JPH0715625B2 - Character figure generation processing method - Google Patents

Description

The present invention relates to a character / graphics generation processing method for generating character / graphics of various typefaces.

[Conventional technology]

As a conventional method for generating a character graphic, “character information described by a line graphic showing the skeleton of the character graphic” and FIG. 9 (Yasaburo Kuwayama, “Character Design”, Graphic Publishing Co., 1971)
"Shape information of basic elements constituting a typeface of a character and a figure" is stored in advance, and "the shape of the constituent element is changed for each character type based on the font information". There is a method of generating a character figure by combining them (see Japanese Patent Application No. 60-202476). In this method, the typeface of the character / graphic used in principle depends on the shape of the "basic element" stored in advance, and therefore the typeface of the generated character / graphic is determined if the shapes of the constituent elements are determined.

[Problems to be solved by the invention]

Conventionally, since the above-mentioned method was adopted, it was not possible to generate a character graphic of a typeface other than that which was stored in advance. In order to generate a character graphic of another typeface, it is necessary to newly design and register the shape of the constituent element corresponding to the typeface. Special knowledge of character / graphic design regarding the combination of constituent elements to generate a graphic was required, and specialized knowledge for registration operation was also required. Therefore, it was difficult for ordinary people without specialized knowledge to change these components.

SUMMARY OF THE INVENTION It is an object of the present invention to solve these problems of the prior art and to easily generate and use a character graphic of any typeface or shape in document creation and the like.

[Means for solving problems]

To achieve this object, the present invention inputs a character graphic having a typeface feature to be generated as a typeface shape sample character graphic, extracts the constituent elements of the input character graphic, and determines the geometrical characteristics of the constituent element. And means for changing the shape of the constituent element stored in advance based on the geometrical characteristics of the extracted constituent element.

[Action]

Since the present invention includes the means, by inputting a sample character graphic having a typeface feature to be generated, the shape of the component stored in advance by using the means is changed to a component having the typeface feature to be generated. The shape is changed, and the character and figure of the typeface desired to be generated are automatically generated using the constituent elements changed in shape.

Example of Invention

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

FIG. 1 is a block diagram of an embodiment of the present invention.

Reference numeral 201 is a constituent element extraction unit, 202 is a constituent element feature extraction unit, 203 is a constituent element shape changing unit, 204 is a character / graphics generation unit, 205 is a constituent shape storage unit, and 206 is a character / graphics configuration information storage unit.

The component shape storage unit 205 stores in advance the shapes of the respective components that form the character graphic. The storage unit 20
An example of the data structure of 5 is shown in FIG. 301 is an identifier assigned to each constituent element (hereinafter referred to as ID), 302 is a center line representing the skeleton of the constituent element (hereinafter referred to as skeleton line, fourth
(See the figure), the number of representative points that represent the shape, 303 is the coordinate position that represents the position of the representative point, 304 is the width of the component at the representative point (see FIG. 4), and 305 and 306 are the end points (skeletons) of the respective components. It is a point attribute value that represents the shape of the start point and end point of the line. Table 1 shows an example of the meaning of the point attribute value.

The character / graphic configuration information storage unit 206 stores in advance information necessary for generating a character / graphic by combining the elements forming the character / graphic. FIG. 3 shows an example of the data structure of the storage unit 206. 401 is an ID given to each character / graphic, 402 is the number of constituent elements that make up the character / graphic, 403 is a constituent element ID that makes up the character / graphic, and 404 is when generating the constituent
Coefficient for transformation such as enlargement or reduction in the coordinate axis direction, 4
Reference numeral 05 is a composite position that indicates the coordinate position where the constituent elements are pasted after being generated and transformed.

