JP2901525B2 - Character creation method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character creation system, and more particularly to a character creation system for creating characters or symbols other than those provided in advance in a personal computer or a word processor.

[0002]

2. Description of the Related Art Generally, characters and symbols (hereinafter referred to as characters, etc.) displayed on a display screen or printed on paper as an output of an information processing device using a computer.
Is represented by a so-called dot pattern composed of a two-dimensional array of black and white dots. Such an information processing apparatus generally holds character pattern data for generating a dot pattern of a required character or the like in the form of a dot font or an outline font (hereinafter, referred to as OLF). . However, the number (character type) and type (typeface) of characters used in business computers that process Japanese language, word processors that handle Japanese and Chinese, etc. increase, and similar characters are used in personal name kanji. However, since different details may be used, it is uneconomical and impractical to prepare all of them in advance in terms of storage capacity. So, for example, for kanji in the case of a Japanese word processor,
In many cases, about the common-use kanji (JIS C6226-1978) is prepared, and others are prepared and used by the user when necessary.

In order to generate character pattern data such as the above-mentioned dot font and OLF, a character pattern designed on paper or the like is read by an image scanner such as a television camera, and the read image signal is binarized. The method is to display the data on the display screen, correct the missing points and extra points manually, and then save the determined pattern data as a dot font, or extract the outline data and save it as OLF. Normal. In addition, there is a method of drawing a character pattern by manually inputting dots, line segments, curves, and the like directly on a display screen. However, creating character pattern data by such a basic method requires extremely specialized skills and is troublesome. Therefore, in a personal computer, a word processor, or the like, a user can relatively easily create characters and symbols (hereinafter, referred to as external characters) other than registered characters (including symbols) prepared in advance in the system. There is provided a character creation method for creating an external character by using a registered character.

As shown in FIG. 3, the basic form of this character creation system is a character reference means 31 for reading one of the registered characters and displaying it in a character creation area on the screen, and a registered character displayed in the character creation area. The user selects and specifies a registered character that is most similar to the external character to be created, and creates the desired external character by using the registered character as much as possible.

A procedure for creating an external character using this method will be described with reference to a specific example with reference to FIG. Reference number 4
The user who intends to create the external character shown in FIG. 5 searches for a registered character that is the closest and has many divertable portions by referring to a character code table or the like. At this time, beside the external character
In addition to the registered character indicated by reference numeral 41, which is diverted as a reference character, in order to refer to the character design when creating an external character deviation (hen), the registered character using the same deviation indicated by reference numeral 42 is examined. Record as After making the above preparations, the type of the font and the character code or the reading of the registered character to be diverted are input to the character reference means 31, and reference characters 4
As shown in FIG. 3, it is displayed in the character creation area (solid line frame). If the read character image cannot be used as it is, the size and position are corrected as indicated by reference numeral 44 by the reduction and movement function of the character correction means 32. Finally, when the bias, which is the remaining required character segment, is created by the new creation function of the character correcting means 32 by imitating the character design, size, and smoothness of the line of the reference character that has been checked, reference numeral 45 Are created. If the character image read by the character reference means 31 cannot be used only by reduction and movement, necessary portions are further added to the character correction means 32.
Modify by deleting and creating new functions.

A further improved system is disclosed in Japanese Patent Application Laid-Open No. 64-2.
090 publication. In contrast to the above-described method, which can specify only one approximate character that can be diverted most, the method described in this publication specifies a plurality of registered characters including character units that can be diverted from registered characters. By designating the coordinates of the character segments to be diverted among these registered characters, extracting and synthesizing the necessary character segments, it is possible to efficiently create a character pattern.

[0007]

Among the above-mentioned conventional character creation methods, the former method shown in FIG. 3 is capable of inputting by designating and inputting one registered character which is the closest to the character or the like to be created. Character units are used as is, but if there is a missing character unit, it is a method of creating it yourself by referring to the approximate character unit, so if you compare it with designing a completely new character Although the number of steps can be reduced and the efficiency has been improved, creating and modifying character segments requires specialized skills, and it is difficult for general users and other people who are not good at character design to make satisfactory characters, There was the disadvantage that it took a long time to create the characters, and in the end it was necessary to endure the use of character designs that were unsatisfactory.

