JPH03129459A - Kana/kanji converter - Google Patents

Abstract

PURPOSE:To precisely convert a word including a character other than a HIRAGANA(cursive form of Japanese syllabary) to a KANJI(Chinese character)- mixed word even when part of an inputted character string is comprised of such word by providing a KANJI conversion part which converts inputted character string data to a corresponding KANJI-mixed text by retrieving a dictionary storage part. CONSTITUTION:A KANA/KANJI conversion part 2 performs KANA/KANJI conversion by retrieving information in the dictionary storage part 3 on character string data inputted from an input device 1 and supplied via a control part 5, and outputs a conversion result. The conversion result is supplied to an output device 4 consisting of a display device such as a liquid crystal display device, etc., via the control part 5, and is displayed on the display device. Thereby, since the word including the character other than the HIRAGANA can be inputted as the reading of the KANJI-mixed word, it is possible to precisely convert the character string, part of which being the KANJI, etc., to the KANJI- mixed character string.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention provides a kana-kanji conversion device for converting character string data containing characters other than kana, such as kanji, into accurate kanji-mixed sentences. It is related to.

(Prior art) In recent years, kana-kanji conversion has become indispensable for text creation devices such as Japanese word processors. Then, we apply kana-kanji conversion to convert it into a sentence containing kanji and create a Japanese sentence.

That is, the sentence creation device stores sentences containing kanji in correspondence with hiragana as a dictionary, and other characters are not converted into sentences containing kanji.

On the other hand, input using handwritten character recognition is attracting attention as a text input device, but when the recognition results from this device are sent to the above-mentioned text creation device to create a sentence containing kanji, the characters given to the text input device must be hiragana. There is a problem in that accurate sentences cannot be created unless input is made with this in mind.

(Problems to be Solved by the Invention) As described above, according to the conventional kana-kanji conversion device, the target of kana-kanji conversion is only hiragana character strings, and for example, when inputting handwritten characters, it is originally written in kanji. If part of the power word was written in hiragana, there was a risk that the conversion to kanji would not be performed accurately.

The present invention was made to solve the above-mentioned problems of conventional kana-kanji conversion devices, and its purpose is to convert character strings that contain characters other than hiragana even when characters other than hiragana are input. To provide a kana-kanji conversion device capable of accurately converting even parts into words containing kanji.

[Structure of the Invention] (Means for Solving the Problems) The present invention converts character string data into a character string containing kanji using information about words containing kanji and their corresponding pronunciations stored in a dictionary storage unit. The kana-kanji conversion device is characterized in that it also accepts characters other than hiragana as readings corresponding to words containing kanji stored in the dictionary storage unit.

(Operation) According to the present invention, even if a part of the input character string is composed of words containing characters other than hiragana, it is possible to accurately convert this word into a word containing kanji.

(Example) Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

A kana-kanji conversion device according to an embodiment of the present invention inputs character string data including kanji as part of a word and converts it into a character string containing kanji. configured.

In FIG. 1, reference numeral 1 indicates an input device for inputting character string data consisting of a character code string, and here it is a device equipped with a handwritten character recognition function. The Kana-Kanji converter 2 is
Information in the dictionary storage section 3 is searched for character string data input from the input device 1 and given via the control section 5, kana-kanji conversion is performed, and the conversion result is output. This conversion result is provided via the control section 5 to the output device 4, which is a display device such as a liquid crystal display device, and is displayed.

The control section 5 performs data transmission/reception control between each device/section, start-up recovery control, etc.

FIG. 2 shows a schematic configuration diagram of the input device 1 in the embodiment device. 11 is a coordinate input device composed of a tablet or the like;
Reference numeral 12 indicates a processing control device, and reference numeral 13 indicates an external storage device storing a recognition dictionary and the like. The processing control device 12 described above includes an input section 1
11, recognition processing unit 112, and position data buffer 1
14, a dictionary data buffer 115, and an output data buffer 116.

Here, the input unit 111 stores a coordinate series of a character pattern input by hand from the coordinate input device 11, and uses the coordinate series for input character recognition processing and text editing processing. Further, the recognition processing section 112 detects the writing stroke from the series of coordinate values stored in the input section 111. and,
Stroke feature information of the input character and external storage device 13
The recognition processing unit 112 calculates the degree of similarity between the stroke feature information of the recognition target character obtained from the recognition dictionary stored in the dictionary data buffer 115 and stored in the dictionary data buffer 115, and recognizes the input character. will be held. The recognition result of the input character recognized by the recognition processing unit 112 is stored in the output data buffer 116.

