JPH02222085A - Recognized character type designating system - Google Patents

Abstract

PURPOSE:To prevent the erroneous designation of a recognized character type and designation omission, to improve a recognition rate, and to reduce the burden of the character type designation by completing the description of one sentence, and then instructing the recognized character type and a character position. CONSTITUTION:The sentence is described in the character frame part of a tablet 11 with the use of a stylus pen, and the respective characters of the sentence are inputted as character data (a) in an input switching part 1. The switching part 1outputs the character data (a) to a character position detecting part 2 and a recognizing part 4, and outputs a switching signal C1 to a recognition result switching device 3 and a recognition result storage part 9. Thus terminals 3a1 and 3a2 of the switching device 3 are connected to the storage part 9, and thus a data write state is prepared. Further the detecting part 2 outputs a character position signal a1 indicating the position of each character through the switching device 3 to the storage part 9, the recognizing part 4 outputs a character type signal a2 to a standard pattern switching device 8, respective standard patterns 5 to 7 are matched with respect to the respective data (a), and recognition data r012 for respective character types are outputted to the storage part 9.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a recognition character type designation system in a character recognition system that recognizes input characters from their handwriting.

[Conventional technology]

Conventionally, there have been the following recognition character type designation systems of this type.

As shown in FIG. 3, the writing surface of the character recognition device is divided into a character frame section 20 for writing characters and a function section 21 for specifying the type of character. A stylus pen 22 is used to write characters or touch the function section 21.

For example, when using this to write down the sentence ``We Japanese people are faced with the need to input Japanese into computers.'', the following operations are performed.

(a) Bent touch [Kanji (Kana)] in the function section 21.

(b) Write “We Japanese,” in the character frame 20.

(c) Bent touch [Katakana] in the function section 21.

(d) Write “Computer” in the character frame portion 20.

(e) Bentouch [Kanji (Kana)] in the function section 21.

(v) Write “I am faced with the need to input Japanese into the text box 20.”

Simultaneously with this operation, the following internal processing is performed, and a recognition result corresponding to the specified character type is output.

The recognized character type is represented by the value of m, where m=o when the recognized character type is kanji/hiragana, m=l when it is katakana, and m when it is alphanumeric.
=2. Furthermore, the standard patterns previously held within the device are divided into a plurality of groups corresponding to each character type.

The standard pattern for kanji and hiragana is B (0) = BO1, BO2, ..., and the standard pattern for BON katakana is B (1) = B11. B12. ..., the standard BIN alphanumeric pattern is B (2) = B21. B22. ..., B2N.

When recognition character type m is specified, standard pattern B (
m) is selected and this standard pattern is used for recognition. For example, when katakana is specified, m=1, and the standard pattern B (m) = B (1) = B11. B12
．． -, BIN is used for recognition, and the recognition result is Katakana.

As mentioned above, specify the character type before writing the character,
After that, by writing down the characters, the specified character type can be recognized as a result.

[Problem to be solved by the invention]

As explained above, in the conventional recognition character type designation method, when writing a character in the character frame section 20 with the stylus pen 22, each time the character type changes, the function section 21 specifies the character type corresponding to that character. This has the following drawbacks:

(b) Specifying the character type during writing becomes a burden on the scribe.

(b) Since the action of specifying the character type is inserted in the middle of character writing,
I often forgot to change the character type and kept writing, only to realize it after I had finished writing the sentence and write the same character again. In this way, errors in specifying character types or forgetting to specify them are likely to occur, resulting in a reduction in the recognition rate.

SUMMARY OF THE INVENTION An object of the present invention is to provide a recognized character type designation method that solves the above problems and allows character type designation to be performed easily and accurately.

[Means to solve the problem]

The recognition character type designation method of the present invention includes a recognition means that includes a standard pattern corresponding to a plurality of character types and outputs recognition results for all character types of input character data based on the standard pattern, and a character type designation means that specifies the character type. , writing the character data recognition result from the recognition means;
The present invention is characterized by comprising recognition result storage means for reading out the character data recognition result for one character type designated by the character type designation means.