The constituent element extraction unit 201 separates and extracts constituent elements patterns constituting the character graphic from the input sample character / graphic pattern for each constituent element. The operation of the component extraction unit 201 will be described with reference to FIG. In FIG. 5, 601 is the input sample character / graphic pattern, 602, 603, 604.
Are the constituent element patterns separated and extracted by the constituent element extracting unit 201 (in FIG. 5, the character / graphic pattern and the inside of the constituent element pattern are not painted for the sake of convenience). From the input sample character / graphic pattern 601 to the component pattern 60
Regarding the method of separating into 2,603,604, a known method, for example, a method using a direction code (Journal of the Institute of Electronics and Communication Engineers, vo
l.J68-D, No.9, PP1680-1687, 1985)). First, the sample character / graphic pattern 601 input by the method or the like is separated into the respective constituent element patterns 602, 603, 604. Next, regarding the separated component patterns 602, 603 and 604, the ID corresponding to the component is determined. There are various methods for this determination, but the following methods (1), (2), and (3) can be considered.

Method (1) The component extraction unit 201 is provided with a character / graphic recognition unit, a component shape generation unit, a pattern matching unit, a sample character / graphic configuration information storage unit, and a sample character / graphic component shape storage unit in advance.

Here, the storage data structure of the sample character / graphics configuration information storage means and the storage data structure of the sample character / graphics component shape storage means may be the same as the data structure shown in FIG. 3 and the data structure shown in FIG. 2, respectively. . The component ID used internally is the same as the component ID used in 206 and 205. An example of a detailed configuration diagram of the component extraction unit 201 in this case is shown in FIG. The operation of FIG. 6 is shown below.

First, a component extraction control unit (hereinafter simply referred to as a control unit) 706
Is the sample character graphic pattern 601 entered as described above.
From each component pattern 602, 603, 604.

Next, the control unit 706 passes the data of the sample character / graphic pattern 601 to the character / graphic recognition unit 701. The character / figure recognizing means 701 recognizes the pattern data 601 by using a recognition dictionary that it has in advance, and passes the corresponding character / figure ID (denoted as Gj) to the control unit 706. The control unit 706 searches the sample character / graphic configuration information storage means 704 using Gj as a key, and reads out the component ID (Ej1, Ej2, ...) Corresponding to Gj from the data structure as shown in FIG. Next, the control unit 706 reads each component ID
, The sample character graphic component shape storage means 705 is searched using the component ID (denoted as Ei) as a key, and the component shape information (first data) corresponding to Ei is calculated from the data structure shown in FIG. In FIG. 2, the data from 302 to 306 corresponding to Ei) is read and the information is passed to the component shape generation means 702. The constituent element shape generation means 702 generates the shape data of the constituent element Ei by the method described later (see the description of the operation of the character / graphics generation section 204), and passes the generated shape data to the control section 706. In this way, the control unit 706 generates the shape data of all the constituent element IDs (Ej1, Ej2, ...) Corresponding to Gj, and separates the shape data of each constituent element from the sample character graphic in advance. The respective data of the element patterns 602, 603, 604 are passed to the pattern matching means 703 to perform pattern matching, and the respective component element patterns 602, 603,
604 is associated with the component ID (Ej1, Ej2, ...).

The sample character / graphics configuration information storage unit 704 and the sample character / graphics component shape storage unit 705 may be the character / graphics configuration information storage unit 206 and the component shape storage unit 205 themselves.

Method (2): The component extraction unit 201 is provided with a pattern matching means, and the sample character / graphic pattern pattern 60 to be input to the 201 is input.
The constituent element ID constituting 1 and the skeleton line shape of the constituent element are associated and given in advance. The skeleton line shape is extracted by thinning each of the separated constituent element patterns 602, 603, 604, and the pattern matching between the skeleton line shape given in advance by the pattern matching means and the skeleton line shape obtained by thinning the 602, 603, 604, etc. Then, each component pattern 602, 603, 604 is associated with the component ID.

Method (3) The constituent element extracting unit 201 is provided with a conversation means, and the constituent element patterns 602, 603, and 604 are interactively given respective constituent element IDs from the outside.

