed by the dot data converter 12 after the contour data is generated.
executed by The printer 14 is a thermal printer that prints out the dot data in the bitmap print buffer 13 line by line.

FIG. 7 is a block diagram showing the configuration of the parts placement device 11. The print code register 11-1 is used to set the character code to be printed, and this character code is passed to the search code register 11-2. The search code register 11-2 stores the character code from the print code register 11-1 or the parts code (or character code) from the main control unit 11-3, and searches the library based on this character code or parts code. Address index 11-4 is font data library 11
The first address corresponding to the character or part in -5 is indexed. The font data library 11-5 stores data structure tables corresponding to various characters and upper-level parts. ” has a corresponding data structure table. Library address pointer 11
-6 is a pointer to which the start address obtained by referring to the library address index 11-4 is set, and the corresponding data structure table of the font data library 11-5 is specified according to this value. The parts pointer 11-7 is a pointer that sequentially specifies the definition contents of the data structure table specified by the value of the library address pointer 11-6 in units of parts from the beginning.
The data read from the controller 5 is taken into the main controller 11-3. The pointer stack memory 11-8 is a memory for saving the values of the library address pointer 11-6 and the parts pointer 11-7 as necessary. Main control unit 11
-3 is a code register 11 that searches for the character code or part code if the data read from the font data library 11-5 is a character or upper part.
-2, but if it is a base part, data regarding this base part (part number, placement position, magnification) is given to the dot data conversion device 12.

FIG. 8 is a block diagram showing the configuration of the dot data conversion device 12. This dot data conversion device 1
2 are search code register 12-1, library address index 12-2, library address pointer 1
2-3, a parts pointer 12-4, a base parts data library 12-5, and a main control unit 12-6.
The part number is input as a search code from , and the placement position and magnification are input from the parts placement device 11 to the main control section 12-6. library address index 12
-2 is the base parts data library 12 based on the parts code set in the search code register 12-1.
The first address corresponding to the corresponding part in -5 is indexed. The base parts data library 12-5 stores data structure tables corresponding to various base parts, and the corresponding data structure table is specified according to the value of the library address pointer 12-3. In accordance with the value -4, the definition contents of this designated data structure table are sequentially designated and read out in skeletal units from the beginning. The data read from the base parts data library 12-5 by this is transferred to the main control unit 12-6.
sent to. The main control unit 12-6 generates contour data based on the base parts data and the magnification data from the parts placement device 11, expands it to the bitmap print buffer 13, and increments the value of the parts pointer 12-4.

FIG. 9 shows the font data library 11-5.
This shows the data structure table stored in . This data structure table is provided for each character and upper part, and the data structure table corresponding to one character or upper part consists of N blocks (N=1, 2...), and one block is 6 bytes, so N The data amount is obtained by adding the header (2 bytes) to the block x 6 bytes. The header variable (16 bits) defines the character, the number of blocks that make up the upper part (7 bits), etc. For example, the character ``Jin'' consists of ``Ninben'' (upper part) and two horizontal lines (
"Ninben" is a 2-block composition consisting of two types of base parts, namely a diagonal sweep and a vertical line, and the number of blocks is defined as a header variable.

[0014] Each block data has a first variable, a second variable...
...consists of a fifth variable. As shown outside the table in Figure 9, the details of the first variable define that when the most significant bit of the first variable (16 bits) is "0", another character font should be called, and the remaining 15 bits are This becomes a character identification code, and the remaining 15 bits can represent all Japanese character codes. Defines that a part should be called when the most significant bit of the first variable is “1”, and when the second most significant bit is “0”, high-order parts such as “hen” or “tsukuri” are called, and when it is “1”, a part is called. Defines what normal one-stroke parts (base parts) should be called. Note that since there are at most 200 types of base parts, the number of bits is sufficient. The second and third variables define the placement position of the parts,
The second variable indicates the X coordinate value, and the third variable indicates the Y coordinate value. The fourth variable and the fifth variable define the magnification of the part, the fourth variable indicates the X-axis direction, and the fifth variable indicates the Y-axis direction.

FIG. 10 shows the base parts data library 1.
2-5 shows a data structure table stored in 2-5. Here, the base part is the smallest character component that cannot be further decomposed, and in addition to one skeleton, there are also base parts that consist of two or more skeletons. For this reason, the data structure table consists of N blocks (N=1, 2...) corresponding to the number of skeletons.
One block is 4 to 14 bytes, so N blocks x
The data amount is 4 to 14 bytes plus the header (2 bytes). The number of blocks (number of skeletons) is defined in 7 bits of the header variable (16 bits).

