JPH0431990A - Recognition device for handwritten character - Google Patents

Abstract

PURPOSE:To reduce the capacity of a storage medium to a minimum and to easily and quickly input a phrase by immediately inputting a desired phrase by inputting one of characters and simple expander information by handwriting even when a desired word is the phrase. CONSTITUTION:This device is provided with a tablet 1 to which character information by handwriting is inputted, a display means 8 on which the character information inputted to the tablet 1 by handwriting is displayed, a first storage means 11 in which a reference character is stored in advance, a second storage means 12 in which an expander is stored in advance, and a third storage means 13 in which the phrase is stored. The character information is qualified based on storage information in the storage means 11, 12, and 13, and when the expander is included in the character information, remaining information except for expander information is qualified. The phrase relating to a qualified result is retrieved based on the expander information, and correspondent one or plural phrases are displayed on the display means 8, and the desired phrase can be selected from one or plural phrases.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a handwritten character recognition device suitable for application to, for example, a word processor device that can input characters by hand.

Summary of the invention: The present invention relates to a handwritten character recognition device suitable for application to, for example, a word processor device, and includes a tablet into which handwritten character information is input, and a tablet into which handwritten character information is inputted into the tablet. a display means for displaying;
a first storage means in which standard characters are stored in advance;
It has a second storage means in which the extension is stored in advance and a third storage means in which the idiom is stored, and the character information is stored in the first, second and third storage means. If the text information contains extension information, the remaining text information is tested excluding this extension information, and the phrases related to the test results obtained are By searching based on the information, displaying one or more corresponding idioms on the display means, and selecting the desired idiom from the one or more idioms, the desired word can be found as an idiom. Even if there is a word, you can input the desired phrase by handwriting at least one character and simple extension information.This minimizes the storage capacity of the storage medium and allows you to input the phrase by hand. can be done simply, reliably, and quickly.

[Conventional technology]

Conventionally, to enter a word (idiom) such as "elementary school" using a word processing device that allows characters to be entered by hand, the user would input the word "elementary school" by hand using a special pen on the tablet. Ta. In addition, the word processor device processes character information stored in an internal storage medium using a method such as a polygonal line approximation method for characters input by handwriting via a tablet. Text information and operator (user)
There is user character information registered from time to time. ] and the input character information, the same or very similar characters or phrases related to these characters are displayed on the monitor screen, and the operator presses a key on the keyboard to display the displayed characters. The operator is allowed to select a desired character or phrase from among several candidate characters or phrases.

[Problem to be solved by the invention]

As is clear from the above, in the past, for example, in order to handwrite a word (idiom) such as "elementary school" into a word processing device, it was necessary to input "elementary school" by hand into a tablet. There was an inconvenience that the speed was slow.

Therefore, for example, if you write "○学○゛" instead of "elementary school", it will automatically be written as "゛'elementary school°',""junior high school school,""
If a word processing device displays candidate phrases such as ``University School'' on its monitor screen, and the operator selects them, the input speed becomes faster.

However, in that case, in addition to the standard character information and the user character information registered by the user from time to time, the standard character information “
It is necessary to store and wipe the information with "○" added to the storage medium and the information with "○" added to the user character information.For example, in addition to the characters such as "Gaku" mentioned above, "○Gaku○" It becomes necessary to separately store information such as this on a storage medium in advance.

Therefore, it places a heavy burden on the operator (user) and increases the storage capacity of the storage medium, which inconveniently increases search time.

The present invention has been made in view of this point, and even if the desired word is an idiom, the desired idiom can be instantly retrieved by simply inputting at least one character and simple extension information by hand. You can input by hand, which minimizes the storage capacity and makes it easier to input idioms.
Furthermore, the present invention attempts to propose a handwritten character recognition device that can be quickly recognized.

[Means to solve the problem]

As shown in FIGS. 1 to 4, the handwritten character recognition device of the present invention includes a tablet (1) into which handwritten character information is input, and a tablet (1) that displays the handwritten character information. a display means (8); a first storage means (11) in which standard characters are stored in advance;
It has a second storage means (12) in which extensions are stored in advance and a third storage means (13) in which idioms are stored, and character information is stored in the first, second and third storage means. Means (11)
, (12) and (13)), and if extension information is included in the character information, the remaining character information excluding this extension information is verified, and this Search for phrases related to the test results obtained by the above based on the extension information, display one or more corresponding phrases on the display means (8), and select the desired phrase from the one or more phrases. This allows you to select idioms.