The constituent element extracting unit 201 obtains constituent element IDs corresponding to the constituent element patterns 602, 603, and 604 by the methods (1), (2), and (3) described above, and the constituent element ID for each of these constituent element patterns. And pattern data (corresponding pattern data among 602, 603, and 604) are associated with each other and passed to the component feature extraction unit 202.

The component feature extraction unit 202 has a pattern (corresponding to the pattern shown in FIG. 4) corresponding to the input component ID (denoted as Ei).
The shape feature information of the component is extracted from the data, and the component ID (Ei) and the extracted shape feature information of the component are passed to the component shape changing unit 203. Although various types of shape feature information of the constituent elements are conceivable, for convenience of explanation, here, if the shape data is the same as the shape data of the constituent elements stored in the constituent shape storage unit 205 as the shape characteristic information, that is, the shape of the skeleton line. Taking the example of the case of a point attribute representing the number of representative points representing the, the coordinate position representing the position of the representative point, the width of the component at the representative point, and the shape of the end points (start and end points of the skeleton line) of the component , Which will be described below (in other cases, the shape feature information needs to be converted, but the description is omitted). The extraction of such characteristic information can be realized as follows using a known method (see FIG. 4).

The skeleton line 503 is obtained from the component pattern 501 by thinning or the like. Next, the feature points (end points, points with large curvature, etc .; Pi1, Pi2, Pi3 in Fig. 4) of the skeleton line are obtained by line tracing, and these feature points are used as representative points representing the shape of the skeleton line. Each coordinate position (Xi1, Yi1), (Xi2, Yi2),
(Xi3, Yi3) and the representative score (Ni = 3) are calculated. Then, the widths (Wi1, Wi2, Wi3) of the constituent elements at the representative point are obtained. This is based on the intersections of the normal to the skeleton line at the representative point and the outlines of the components (hereinafter referred to as contour lines; 502) (Qi1, Ri1, ... respectively), and the representative point and the intersection point. It is calculated by taking the distance of as the width. As the end point shape, the point attributes as shown in Table 1 are given corresponding to the shape pattern of the curve Qi1Ri1 (or the curve Qi3Ri3). This can be easily realized by using a pattern matching means which internally has an end point shape pattern corresponding to a point attribute.

It should be noted that the coordinate position of the representative point and the width of the component at the representative point are normalized in advance (for example, in the coordinate position, the starting point is the coordinate origin, and the size is the length of the skeleton line is L0). Normalize by scaling etc.).

The component shape changing unit 203 receives the component ID (Ei) and the shape feature information of the component from the component feature extraction unit 202, and shows the component ID and the shape feature information in FIG. It is registered in the changed component shape storage table 801. 7th
In the figure, 802 is the registered component shape information,
The data structure may be the same as that shown in FIG.

In addition, the component ID input from the component feature extraction unit 202
There is also a method of directly rewriting the storage area of the corresponding component ID of the component shape storage unit 205 without registering the shape feature information of the component with the changed component shape storage table 801, but this can be easily realized. Yes, and the description is omitted.

The character / graphics generation unit 204 has a constituent element shape generation means and an area filling means by interpolating between representative points, and stores the character / graphics configuration information corresponding to the input character / figure code (character / figure ID). The relevant character / graphic configuration information (number of components, component ID, coefficient, composition position) is read from the unit 206, and the shape of the component required based on this information is transmitted to the component shape changing unit 203. Shape memory 2
It reads from 05, generates and transforms the constituent elements, and combines the generated constituent elements to generate character graphics. The specific operation is shown below.

First, as shown in FIG. 1, an input character / graphic code (hereinafter referred to as a character / graphic ID) is Gj. Since the character / graphics generation unit 204 has a character / graphic ID of Gj, the character / graphic structure information storage unit 206 uses the Gj as a key as the corresponding configuration information, as shown in FIG. The element ID (Ejk), the coefficient (Lxjk, Lyjk), and the combined position (xjk, yjk) (k = 1, 2, ..., Nj) are read.