The amount of data for each block data differs depending on the shape of the skeleton. In other words, if the skeleton is a straight line, the amount of data will be from the 1st variable to the 10th variable, and if the skeleton is a Bézire curve, the amount of data will be from the 1st variable to the 14th variable.
In the case of a part of an ellipse, the amount of data is from the first variable to the 12th variable, and in the case of a small circle, it is the amount of data from the first variable to the fourth variable. The first variable (8 bits) is its lower 2
The shape (type) of the skeleton is defined by the bit, and its value is “00”.
” indicates a straight line, “01” indicates a Bézire curve, “10” indicates a part of an ellipse, and “11” indicates a small circle.The other bits in the first variable, that is, the upper 6 bits, indicate the shape of the line end. The upper 3 bits of these 6 bits indicate the shape of the start end, and the other 3 bits indicate the shape of the end end.The line end shape of the first variable is as shown outside the table in FIG. 000" indicates a straight line cut, "001" indicates a semicircular cut, and..."111" indicates a brush stroke at the end of a vertical line. In this way, eight types of line end shapes are prepared in advance, and the specific shapes are shown in the figure. 11.

The second to fourth variables indicate fleshing data representing the thickness when the skeleton shape is a straight line, Bezire curve, or part of an ellipse, and indicate the size when the skeleton shape is a small circle. When the skeleton is a straight line, the fifth to eighth variables are skeleton data indicating the position coordinates of the starting point and ending point of the line segment. When the skeleton is a Bézire curve, the fifth to twelfth variables are skeleton data and indicate the position coordinates of the starting point, first control point, second control point, and end point of the line segment. If the skeleton is part of an ellipse, the fifth to tenth variables
The variables indicate the center coordinates, radius, starting angle, and ending angle of the ellipse. The 9th and 10th variables when the skeleton is a straight line, the 13th and 14th variables when it is a Bézire curve, and the 11th and 12th variables when it is part of an ellipse are the straight line cuts at the starting point, respectively. The correction value indicates the straight line cut correction value at the end point, and the remaining 6 bits excluding the upper 2 bits indicate the correction value (+
31 to -32) are defined. This correction value indicates how much the cut surface is inclined when the line end is cut in a straight line, and the line end component is added along this cut surface. On the other hand, the upper two bits are control data for changing the font.
This control data is used when changing Mincho font to Gothic font.

Next, the operation of this embodiment will be explained with reference to FIGS. 12 to 21. Now, the character "Jin" is selected as a printing target in the print code register 11- of the parts placement device 11.
1 is input. Here, FIG. 12 shows the character "Jin" assembled according to the way the parts are held in this example. position,
It is made up of data that defines the magnification.

First, when the character code of "Jin" is set as a search code in the search code register 11-2, the operation according to the flowchart of FIG. 13 is started. That is, when the character code "Jin" is set in the search code register 11-2, the font data library 1
The first address specifying the data structure table of the character font "Jin" stored in 1-5 is indexed from the library address index 11-4 and set in the library address pointer 11-6 (step A1).
. Next, the initial value "1" is set to the parts pointer 11-7, and the first part data is specified (step A2
). As a result, "ninben", which is the first part of "jin", is read out from the font data library 11-5 (step A3). Then, the main control unit 11-3 checks whether the part data read from the font data library 11-5 is another character font or a higher-order part (step A4); in this case, the first part is a higher-order part. Therefore, the process advances to step A5, and the library address pointer 11-6 and parts pointer 11-7 are
The value of is saved in the pointer stack memory 11-8. Next, when the main control unit 11-3 sets a code indicating "Ninben" in the search code register 11-2, the leading part that specifies the data structure table stored in the font data library 11-5 corresponds to this code. The address is indexed from the library address index 11-4 and set in the library address pointer 11-6 (step A1). Next, in order to specify the first part of this data structure table, the initial value "
1" (step A2). Here, since "Ninben" is composed of two lines, a diagonal line and a vertical line, the first part, the diagonal line, is first read out from the font data library 11-5 (step A3).
). Since this diagonal line is a base part that cannot be further disassembled, this is detected in step A4 and the process proceeds to step A6, where the data regarding this base part (part identification code, placement position, magnification) is sent to the dot data converter. Send on 12th.