In the latter method described in Japanese Patent Application Laid-Open No. 64-2090, a plurality of registered characters including a character segment that can be diverted are designated, and a character segment desired to be diverted is taken out of these and synthesized. It is no longer necessary for a general user to create a character segment by himself. However, it is necessary to select a plurality of registered characters including character units that can be diverted and to input them by character codes or readings, and it takes time to select and diverge characters when input is not known. Remains.

An object of the present invention is to solve the above disadvantages,
Anyone can create the skeleton of the character they need and break it down into appropriate character segments, and then combine them using the prepared character units for each typeface to create a beautiful and unified design in a short time. An object of the present invention is to provide a character creation method that can easily create characters.

[0010]

According to a first aspect of the present invention, there is provided a character creating method for inputting a skeleton of a character created by handwriting and displaying the skeleton in a character creating area, and a typeface input for designating a typeface of the character to be created. Means, a skeleton character unit dictionary for search storing pattern data of a non-fleshed skeleton character unit for search in association with identification information of character units constituting the character, and identification information of the character unit. A character segment dictionary unit that stores pattern data of a fleshed character segment of each typeface in correspondence with the character skeleton displayed in the character creation area according to a user instruction. Extracting means for extracting the size, position, and pattern data of each character segment as skeleton unit data, and extracting the skeleton character unit dictionary for search based on each skeleton unit data extracted by the extracting unit Inspect the department A pattern recognition unit for acquiring identification information of the corresponding character unit; and a character unit dictionary having a fleshed-out character type designated by the typeface input unit based on the identification information acquired by the pattern recognition unit. And a unit for synthesizing the character pattern data of the typeface by replacing the skeleton unit data extracted from the unit with the size and position of the skeleton unit data created by the extracting unit.

According to a second aspect of the present invention, in the character creation system according to the first aspect, the handwriting input means reads a handwritten skeleton character written in a designated frame on a sheet of paper with an image scanner, and reads the external character. It is displayed in a character creation area on the creation screen.

According to a third aspect of the present invention, in the character creation system according to the first aspect, the handwriting input means converts a skeleton character entered in a frame on the input screen by a pen input device on the external character creation screen. Characters are displayed in the character creation area.

According to a fourth aspect of the present invention, there is provided the character creating method according to the first aspect, further comprising a character correcting means for correcting the character pattern data synthesized by the unit combining means on an external character creating screen. Features.

[0014]

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of one embodiment of the present invention.

As shown in FIG. 1, the character creation method according to the present embodiment includes a handwriting input unit 1 for inputting a skeleton of a character created by handwriting and displaying the skeleton of the character in a character creation area. The extracting means 2 for extracting the size, position, and pattern data by decomposing into character units such as biased and adjacent sides in accordance with the instructions of, and the character unit extracted by the extracting means 2 is used in any of the prepared character units. A pattern recognizing means 3 for checking for a match; and a skeleton character segment dictionary unit 4 for search storing pattern data of fleshless skeleton character segments for search which are easy to compare and match in the pattern recognizing unit 3.
Typeface input means 5 for inputting the typeface of the character to be created,
The character unit dictionary unit 6 storing the pattern data of the fleshed-out character units of each typeface and the pattern data of the specified typeface of the character unit recognized by the pattern recognition unit 3 are read out from the character unit dictionary unit 6. And a unit for synthesizing the external character of the designated typeface by adjusting and replacing the size and position of the character unit of the handwritten input character.

The handwriting input means 1 reads a non-fleshed skeleton character written in a frame of a predetermined size on the paper by an image scanner, or directly reads a skeleton character image in a designated frame of an input screen by a pen input device. To be input to a personal computer or word processor, and displayed in a character creation area on the external character creation screen.

The cut-out means 2 decomposes the skeleton character displayed in the character creation area into character segments in accordance with a user's instruction, and converts the information of each decomposed character unit into data. For example, if a skeletal character as indicated by reference numeral 20 in FIG. 2 is displayed and the user instructs to decompose the character into two adjacent character segments as indicated by dotted lines, the reference numerals 21 and 22 indicate The size, shape, position, and pattern data of each decomposed character segment as shown are extracted and converted into data. The instruction for disassembly into character segments can be given not only by the vertical division as illustrated, but also by horizontal division or a combination of vertical and horizontal straight lines such as L-shape, inverted L-shape, and rectangle.