A component element ID is sent to the component element shape changing unit 203 for each read component element ID, and the component element shape changing unit 203
Then, using the component ID as a key, the changed component shape storage table 801 shown in FIG. 7 is searched first in the order of the component shape storage unit 205, and then the component ID is searched from the position where the component ID is first found. The shape information of the constituent elements is read and passed to the character / graphics generation unit 204. The case where the component ID is Ei will be described as an example. The character / graphics generation unit 204 uses the Ei as a key through the component shape change unit 203 as the corresponding shape information and shows the number of representative points Ni and the coordinate position (Xis, Yis) of each representative point Pis shown in FIG. , The width of the component at the representative point Wis (s =
1,2, ..., Ni), and each point attribute that represents the end point shape at both end points (start point, end point) of the component.

An example in which the character / graphics generation unit 204 generates the shape of the component from the read coordinate position and the point attribute of the representative point using the component shape generation means will be described with reference to FIG. In the example of FIG. 4, the number of representative points is Pi1, Pi2, and Pi3 = 3, and the widths of the constituent elements at the representative points are Wi1 and Wi, respectively.
2, Wi3, the point attribute at both end points Pi1 and Pi3 of the component is 1 (round end point, small curvature) shown in Table 1 above. First, a curve (for example, a spline curve) is interpolated between Pi1, Pi2, and Pi3 to obtain a skeleton line 503. Next, at each of the representative points Pi1, Pi2, Pi3, the distance in the normal direction of the skeleton line 503 is Wi1, Wi2, Wi3, respectively.
Points away from each other (Qi1, Ri1, Qi2, Ri2, Qi3, Ri shown in Fig. 4)
3) ask. In addition, Qi1, Qi2, Qi3 are
Qi2Qi3 and the curve Ri1Ri2Ri3 are obtained in the same manner. The end point shape can be easily generated by previously defining the end point shape generation method corresponding to Table 1 above. For example, P
The end point shape on the i1 side is generated as a circular arc with curvature r1 with Qi1 and Ri1 as the end points of the chord (the Pi3 side is also generated in the same way).
As described above, the shape of the contour line 502 of the component can be generated.

The character / graphics generation unit 204 fills the generated inner area of the contour line 502 with the area filling means to generate the constituent elements.

In the character / graphics generation unit 204, the components generated as in the example of FIG. 4 are transformed as follows. Component ID is Ejk
In the case of, the coefficient Lxjk is multiplied in the X-axis direction, which is the coordinate axis, and Lyjk is multiplied in the Y-axis direction, and after transformation such as enlargement or reduction, (xjk,
yjk) position. This translates (X, Y) into (Lxjk *
It is nothing but the coordinate conversion to X + xjk, Lyjk * Y + yjk). Although only the enlargement, reduction, and the like are illustrated as the deformation operation here, the deformation by affine transformation such as rotation may be performed by adding a coefficient for the deformation.

In the character / graphics generation unit 204, all the constituent elements Ejk (k = 1, 2, ...
, Nj) by repeating the generation, deformation, and pasting of the shape of the component, and the input character graphic ID
A shape corresponding to (Gj) can be generated.

In order to operate the embodiment of FIG. 1, first, by inputting a sample character / graphic pattern corresponding to the typeface of the character / graphics to be generated to the component extracting unit 201, the change configuration of the component shape changing unit 203 is changed. The element shape storage table 801 registers the element shape information corresponding to the typeface for each element. FIG. 8 shows a modified component shape storage table 80 for the shape of a square Gothic component stored in advance in the component shape storage unit as described above.
Figure 1 shows an example of changing the shape of a component by registering the shape of a round Gothic component. If there are multiple sample character graphics, repeat for a certain amount.

After that, if the character / graphic code (ID) to be generated is input to the character / graphic generation unit 204, the character / graphic generation unit 204
By the method described in the explanation of the operation of 1., it is possible to generate the character graphic of the typeface in which the constituent elements whose shapes are changed are combined.