The editing processing unit 113 also detects the input position of the input character recognized by the recognition processing unit 112 on the coordinate input device 11, that is, the input position of the character in the character input possible area (tablet surface) of the coordinate input device 11. The position detection result of this input character is stored in the position data buffer 114 and used for display control in the output device 4.In other words, the coordinate input device 11 and the output device 4 are connected to the position coordinate area (character input area). and character display area) are set to be equal, and the input character pattern or recognized character pattern is displayed corresponding to the character input position.

Also, the coordinate input device 11. For example, it is made of a transparent tablet, and the output device 4 is made of a liquid crystal display device. For example, these are provided one on top of the other as shown in FIG. 3, and the character information displayed on the output device 4 can be visually checked via the coordinate input device 11.

Furthermore, as shown in FIG. 1, the dictionary storage unit 3 has reading data 3a corresponding to each kanji word data 3b.
is stored, but this reading is not only written in hiragana, but also partially or entirely in kanji.

The control unit 5 in the kana-kanji conversion apparatus configured as described above is configured by, for example, a microcomputer, and executes the program of the flowchart shown in FIG. 4, so this will be explained.

First, it starts by turning on the power, etc., and each register,
Buffers and the like are initialized (41). Next, the resulting characters (codes) recognized by the input device 1 as explained using FIG. 2 are taken in from the input device 1 (42),
It is determined whether the code is a command (43). If it is not a command, the character corresponding to the character information (code) in the output data buffer 116 is transferred to the position data buffer 11.
4 is displayed on the output unit 4 in accordance with the position information indicated by 4 (44
). The code of the displayed character is stacked in an input buffer (not shown) in the control section 5 and is subjected to the kana-kanji conversion process by the kana-kanji conversion section 2 (45). On the other hand, in step 43, when the arrival of a command is detected, it is analyzed whether it is a command that instructs the next candidate (46),
If the command is a next candidate instruction, a next candidate request is issued to the processing control device 12 of the input device 1, the character code of the next candidate is received as a recognition result (47), and the process proceeds to display step (44). If the command is not an instruction for the next candidate, it is analyzed to see if it is a command for instructing kana-kanji conversion, and if it is an instruction for kana-kanji conversion, it is caused to perform kana-kanji conversion by the kana-kanji conversion unit 2 (49), and the resulting conversion is The results are sent to the output device 4 for display (50), and further, the control unit 5
The input buffer (not shown) is initialized in preparation for processing the next character (51), and the process returns to step 42.

In response to character input in step 42 during processing by the control section 5, the processing control device 12 of the input device 1 executes a program shown in the flowchart shown in FIG. 5 to perform character recognition and the like.

That is, first initialize each register etc. Stella 1A),
Detects 0N10FF of the pen on the coordinate input device 11 (
Step B>. Here, a series of position coordinate data input when the pen is ON is sequentially stored in the input unit 111 (step C). This input data storage process is repeatedly performed by determining the presence or absence of an instruction input, for example, until completion of data input for one character is instructed by an OFF operation of the pen (step D).

Note that the input of one character's worth of data is completed by turning off the pen. In this case, step E
), it is determined whether the input coordinate data at that time indicates a character input area or a command area (Step F>. Then, if the input coordinate data indicates a character input area, this is data input for one character. is given to the determination processing section shown in step D above as end information.

Each time data for one character is input in this way, the recognition processing unit 112 performs analysis such as determining the characteristics of the writing stroke indicated by the series of position coordinates, and stores the recognition results in the external storage device 13. The input character is recognized by comparing it with a recognition dictionary, etc. (Step G).

Thereafter, the editing processing unit 113 obtains the input coordinate position (x, y) of the character that has been recognized (step H). The input position of this character is obtained, for example, as position information of the square in which the input character is written, among the squares set in a matrix on the character input surface of the coordinate input device 11.

Note that if the input position data when the pen is OFF indicates a command area for selecting input character candidates or instructing execution of kanji conversion, the input position data is processed as input for the corresponding command (step ■).

Through such a series of character input procedures, the input character or command is recognized from the series of coordinate values input from the coordinate input device 1, and the character input position (x, y
) is required.