[Effect]

According to the present invention, character information is stored in the first, second and third
It is verified based on the storage information of storage means (11), (12) and (13), and when extension information is included in the character information, the remaining character information excluding this extension information is verified. The phrases related to the test results obtained through the test are searched based on the extension information, the one or more corresponding phrases are displayed on the display means (8), and the one or more phrases are displayed on the display means (8). Now you can select the desired phrase from the list of phrases, so even if the desired phrase is a phrase, you can instantly input the phrase by handwriting at least one character and simple extension information. This allows the capacity of the storage medium to be minimized, and the handwritten input of idioms can be performed simply, reliably, and quickly.

〔Example〕

Below, an example of the handwritten character recognition device of the present invention will be described in detail with reference to FIG.

In FIG. 1, (1) is a tablet, and desired characters are written on this tablet (1) using a dedicated ben [2]. (3) is a tablet interface for supplying output information from the tablet (1) to the CPU (6). (7) is a display controller that supplies character information from the tablet (1) or CPU (6) to, for example, display B (8) whose display section is composed of a liquid crystal panel, and controls the display section (8). To display characters on a display section.

The CPU (6) has a tablet interface (3
) is subjected to predetermined processing such as polygonal line approximation, and furthermore, based on information such as the order in which it was written, character recognition and other various controls are performed.

(5) is the data, address, and control bus carinal of this CPU (6).

(9) is, for example, a ROM in which the program of this CPU (6) is stored, and (10) is, for example, a ROM that stores the program of this CPU (6).
This is RAM used when U (6) performs processing.

(11) is a character ROM in which general character information such as kanji, katakana, and hiragana is stored. (
12) is an extension ROM in which code shape information such as an asterisk "*" or a round "○" is stored. (
13) is an idiom such as “elementary school” or “drought” (
This is a ROM for phrases in which phrases (or compound words) are stored. (1
4) is, for example, a user ROM (EEFR) in which the user can store arbitrary character information at any time.
OM etc.). Note that this does not have to be a ROM, but may be a RAM with battery backup. In addition, if an external storage device (for example, magneto-optical disk storage) is provided, R
OM(11), (12), (13), (14)
may be RAM as a buffer.

In addition, each of the above-mentioned ROMs (11), (12), (13)
) and (14) need not be separate ROMs, but may be, for example, ROMs with a large capacity. In that case, that one ROM
For example, the areas in which each piece of information is stored may be divided by address, such as character information of symbols or codes from OOH to IFFFH.

(15) is an output circuit (interface circuit), and C
The data supplied from P U (6) via the bus (5) is sent via the terminal (16) to a word processor device or the like which cannot recognize other handwritten characters in a format such as RS-232C, for example. Enable handwritten input using the device.

(17) is an optional keyboard on which, for example, a return key, shift key, cursor key, etc. are arranged. In the above-mentioned handwritten character recognition device, when "■" is handwritten into the tablet (1) as shown in FIG. 2A, "■" is automatically displayed on the display section (8).
Idioms starting with ``金'' such as ``Friday mill'', ``Kinboshi'', and ``Kinmochi'' are displayed, and when the user selects them on the keyboard or points at the idiom using the special pen (2),
The selected idiom can be used as input character information.Alternatively, use the keyboard as an open-yeon keyboard as a means to select an idiom, and move the cursor to the desired idiom using the cursor keys on the keyboard (17). When a return key or the like is pressed, the selected phrase is treated as input character information.

In addition, as shown in Figure 2, when ``゛'○学○'' is handwritten into the tablet (1), the display section (8) will automatically display ``Elementary School'', ``Junior High School'', and ``University School''. A three-letter idiom with the central character “gaku” is displayed, such as “Gaku”.
Here, when the user points at the display area of this phrase on the display section (8) with a dedicated pen (2), the selected phrase is treated as input character information.

In other words, when inputting handwritten data into the tablet (1), by inputting codes and symbols in a predefined format along with the characters on the main body that should be input, for example, if you want to input "elementary school", but you want to input "school" If you can write `` but don't know the other kanji, all related idioms will be displayed on the display, and you can choose from them. Also, even when you don't have time, you can simply and quickly enter the desired idiom because you can just write, for example, "academics" instead of writing "elementary school" in its entirety.

FIG. 3 shows an example of the order in which characters are written in handwritten input of, for example, "■" or "○学○". In this figure, the numbers from 1 to 10 are the number of strokes, and the order in which input characters are written is shown by solid lines according to these numbers.

Here, the operation of the handwritten character recognition device will be described with reference to FIG. 4, taking as an example a case where "[Yes]" is handwritten and input into the tablet (1).