The pattern recognizing means 3 receives the data of each of the decomposed character segments from the cutout means 2 and checks which of the previously prepared character units matches the data.
Therefore, the size of each received character segment is standardized to the size of one character, and the standardized pattern data is compared with the reference pattern data of the character units stored in the skeleton character unit dictionary for search 4. The comparison is performed sequentially, and the identification information of the corresponding character segment is obtained and passed to the unit combining means 7. That is, FIG.
Is standardized to the size of one character as shown by reference numerals 24 and 25, and the pattern data and the skeleton character unit dictionary for search 4 Are sequentially compared with the reference pattern data stored in the character pattern by a well-known pattern matching method, a character segment whose matching degree exceeds a certain reference and is the highest is extracted, and its identification information is obtained.

The search skeleton character segment dictionary unit 4 is a search dictionary which stores reference pattern data of only a non-fleshed skeleton character segment for search, which facilitates pattern matching in the pattern recognition means 3. 2, reference numerals 24, 25
The reference pattern data of a character unit having only a skeleton as shown in (1) is stored in association with the identification information (character unit identification code) of the character unit.

[0021] typeface input means 5 is a means to specify the typeface of the external character you try to create, Mincho, Gothic, specified by selecting one of a typeface that has been prepared in advance of the textbooks and the like I do.

The character segment dictionary unit 6 is a reference dictionary storing pattern data of fleshed character segments of each typeface, and is a fleshed-out character segment as indicated by reference numerals 23 and 26 in FIG. Are stored with the identification information of a character segment as a primary key and a typeface such as Mincho, Gothic, textbook, etc. as a secondary key.

Based on the identification information of the character segment passed from the pattern recognizing unit 3 and the font type specified by the font input unit 5, the unit synthesizing unit 7 fills the specified font from the character unit dictionary unit 6. Is read out, converted to match the size and position data of the skeleton character unit from the cut-out means 2 and replaced with the character unit of the skeleton character displayed in the character creation area. To compose the external character of the designated font. The character segments indicated by reference numerals 23 and 26 in FIG.
The size and the position are adjusted as shown in FIG. 8, and these are combined to form the external character of the designated typeface having a flesh as indicated by reference numeral 29.

Next, referring to FIGS. 1 and 2, a series of operations of the external character creation processing according to the present embodiment will be described. To create an original character using this character creation method, first, a fleshless handwritten skeleton character is read by an image scanner, or a skeleton character image is written by a pen input device and input by the handwriting input means 1. At the same time, the font is designated by the font input means 5. The input skeleton character image is, as indicated by reference numeral 20 in FIG.
Since the size is adjusted and displayed in the character creation area, as shown by the dotted line, an instruction is given to decompose the partial and partial character segments.

The extracting means 2 decomposes into character segments as indicated by reference numerals 21 and 22 according to the instruction, converts the size, shape, position and pattern data of each character unit as skeleton unit data, and Hand over to recognition means 3. The pattern recognizing means 3 receiving this standardizes each skeleton segment data to the size of one character as shown by reference numerals 24 and 25, and then uses the conventionally recognized pattern matching method to search. The reference information is compared with the reference pattern of the skeleton character unit dictionary unit 4 to determine the identification information of the character unit.

Based on the identification information of each character segment determined by the pattern recognizing means 3, the unit synthesizing unit 7 converts the character unit dictionary unit 6 to fill in the designated typeface as indicated by reference numerals 23 and 26 in FIG. The pattern data of a certain character segment is read out and adjusted according to the size and position of the character segment of the original handwritten skeleton character as indicated by reference numerals 27 and 28, and these are synthesized and indicated by reference numeral 29. The external character of the design of the designated typeface with flesh is synthesized.

In this way, by replacing the character units of the skeleton character input by handwriting with the character units of the designated typeface prepared in advance to create a composite character, even a person who is not good at character design can use the character required. If you make a skeleton and decompose it into character segments, you can use the prepared character units of the typeface to combine them, and easily create beautiful external characters with a unified design in a short time .