In the embodiment of the present invention, as the shape representation of the constituent elements of the character graphic, the representative points of the starting point, the middle point, the middle point, ..., The end point are set on the skeleton line (center line) of the constituent elements, and the coordinates of the representative points are set. A method (skeleton fleshing method) of generating a contour line from position information, width information of a constituent element at the representative point, end point shape information, and the like, and generating the constituent element information at the time of generating the constituent element is shown. I don't care. For example, there is a method (contour filling method) in which the contour shape of a constituent element is represented by a representative point sequence and the constituent element is generated by interpolating this. In this case, the component shape feature information, the data structure of the component shape changing unit and the component shape storage unit, the data structure of the character / graphic configuration information storage unit, the component shape generation method, the component feature extraction method, etc. Although it needs to be changed depending on the shape representation method of the constituent elements of the figure, it is obvious that the same can be realized in the same manner as the embodiments of the present invention described above.

〔The invention's effect〕

As described above, according to the present invention, when a character graphic of a typeface that is not stored in advance is generated and used, a sample character is simply input without newly registering the constituent elements of the typeface. With, it is possible to automatically change the shape of a component registered in advance to a component shape that produces a target typeface. Further, it is possible to easily generate a character graphic of a target typeface without requiring specialized knowledge about character / graphic constituent elements or character / graphic constituent information combining the constituent elements and information registration operation.

[Brief description of drawings]

FIG. 1 is a diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing an example of the data structure of the component shape storage unit 205, and FIG. 3 is an example of the data structure of the character / graphic configuration information storage unit 206. FIG. 4, FIG. 4 is an explanatory diagram regarding a component element pattern, FIG. 5 is a diagram illustrating a sample character / graphic pattern pattern and a component element pattern separated and extracted, and FIG. 6 is a component element extracting unit 20 in the present invention.
1 is a diagram showing an example of a detailed configuration of FIG. 1, FIG. 7 is a diagram showing an internal configuration example of a component shape changing unit 203 in the present invention, and FIG. 8 is an example of a change in component shape according to an embodiment of the present invention. FIG. 9 and FIG. 9 are diagrams showing examples of constituent elements of a character graphic. 201: component extraction unit, 202: component feature extraction unit, 20
3 …… Component shape change unit, 204 …… Character / Figure generation unit, 205
...... Constituent element shape storage unit, 206 …… Character / graphic configuration information storage unit, 301 …… Component element ID, 302 …… Representative point number, 303 …… Representative point coordinate position, 304 …… Representative element at representative point Width of, 3
05,306 …… Point attribute value that represents the shape of the end point of the component, 401…
Character / figure ID, 402 …… Number of components, 403 …… Component ID, 404
…… Deformation coefficient, 405 …… Composite element bonding position (composite position), 501 …… Component pattern, 502 …… Contour line, 503 ……
… Skeletal line, 601 …… Sample character figure pattern, 602,603,604 ……
Separated / extracted component pattern, 701 ... Character / figure recognition means, 702 ... Component shape generation means, 703 ... Pattern matching means, 704 ... Sample character / figure configuration information storage means, 705 ...
... Sample character / figure component element shape storage means, 706 ... Component element extraction control unit.

Claims (1)

[Claims] 1. Preliminarily storing shape information of basic constituent elements forming a character graphic and character graphic configuration information for generating the character graphic by combining the constituent elements for each character graphic, On the basis of the shape information of the constituent element and the character / figure constituent information, the constituent element of the character / graphic object to be generated is read to generate the constituent element, and enlarged.
In the character / graphics generation processing method in which the character / graphics are generated by combining the deformed constituents by performing deformations such as reduction and rotation, the constituents of the character / graphics are extracted from the character / graphics input as the shape sample. Based on first means, second means for extracting the geometrical characteristics of the constituent element extracted by the first means, and based on the geometrical characteristics of the constituent element extracted by the second means. A character and figure generation having a third means for changing the shape information of the constituent elements stored in advance in the character shape and generating a character and figure by using the constituent element whose shape is changed by the third means. Processing method.