Here, the dot data converting device 12 is shown in FIG.
It operates according to the flowchart of No. 4. Now, when the parts identification code of the diagonal lines that make up "Ninben" is set in the search code register 12-1, the data structure table stored in the base parts data library 12-5 is specified in correspondence with this code. The first address to be read is indexed from the library address index 12-2 and set in the library address pointer 12-3 (step B1). Next, in order to designate the first skeleton of this base part, an initial value "1" is set in the parts pointer 12-4 (step B2). In addition, this “Ninben”
The diagonal line consists of one skeleton, but some base parts consist of two or more skeletons even in one stroke, so first specify the first skeleton. As a result, when base part data regarding the first skeleton is received from the base parts data library 12-5 (step B3
), the main control unit 12-6 checks whether a short vector has occurred based on the above-mentioned font change control data (step B4). Note that, as will be described later, generation of short vectors is not required when changing the font from Mincho to Gothic. Normally, generation of short vectors is required, so proceed to step B5 and generate short vectors. A process is performed to generate the data and write it into the bitmap print buffer 13.

FIG. 15 shows the conversion to this short vector.
7 is a flowchart showing buffer write processing. Note that since the base part consists of skeleton data, fleshing data, and both end shape data, it is converted into three types of outline data accordingly. That is, the first is the contour (main body) of the fleshed-out shape around the skeleton, the second is the part that shows the shape of the line end at the starting point of this main body, and the third is the shape of the line end at the end point of the main body. It is a part. In this way, the base part automatically creates three pieces of outline data individually and overlays them to create the bitmap print buffer 1.
3. At this time, outline data is first generated for the main body, and then outlines for the line end portions are generated. Now, as shown in Figure 11, "Ninben" consists of a main body (diagonal line) and a line end part (brush mark at the beginning of the diagonal line), so first, refer to the skeletal shape and its variables (step C1), and then The contour vector data is generated using the image data (step C2), and written into the bitmap print buffer 13 (step C3). Next, the line end shape and variables are referred to (step C4), contour vector data is generated using it (step C5), and written to the bitmap print buffer 13 (step C6).

[0022] Figure 16 shows diagonal lines constituting “Ninben”;
FIG. 17 shows how the starting brushstroke component is developed on the bitmap print buffer 13. First, as shown in FIG. 16, a Bezire curve is used from S0 to S1 as a skeleton, and a quadratic equation is used as an equation representing the thickness. That is, S0 is the fulcrum (parameter variable i=0), S1 is the end point (parameter variable i=1), and the interval therebetween is expressed as V=(VX, VY). Note that VX and VY are functions of i, and VX=P
i3+qi2+ri+s, VY=ti3+ui2+wi
+z. Then, Ai in the direction perpendicular to the skeleton with respect to i
The point moved by 2+Bi+C becomes the contour. In addition, A
, B, and C are as shown in Figure 10, variable 2=A, variable 3=B
, variable 4=value of C. Here, for each i, if the points on both sides of the skeleton are X and Y, then

[Formula]

It can be expressed as Therefore, contour data (short vector format) is obtained by obtaining X and Y while changing i at intervals that match the Bezire and A values of the skeleton. Note that the feeling includes many vertical straight lines and horizontal straight lines, in this case, if it is a vertical straight line, p = q = r = 0 and t = u = 0, and if it is a horizontal straight line, t
=u=w=0 and p=q=0, or there is no Bezire control point. Furthermore, in that case, A=B=0 unless the product is very particular about quality. Therefore, it is only necessary to draw a straight line whose thickness is clearly represented by c, and there is no need to execute the above calculation formula. Furthermore, when there are many horizontal lines and when they are reduced to a small size, the thickness of the lines may vary due to quantization errors in coordinate calculations, but this time, the thickness of the lines will be reduced in the same way. Because the same thickness is clearly displayed, it is possible to prevent variations in thickness, and because the straight line is simply drawn repeatedly, it is much faster to draw than the conventional method (edge detection → filling). becomes possible.

Next, regarding the development of line end parts, since the type of line end parts is only indicated, they must be generated automatically. In the example shown in FIG. 17, a "010" type part is attached as shown outside the table in FIG. In addition, in the data structure table of the base part, the upper 6 bits of the first variable are the line end information, and the others are the main body information, so in the case of Fig. 17, the starting edge is the stroke of the diagonal line, and the ending edge is the thickness. is 0, so it is a straight line cut (a semicircular cut is also fine), and the skeleton is Bezire, so the first variable is "01000001"
becomes. The values of P1' and P2' are calculated from the upper 6 bits of these and the contents of the 13th variable. In addition, since P1 and P2 are Bezire starting points, X when the parametric variable = 0,
The vertical direction of the center line is determined from the differential value of the parameter of each function Y, and it is determined by moving the starting end of the center line in that direction by the thickness of the starting end. P1', P2' are the values of the parametric variables by the value of the 13th variable - (P1'), + (P
2') and calculate it. Next, P1'
By defining the contour vector as a straight line whose starting and ending points are P2' and P2', whose thickness gradually increases from 0, and which is connected to a semicircle at the ending point, the "010" type line end part is It can be treated as a base part whose first variable is "00000100", and its outline data can be developed on the bitmap print buffer 13.