Next, with reference to FIG. 6, a description will be given of the process when a handwritten character string "Jiyatsuboshi wo jijisuru" is inputted from the input device 1. Here, among the above character strings, "dried"
The processing including kana-kanji conversion will be explained using the character as an example.

The user inputs "jiyatsuboshi" (5a) because the character for "dried" is easier to write than "kan" in hiragana.

At this time, in order to indicate the character related to the current input, the control unit 5
controls so that the cursor 51 is displayed.

When "Kan" is input, the recognition processing unit 112 performs character recognition as explained in step G of FIG.
The following four characters are obtained as recognition results: ``Hoshi'', ``Te'', and ``〒''. Then, the code of the first candidate "1000" is output buffer 1
16, and display based on this is performed as described in step 44 of FIG. 4 (5b). At this time, the control unit 5 displays the next candidate icon (52>) in the command area in order to allow the user to select the next candidate.Here, the user selects another candidate because the character he/she input is not "1000". Request a candidate and operate the next candidate icon (52).

In response to this, as explained in steps 46 and 47 in FIG. 4, the code for the next candidate "Hoshi" is stored in the output data buffer 116, and "Hoshi" is displayed (5c>
. This is the character entered by the user, so next
Proceed to input. As soon as the next character is input, the character for ``dried'' is confirmed, and the next candidate icon (52) for selecting another candidate is deleted (5d). In this way, the character recognition process proceeds. However, when kana-kanji conversion is instructed as a command (by operating a predetermined area on the input surface of input device 1, etc.), the process proceeds to steps 48 and 49 in FIG. The processing operation by unit 2 will be explained with reference to the flowchart in FIG.

After starting the Kana-Kanji conversion, the output data buffer 11
Conversion is performed until there are no more character strings (codes) in 6 (
Step a). First, the pointer is advanced from the first code (step eL) and the dictionary is searched for each character of the code (step b), and this is repeated until an independent word is detected (step C).

In this example, since "jiyatsuboshi" is an independent word, it is necessary to search for "slightly" in the dictionary storage unit 3 in FIG. and displayed on the output device 4. Thereafter, the pointer is advanced (step e), and processing proceeds to "no", but since "no" is an attached word, attached word analysis is performed (step d), and it is processed as an attached word.

Below, the pointer is advanced (step e) and the "Modify..."
・'' is processed, ``revise'' is given to ``revise'', ``instruction'' is given to Shiji J, and finally a sentence containing kanji such as ``instructs to make some corrections'' can be obtained.

In this way, the hon-kana-kanji conversion device allows the inclusion of kanji in the reading, so it is possible to accurately convert a character string containing kanji into a complete character string containing kanji.

Furthermore, since the kanji character string is searched by dictionary search as described above, there is an advantage that the kana-kanji conversion process can be performed easily and the processing speed is not impaired.

Note that the present invention is not limited to the above embodiments,
If characters such as "su" and "he" that are difficult to distinguish between "hiragana" and "katakana" are stored in a dictionary with their readings associated with words, the correct hiragana or katakana characters can be obtained through kana-kanji conversion. , extremely convenient. Further, the input device 1 may be an OCR device or a storage device in which character strings containing kanji are stored in advance.

In such cases, the next candidate command for the input character can be input by providing a separate keyboard input device, but the candidate should be selected before inputting the character string to the relevant kana-kanji conversion device, and the next candidate command will not be used. It depends on what you do.

[Effects of the Invention] As explained above, according to the present invention, characters other than hiragana are allowed to be included in the pronunciation of words containing kanji. Even if you select various input devices, you do not need to be conscious of inputting in hiragana.

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram of a kana-kanji conversion device according to an embodiment of the present invention, FIG. 2 is a block diagram of an input device, FIG. 3 is a configuration diagram of an input device, and FIGS. 4, 5, and 7 Figure 6 is a flowchart for explaining the operation of an embodiment of the present invention.
The figure is a diagram showing an example of a display when characters input from an input device are recognized.

Claims (1)

[Scope of Claims] An input device for inputting character string data consisting of a character code string; a dictionary storage unit in which words containing kanji and pronunciations containing characters other than hiragana are stored in correspondence; and the input device. 1. A kana-kanji conversion device, comprising: a kanji conversion section that searches the dictionary storage section for character string data input from the kanji storage section and converts it into a corresponding kanji-mixed sentence.