〔Example〕

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

FIG. 1 is a block diagram showing a recognition character type designation system according to an embodiment of the present invention.

The recognition character type designation system of this embodiment includes a manual switching section 1, a character position detection section 2, a recognition result switching section 3, a recognition section (recognition means) 4, a kanji/hiragana standard pattern 5, and a katakana standard pattern 6. and alphanumeric standard pattern 7, standard pattern switch 8, and recognition result storage unit (recognition result storage means)
9, and a character type designation section (character type designation means) 10.

The input switching section 1 has a function of determining whether data from the tablet 11 is character data or data obtained by Bentouch of a function section (not shown). When this input switching unit 1 determines that it is character data, it outputs the character data a to the character position detection unit 2 and the recognition unit 4, and also outputs a switching signal c to the recognition result switching unit 3 and the recognition result storage unit 9. death,
If it is determined to be function data, it has a function of outputting function data b to the character type specifying section 10 and outputting a switching signal Ct to the recognition result switching device 3 and the recognition result storage section 9. The input switching unit 1 also has a function of outputting character position data P+, Pg input after inputting the function data to the character type designation unit 1o.

The character position detection section 2 has a function of detecting the position of the character data a input from the input switching section 1 within the character frame and outputting a character position signal a1 indicating the position to the recognition result switching device 3.

The recognition result switch 3 connects the terminal 3a+ when the switching signal is input.
and 3a2 are connected to the recognition result storage section 9, and the terminals 3b+ and 3bz are connected to the recognition result storage section 9 when Ct is input.

The recognition unit 4 outputs the character type signal a2 to the standard pattern switch 8, matches the input character data a with the standard patterns 5 to 7 of all character types, calculates the degree of similarity, and converts the recognition data r0, 2 It has a function of outputting a character type signal a2 to the recognition result switching device 3 as well as outputting the character type signal a2 to the recognition result storage section 9. Therefore, the recognition data rots is for each character data a
Recognition results using Kanji/Hiragana Standard Pattern 5.

The content includes three types of recognition results: recognition results based on the katakana standard pattern 6 and recognition results based on the alphanumeric standard pattern 7.

The character type specifying unit 10 outputs a character type signal bt to the recognition result switching unit 3 based on the function data b from the input switching unit 1, and also outputs character position data P,.

It has a function of outputting a character position signal S indicating the character position from Pt to Pz based on Pt to the recognition result switch 3.

When the recognition result storage unit 9 receives the switching signal C5,
Recognition data r from the recognition unit 4. I! and recognition result switch 3
Input character position signal a1 and character type signal at from
Recognition data r61□ is converted into character position signal a1 and character type signal a
It has a writing function corresponding to 2. Further, when the recognition result storage section 9 receives the switching signal c2, the recognition result storage section 9 receives the character position signal S and the character type signal b2 from the recognition result switching device 3, and based on these signals, recognizes the recognition data r01. It has a function of reading out necessary recognition data R from among the following. That is,
Extract the characters between 21 and 22 indicated by the character position signal b1,
It has a function of specifying and extracting one of the three types of character types included in the recognition data of these characters using the character type signal bt.

Next, the operation of this embodiment will be explained. To make it easier to understand, I will explain the previous example [We Japanese people are faced with the need to input Japanese into computers. ” will be explained below.

First, the above sentence is written in the character frame of the tablet 11 using a stylus pen. The written text is input into the manual switching unit 1 as character data a for each character.

The input switching section 1 outputs character data a to the character position detection section 2 and recognition section 4, and outputs a switching signal C to the recognition result switching device 3 and recognition result storage section 9. As a result, the recognition result switch 3 switches to the terminal 3a.