As a result, bitmap print buffer 1
In 3, diagonal lines with brush marks forming the “Ninben” are developed. When the short vector generation/buffer writing process (step B5 in FIG. 14) is completed, the process proceeds to step B6, where "1" is added to the part pointer 12-4 to update its value, but "Ninben" Since the diagonal line consists of one skeleton, the end of the skeleton is detected in step B7, and the dot data conversion device 12 ends the expansion operation for the diagonal line.

Next, the parts placement device 11 proceeds to step A7 of FIG. 13, and adds "1" to the parts pointer 11-7 to update its value. As a result, the value of the part pointer 11-7 becomes "2", but since "Ninben" consists of two parts, the end of the part is not detected in step A8, and the process returns to step A3, where the second part of "Ninben" is Load the vertical line part. In this case, since the vertical line is the base part, step A4 to step A
6, the base part data regarding this vertical line is sent to the dot data converter 12. As a result, the dot data converter 12 generates a vertical line outline font and develops it on the bitmap print buffer 13. Next, the process proceeds to step A7, where the parts pointer 11-
When the value of 7 is updated, the value exceeds the number of parts "2" in "Ninben", so the end of parts is detected in step A8.

After "Ninben" has been expanded onto the bitmap print buffer 13 in this way, it is checked whether the data has been saved in the pointer stack memory 11-8 (step A9). Since the pointer value specifying the first part of the character has been saved in the pointer stack memory 11-8, it is returned from the pointer stack memory 11-8 (step A10).
, updates its value (step A7). by this,
The second part of “Jin” is specified. In this case, the second part is the upper horizontal line of the two upper and lower horizontal lines, and since this horizontal line is also a base part, its contour data is developed on the bitmap print buffer 13. Next, the third part is specified and the lower horizontal line is developed on the bitmap print buffer 13, but in this case, the same base part is printed on the bitmap print buffer 1 with the skeleton length and placement position changed.
It is expanded on 3.

As described above, the character font "Jin" consists only of data that defines the part identification code, placement position, and magnification corresponding to one upper part and two base parts, so it cannot be used as before. The amount of data can be significantly reduced compared to the case where the outline of a character is stored as a coordinate string.

FIG. 18 shows that another character "iii" is constructed by arranging three character fonts "i" horizontally. Similarly, "Kan" shown in FIG. 2 can be constructed using "Ukan" and another character font "Gen". In this way, in this embodiment, other character fonts can also be used as parts, making it possible to significantly reduce the overall amount of data.

On the other hand, since we have separate data indicating the thickness of the skeleton, which is the center of the lines that make up the character, and the thickness from the skeleton that fleshes out this skeleton, and we express the thickness using a quadratic expression, The thickness is constant, and there is no variation in the thickness of horizontal and vertical lines when reduced, making it possible to output high-quality fonts faster than conventional dot fonts. Furthermore, since the skeleton and thickness are independent data, it is possible to change the shape without changing the thickness, or change the thickness without changing the shape.

On the other hand, base parts, which are character constituent elements, are stored in the form of line bodies and line end parts, and line end parts are automatically generated just by defining their shape, and the size of this line end part is Since it depends on the thickness of the line body to which it is connected, there is no need to prepare line end parts according to their types and sizes, and it is possible to delete the overall amount of data. Become. In addition, since everything is a contour vector, it is easy to create parts. Next, a case will be described in which the Mincho font is changed to the Gothic font.

FIG. 19 shows a specific example of a Mincho font. By using this Mincho font as a basic font, and keeping the thickness constant and making both ends of the lines circles, a Gothic font as shown in FIG. 20 can be obtained. Can be done. In this case, in the base parts data library 12-5 mentioned above, the lower two bits of the first variable are "1".
1", that is, when the shape of the skeleton is a straight line, Bézire curve, or part of an ellipse, the second and third variables are set to "0", the fourth variable is set to a constant value, and the first variable is set to "0". A Gothic font can be obtained by replacing the line end shapes "010", "110", etc. with "001", but as shown in Figure 20
In the Gothic typeface shown in , unnatural lines (indicated by hatching in the figure) occur where the vertical line of the letter "a" is placed and where the horizontal line jumps.