Of the entire handwriting input by hand, the first side number is Kb, the last side number is Ke, and the first side number is Kb.
RAM (10) in the figure includes a character candidate area with an extension (modifying code shape, that is, a modification extension) (this will be referred to as a buffer), a prefix extension area (
This will be referred to as a buffer) and a postfix extension area (this will be referred to as a buffer) are provided.

First, in step (100), the character candidate buffer with extension that stores the recognition results is cleared.

Then, the process moves to step (101).

In step (101), the first side number K is used to specify the range of stroke numbers to be verified as characters within the input handwriting.
b is the first pair, and the prefix extension buffer is ``
"Do not modify" information is stored. Then, the process moves to the next step (102>).

In step (102), the last number Ke is set to the ninth position, and information indicating "no modification" is stored in the postfix extension buffer. and step (103)
to move to.

In step (103), the entire "[exist]" from the 1st to the 9th picture and each ROM (11), (14) (
(See Figure 1). Then, the process moves to step (104).

In step (104), step (103
), it is determined whether there is a candidate.

If “YES”, step (105)
If it is rNQ4, the process moves to step (106).

Here, since "■" is not included in the standard character information, the answer is "NO" and the process moves to step (106).

In step (106), (N-Ke), that is, (
9-9) is the maximum number of strokes in the extension "1"'("○" is the number of strokes "
1”) or more is determined, rYES.

If so, move to step (110) and select “No,
), the process moves to step (107). Since it is small here, it becomes rNQ-, and step (107)
to move to.

In step (107), the last number Ke is set to “1”.
is set to the value obtained by subtracting , that is, "8". And step (
108).

In step (108), (Ke-i-1) is “9-
Since N is also "9", the ninth to ninth strokes, that is, the last stroke written, are compared with the extension information stored in the ROM (12). and step (109)
to move to.

In step (109), it is determined whether a new extension candidate was obtained as a result of the comparison in step (108), and if ryts, step (103) is performed.
If it is 'N mouthwork, step again (106)
to move to. Here, since the ninth image is not an extension, the determination is NO, and the process moves to step (106).

In step (106), (N-Ke), that is, (
In 9-8), it is determined whether the maximum number of strokes in the extension is "1''' or more. If it is ryEs J, the process moves to step (110), and if it is 'NDj, the process moves to step (107).

Here, (9-8) is "1" and they are equal, so it is determined that rYES, and the process moves to step (110).

In step (110), (Kb-1), that is, (
1-1), it is determined whether the maximum stroke number of the extension is "1" or less, and if rYES, the process moves to step (114), and if "NO", the process moves to step (111).

Here, since (1-1) is "0", the determination is "NO", and the process moves to step (111).

In step (111), the leading surface number Kb is set to a value obtained by adding "1", that is, "2". And step (
112).

Step (112) Then, from the first side to the (Kb-
1) The picture (here 1"), i.e. "○" on the first side
" is compared with the information stored in the extension ROM (12). As a result, "O" is obtained as a prefix modification (extension) candidate. Then, the process moves to step (113).

Step (113) Now step (11'2)
At , it is determined whether or not a prefix candidate has been obtained, and if rYES J, the process moves to step (102), and if "NO", the process moves to step (110). Here, the answer is "YES", so step (10)
2) Move to.

In step (102), the next stroke number Ke of the character is
Make it the 9th stroke again. Further, information indicating "not modified" is stored in the post-extension buffer.

Then, the process moves to step (103).

In step (103), Kb is “2” and Ke
is "9", so " from the second side to the ninth side"
The “gold” part is the character ROM (11), (14)
is compared with the stored information, and "gold" is obtained as a character candidate. Then, the process moves to step (104).

In step (104), it is determined whether or not a candidate has been obtained, and if rYES, step (10
5), and if “NO”, step (106)
to move to. In this case, it is determined that ryas, so the process moves to step (105).

In step (105), a new character candidate “kin” is selected.
By combining the prefix modification candidate “○” and the postfix modification candidate “Do not modify!” information, the character with extension, that is, (“○”
, "gold"/"unqualified"), and adds this to the character candidate buffer with extension and stores it.

Then, the process moves to step (106).

In step (106), (N-Ke), that is, (
99), it is determined whether the maximum stroke number of the extension is "1" or more, and if rYES, the process moves to step (110), and if "NO", the process moves to step (107).

Here, the judgment is "NO", so step (107)
to move to.

In step (107), the last stroke number Ke is set to a value obtained by subtracting "1", that is, "8". Then, the process moves to step (108).