3az to the recognition result storage unit 9, and the recognition result storage unit 9
prepares to write data.

The character position detection section 2 outputs a character position signal a1 indicating the position of each character from the character data a to the recognition result storage section 9 via the recognition result switch 3. For example, when the above sentence is written as shown in FIG. 2, a signal a1 indicating the character position (1, 1) is output for the character data a of "I".

The recognition unit 4 outputs the character type signal a2 to the standard pattern switch 8, and selects each standard pattern 5 to 7 for each character data a.
The recognition data ".1□" for each character type is output to the recognition result storage unit 9. For example, when character data a is "wa", the recognition results of kanji/hiragana and katakana for "wa" are matched. The recognition result and the alphanumeric recognition result are output as recognition data r (11□.The recognition unit 4 outputs the recognition result storage unit 9 via the recognition result switch 3
A character type signal a2 indicating three types of characters is outputted.

The recognition result storage unit 9 stores the input recognition data rGl□ at an address indicating the character position and character type based on the character position signal a and the character type signal a2.

This storing operation is performed for all characters of the above sentence.

Next, as shown in FIG. 2, the function section is pen-touched to designate the character type and character range.

This instruction specifies the character range P1 (“wa”) to P2 (“,”) with the function section set to “Kanji (Kana)”.
Change the function part to "Katakana" and P2 ("Ko")
) ~ P3 ("Yu"), and finally set the function section to "Kanji (Kana)" and press P.

(“ni”) to P4 (“.”).

When the function section is instructed to "Kanji (Kana)", the input switching section 1 outputs a switching signal C2 to the recognition result switching device 3 and the recognition result storage section 9. As a result, the recognition result switching device 3 switches the connection of the recognition result storage section 9 to the terminals 3b, 3bt, and the recognition result storage section 9 prepares to read data. Further, the input switching section 1 outputs the function data b to the character type specifying section IO.

The character type designation unit 10 receives a character type signal b indicating "Kanji (Kana)".
! is output to the recognition result storage section 9 via the recognition result switch 3.

When the character range P I""' Pz is designated, character position data P+, Pg are output to the character type designation section 10.

The character type specification section 10 specifies character positions (1, l) to (1°7).
A character position signal S indicating ``S'' is output to the recognition result storage section 9 via the recognition result switch 3.

The recognition result storage unit 9 specifies "I" to "," from among the stored recognition data based on the input character position signal. Then, based on the character type signal bt, "wa" ~ "
, '', only the recognition data of ``Kanji (Kana)'' is selected and read out.

As a result, from the recognition result storage unit 9, "I" ~ ","
Kanji and Kana characters are output in chronological order.

The same operation as above is performed for the instructions P2 to P, and P, to P4.

〔Effect of the invention〕

As explained above, the recognition character type designation method of the present invention is
- Since the system specifies the type and position of characters to be recognized after writing down the text, it is possible to prevent mistakes or forgetting to specify the type of characters to be recognized. As a result, the recognition rate is improved and the burden of specifying character types is reduced. There is an effect that can be done.

[Brief explanation of the drawing]

Figure 1 is a block diagram showing a recognition character type designation system according to an embodiment of the present invention, Figure 2 is a diagram showing where character types are collectively designated in the recognition character type designation system of Figure 1, and Figure 3 is a conventional recognition method. FIG. 2 is an explanatory diagram of a character type designation method. 4...Recognition section 5...Kanji/Hiragana standard pattern 6...Katakana standard pattern 7...Alphanumeric standard pattern 9...Recognition result storage section 10...Character type designation section Hi4 Figure 2

Claims (1)

[Claims] (1) A recognition means that is provided with a standard pattern corresponding to a plurality of character types and outputs recognition results for all character types of input character data based on the standard pattern; A character type designation means that specifies a character type; Write the character data recognition result,
A recognized character type designation system comprising: recognition result storage means for reading out the character data recognition result for one character type designated by the character type designation means.