In order to delete such a line, that effect is set in the font change control data described above. That is, in FIG. 10, TS (straight line cut correction value at the starting point), T
The upper two bits of the variable that defines E (straight line cut correction value at the end point) are control data for changing the font, and if this data is used to declare that short vectors do not need to be generated, step B4 in FIG. This is detected in step B5, and the short vector generation process in step B5 is skipped. Therefore, if the horizontal line of "A" consists of three skeletons, that is, the first skeleton...the part where the brush is placed, the second skeleton...the part of the horizontal line, and the third skeleton...the bouncing part, then about the third skeleton. It is set that short vectors do not need to be generated. Similarly, if the vertical line of "a" consists of two skeletons, that is, the first skeleton...the part of the brush stroke, and the second skeleton...the vertical line part, then the generation of short vectors is set to be unnecessary for the first skeleton. I'll keep it. As a result, well-balanced Gothic characters without unnatural lines can be obtained as shown in FIG. 21. That is,
When using one typeface as a base and processing its contour data to obtain characters from another typeface, it is possible to erase the unnatural lines of the original typeface, making it possible to print well-balanced characters. can.

Note that in the above embodiment, 2 is used to represent the thickness.
Although the following equation is used, it may also be a polynomial of degree 3 or higher, in which case it is possible to change the thickness more widely.

[0034]

[Effects of the Invention] According to the present invention, simply by specifying the line end shape, the specified type of line end part is automatically generated, and the line end part is automatically generated according to the thickness of the line to which this line end part is connected. Since the size of the line end parts can be controlled, the amount of data can be significantly reduced.

[Brief explanation of the drawing]

FIG. 1 is a functional block diagram of the present invention.

FIG. 2 is a diagram showing a specific example of constructing the character "Kan" by assembling various parts in the embodiment.

FIG. 3 is a diagram showing an overview of the data structure of base parts and upper-level parts.

FIG. 4 is a diagram showing an example of deforming the skeleton of the base part.

FIG. 5 is a diagram showing another example in which the skeleton of the base part is deformed in the same way as in FIG. 4;

FIG. 6 is a block diagram showing the overall configuration of a character font output device in an embodiment.

FIG. 7 is a block configuration diagram of the parts placement device 11 shown in FIG. 6.

FIG. 8 is a block configuration diagram of the dot data conversion device 12 shown in FIG. 6.

[Figure 9] Font data library 11- shown in Figure 7
5 is a diagram showing a data structure table stored in 5.

10 is a diagram showing a data structure table stored in the base parts data library 12-5 shown in FIG. 8. FIG.

FIG. 11 is a diagram specifically showing the line end shape shown in FIG. 8;

FIG. 12 is a diagram showing how the character "Jin" is constructed by assembling parts.

FIG. 13 is a flowchart showing the operation of the parts placement device 11.

FIG. 14 is a flowchart showing the operation of the dot data conversion device 12.

FIG. 15 is a flowchart showing details of the short vector generation/buffer write process shown in FIG. 14;

FIG. 16 is a diagram for explaining the processing content of FIG. 15, taking as an example the case of developing diagonal lines.

FIG. 17 is a diagram for explaining the processing contents of FIG. 15, taking as an example a case where a diagonal line starting brush holder component is developed.

FIG. 18 is a diagram showing a specific example where another character is constructed using the character itself as a part.

FIG. 19 is a diagram specifically showing the character form of the Mincho typeface.

FIG. 20 is a diagram showing an example in which the Mincho typeface characters in FIG. 19 are changed to Gothic typeface characters.

21 is a diagram showing an output example of a Gothic font obtained by removing unnatural lines from the Gothic font shown in FIG. 20; FIG.

FIG. 22 is a diagram for explaining a conventional example.

FIG. 23 is a diagram for explaining the problem to be solved by the invention.

[Explanation of symbols]

11 Part placement device 12 Dot data conversion device 12-2 Library address index 12-5
Base parts data library 12-6 Main control unit 13 Bitmap print buffer 14 Printer

Claims (1)

[Claims] 1. A first storage means for storing skeleton data and thickness data for each character component; a second storage means for storing a type of line end shape for each character component; a first contour data creation means for creating outline data of a component from the skeleton data and thickness data stored in the storage means; and a type of line end shape stored in the second storage means. a second contour data creation means for creating contour data having a size corresponding to the thickness of the line end of the component when creating the contour data correspondingly; and the first contour data creation means. A vector font output comprising: output means for adding and outputting line end contour data created by the second contour data creation means to the contour data of the component created by the means. Device.