In step (108), (Ke↓1) is "9-N is also"
9", the 9th stroke to the 9th stroke, that is, the last stroke written, is compared with the extension information stored in the ROM (12). Then, the process moves to step (109).

In step (109), it is determined whether a new extension candidate was obtained as a result of the comparison in step (108), and if ryEs J, then in step (103)
If “NO”, step (106)
to move to.

Here, since the ninth picture is not an extension, the determination is "NO" and the process moves to step (106).

In step (106), (N-Ke), that is, (9
-8), it is determined whether the maximum stroke number of the extension is "1" or more, and if rYES, the process moves to step (110), and if "NO", the process moves to step (107). Here, (9-8) becomes "1" and is equal, so it is determined that ryEs, and step (11
0).

In step (110), (Kt+-1), that is, (2
1)] Note that Kb was previously set to "2" in step (111)] is determined to be greater than or equal to the maximum number of strokes in the extension "l", and if 'YES', step (
114), and if “NO”, proceed to step (11
Move to 1). Here, (2-1> is “1”, so
ryes -, and the process moves to step (114).

In step (114), it is determined whether the character candidate buffer with extension has been cleared or not. If rYES, the process moves to step (117), and if rNOJ, the process moves to step (115). Here, r
No.

It is determined that this is the case, and the process moves to step (115).

In step (115), step (105
), idiom expansion is performed based on the information stored in the character candidate buffer with extension, and all the results are displayed on the display section (8). And step (1
16) Move to.

In step (116), the user
The user waits for pointing with the dedicated pen (2), and when this is done, the phrase pointed to is recognized and confirmed as the phrase to be input. And it ends.

Note that the explanation with reference to the flowchart in Fig. 4 regarding the case where "Ogaku O" is input by hand is omitted, but "■"
It is easy to understand that this can be recognized in the same way as in the case of .

As is clear from the above, for example, if you want to input "elementary school" and handwrite "○ Gaku O", "elementary school" will be entered immediately.
This allows you to input handwritten phrases easily, reliably, and quickly. In addition to the standard characters such as "Gaku", R also goes out of its way to add the extension "O" to "Gaku", such as "○ Gaku○" and "○○ Gaku".
Since there is no need to store data in OM, the capacity of ROM can be minimized.

In the above embodiment, the extension is set to “○” and “
■'' and ``OgakuO'' are input by hand, but for example, ``Engineering'', ``×Science×'', and ``金'' are input.

It is easy to understand that even if the extension is handwritten as "x" or "1", such as "XX Studies XX", it can be input in the same way. Also ROM (9), (11).

(12) and (13) may be used by dividing one ROM by address.

Furthermore, it goes without saying that the present invention is not limited to the above-described embodiments, and that various other configurations can be taken without departing from the gist of the present invention.

〔Effect of the invention〕

According to the present invention described above, character information is verified based on the storage information of the first, second, and third storage means, and when extension information is included in the character information, this extension information is verified. The remaining character information except for the above is tested, the phrases related to the test results obtained are searched based on the extension information, one or more corresponding phrases are displayed on the display means, and this Now you can select a desired phrase from one or more phrases, so even if the desired word is a phrase, you can instantly select it by handwriting at least one character and simple extension information. This allows the user to input handwritten phrases easily, reliably, and quickly while minimizing the capacity of the storage medium.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of the use of the handwritten character recognition device of the present invention, FIG. 2 is an example of handwritten characters used for explanation, and FIG.
The figure is an explanatory diagram for explaining the process, and FIG. 4 is a flowchart. (1) is the tablet, (3) is the tablet interface, (5) is the bus, (6) is the CPU, (7) is the display controller, (8) is the display unit, (9)
, (11). (12), (13), (14) are ROM, (lO)
is RAM. , the original θ month hand to length palm for Mokiyoshi Dai Mei J's eve.) 1 large - 51 nose ha d O 1st figure + writing method i's 2nd figure 3rd figure

Claims (1)

[Scope of Claims] A tablet into which handwritten character information is input, a display means for displaying the handwritten character information on the tablet, and a first storage means in which standard characters are stored in advance. It has a second storage means in which the extension is stored in advance, and a third storage means in which the idiom is stored, and the character information is stored in the first, second and third storage means. If the above character information contains extension information, the remaining character information excluding the extension information is tested, and the idioms related to the test results obtained are A search is performed based on the extension information, one or more corresponding idioms are displayed on the display means, and a desired idiom can be selected from the one or more idioms. Handwritten character recognition device.