In the above description, as exemplified in FIG. 2, the external character to be created can be decomposed into two parts, and the adjacent "between" is a character unit, and the skeleton character unit dictionary unit 4 for retrieval is used.
And stored in the character segment dictionary unit 6.
Although "ma" is a widely used kanji included in common kanji, it is not a character frequently used as a character segment and registered as a radical in a general dictionary. When a character segment that is not registered in the skeleton character segment dictionary unit 4 for search is specified, the pattern recognition unit 3 reports that recognition is not possible because a matching degree higher than the reference is not obtained, and requests the user to reconsider. Come. Therefore, the user decomposes “between” into two character segments “gate” and “day” and instructs re-execution. In addition,
When multiple character segments with the same or more than the same degree of match are found as a result of pattern recognition, the identification information of multiple character units is stored and the character unit with the highest degree of match is displayed preferentially. However, as in the case of character recognition by normal pattern recognition, confirmation or selection of the user may be requested.
However, the target of pattern recognition is a character segment, the number of target characters is limited as compared with normal character recognition, and the number of complicated characters is small, so the recognition rate is extremely high.

In the above description, as the handwriting input means 1, a case where a handwritten skeleton character is input by an image scanner for inputting a still image and a case where a handwritten input is performed using a pen input device have been described. It is also possible to use a video camera capable of inputting moving images to input handwritten characters.

Further, the pattern combining unit 7 combines the pattern data of the fleshed-out character units obtained from the character unit dictionary unit 6 by adjusting the size and the position, and synthesizes the external characters of the design of the designated font. Although it has been completed, if a correction process by a conventional method can be added to the character pattern data after synthesis, a more beautiful character can be created.

[0031]

As is clear from the above description, according to the character creation method of the present invention, if the required character skeleton is formed and decomposed into character segments, the character elements of the prepared typeface will be By combining pieces, it is possible to easily create beautiful external characters that are unified in design in a short time. Therefore, even if a person who is not good at character design does not know the reading and names of similar characters, reference characters, and character fragments, anyone can easily create beautiful characters with a uniform design of the desired typeface. is there.

[Brief description of the drawings]

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

FIG. 2 is an explanatory diagram illustrating an example of a transition of a character creation process according to the present embodiment.

FIG. 3 is a block diagram showing a configuration of an example of a conventional character creation method.

FIG. 4 is an explanatory diagram showing an example of transition of a character creation process according to a conventional character creation method.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 handwriting input means 2 cutout means 3 pattern recognition means 4 skeleton character unit dictionary unit for search 5 typeface input unit 6 character unit dictionary unit 7 unit synthesis unit 31 character reference unit 32 character correction unit

Claims (4)

(57) [Claims] 1. A handwriting input unit for inputting a skeleton of a character created by handwriting and displaying the skeleton in a character creation area, a font inputting unit for designating a font of a character to be created, and identification information of a character segment constituting the character A searchable skeletal character unit dictionary storing pattern data of a non-fleshed skeleton character unit for search in association with a character unit having a flesh of each typeface in association with identification information of the character unit. A character segment dictionary storing pattern data, and decomposing the skeleton of the character displayed in the character creation area into character units such as unbalanced and adjacent according to a user's instruction, and the size and position of each character unit Extracting means for extracting pattern data as skeleton unit data, searching the skeleton character unit dictionary for search based on each skeleton unit data extracted by the extracting unit, and identifying identification information of the corresponding character unit. Dad to get The character unit pattern data having the flesh of the typeface designated by the typeface input unit is extracted from the character unit dictionary unit by the turn recognition unit and the identification information acquired by the pattern recognition unit, and is created by the cutout unit. A character synthesizing means, comprising: a unit synthesizing means for performing substitution according to the size and position of skeleton unit data to synthesize character pattern data of the typeface. 2. The method according to claim 1, wherein said handwriting input means reads a handwritten skeleton character entered in a designated frame on a paper surface by an image scanner and displays the skeleton character in a character creation area on an external character creation screen. 1. The character creation method described in 1. 3. The character according to claim 1, wherein said handwriting input means displays a skeleton character entered in a frame on an input screen by a pen input device in a character creation area on an external character creation screen. Creation method. 4. A character creating method according to claim 1, further comprising a character correcting means for correcting the character pattern data synthesized by said segment synthesizing means on an external character creating screen.