JP2001243424A - Character input and recognition system and method for small-sized electronic equipment device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To intuitively and easily input many characters at low cost by using only a ten-key as an input device. SOLUTION: The character input and recognition system for the small-sized electronic equipment device 11 having an input device 12 with plural keys is equipped with an input data conversion part 102 which converts input data of keys pressed in order when a character is handwritten on the keys into vectors, a reference data storage part 103 which stores reference data on characters grouped by features of the lines constituting the character in the form of plural tables, and a character estimation part 104 which estimates and recognizes the character from the lines corresponding to the vectors converted by the input data conversion part by referring to the tables in the reference data storage part.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a small electronic device such as a portable telephone. In particular, the present invention relates to a character / recognition system and method for a small electronic device that enables recognition of handwritten input characters drawn on a numeric keypad.

[0002]

2. Description of the Related Art Conventionally, as described in Japanese Patent Application Laid-Open No. 6-28085, in an electronic apparatus which has been reduced in size and cost, only a ten-key input is mainly used as an input device. At the time of character input, for example, (A, B, C) is set to “2” of the numeric keypad, (D, E, F) is set to “3”,
(G, H, I)... Is assigned to “4”, and the character to be input and the number corresponding to the character are searched for. Then, the number is pressed several times until the desired character is displayed on the display device.

[0003]

However, in the above-described character input, the number must be pressed many times until the desired character is displayed, and the character cannot be input intuitively. There is. Accordingly, the present invention has been made in view of the above problems, and provides a character input / recognition system and method for a small electronic device capable of intuitively and easily inputting a large number of characters using only a numeric keypad as an input device. The purpose is to:

[0004]

In order to solve the above-mentioned problems, the present invention provides a character input / recognition system for a small electronic device having an input device comprising a plurality of keys. An input data conversion unit that converts input data of a key that is sequentially pressed when a character is drawn into a vector, and reference data that stores, as a plurality of tables, reference data in which characters are grouped by characteristics of lines constituting the characters. A storage unit; and a character estimating unit that estimates and recognizes a character from a line corresponding to the vector converted by the input data conversion unit with reference to the table of the reference data storage unit. Provided is a character input / recognition system for a small electronic device.

[0005] By this means, it is possible to input a large number of characters at low cost, intuitively and easily simply by drawing a character on the numeric keypad with a finger using only the numeric keypad as an input device. Preferably, the input data conversion unit sets a direction of up to eight keys adjacent to an arbitrary key of the input device as a basic vector. By this means, characters drawn by hand on a plurality of keys are converted into vectors.

Preferably, the basic vectors are upward basic vectors, downward basic vectors, left basic vectors, right basic vectors, right diagonal upward basic vectors,
It consists of a diagonally right downward basic vector, a diagonally left upward basic vector, and a diagonally left downward basic vector. By combining the above basic vectors, it is possible to convert the lines constituting the character into vectors.

[0007] Preferably, one vector of a key pressing locus is obtained from a relationship between a combination of the basic vectors and a key pressing time of the input device. By this means, it is possible to discriminate between vectors based on the relationship between the time when a key is released once and the time when the next key is pressed, and it is possible to obtain the number of strokes of a character.

[0008] Preferably, the combination of the upward basic vector and the downward basic vector, or the combination of the left basic vector and the right basic vector is used to determine one of the key press trajectories.
First, a curve is required. By this means, a curve can be identified from a vertical line, a horizontal line, and an oblique line, and a rounded character can be recognized.

Preferably, the reference data storage unit prepares a plurality of line features constituting the character with respect to the same character, and more preferably, a vertical line, a horizontal line, and a vertical line constituting the character.
Characters are grouped by using at least one of the following: diagonal line, type of line like curve, number of vertical lines, number of horizontal lines, number of diagonal lines, number of strokes, first pressed key in stroke order, last pressed key I do.

By this means, character information is provided in a table of a plurality of reference data having different concepts, so that accurate character recognition can be performed even with an input device having a small number of keys. Preferably, the reference data storage unit further groups keys adjacent to each pressed key with respect to the key sequence of the input device pressed according to the writing order of each character, and weights the keys in the group. Is stored, the character estimation unit refers to the reference data stored in the reference data storage unit from the input data of the pressed key input from the input data conversion unit,
The character is recognized by guessing the pressed key.

By this means, the input ten key 12
By storing the order of the keys, the comparison with the reference data in the writing order becomes possible. Preferably, the input device comprises at least two rows and two rows of keys. More preferably, the input device includes three rows of keys and four rows of keys. This means can be applied to various input devices.

Preferably, the input data conversion unit converts input data of English characters drawn by hand on a plurality of keys into a vector. More preferably, the input data conversion unit converts input data of katakana characters drawn by hand on a plurality of keys into vectors. By this means, input can be easily performed according to the user's preference.

Further, the present invention relates to a method for inputting and recognizing characters in a small electronic device having an input device comprising a plurality of keys, wherein the input of a key which is sequentially pressed when a character is drawn by hand on the plurality of keys. Converting the data into vectors, storing reference data in which characters are grouped by the characteristics of the lines constituting the characters as a plurality of tables, and referencing the tables to correspond to the converted vectors. And a step of estimating and recognizing a character from a drawn line.

According to this means, similarly to the above-described invention, it is possible to input a large number of characters at low cost, intuitively and easily simply by drawing a character on the numeric keypad with a finger using only the numeric keypad as an input device.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a view schematically showing a small electronic apparatus according to the present invention. The character input / recognition system for a small electronic device shown in FIG. 1 includes a small electronic device 11, a numeric keypad 12 used as an input device of the small electronic device 11, and a display for displaying characters input from the numeric keypad 12. And 13.

The numeric keypad 12 is an input device comprising a total of twelve keys of three horizontal rows and four vertical rows to which * and # are added in addition to numeric keys of 1 to 9, 0. As shown in the figure, the ten-keys 12 are sequentially pressed by handwriting, that is, by directly drawing, for example, an English character (alphabet) 14 of “A” on the ten-keys 12 with the fingers of the hand. The input data of the numeric keypad 12 is compared with a plurality of reference data of character information prepared in advance, the result is calculated, the character is recognized, and the display unit 13 displays the English character "A".

Here, the English characters 14 refer to the letters A to Z of the uppercase alphabet. This makes it possible to input a large number of English characters at low cost, intuitively and easily, using only the numeric keypad 12 as an input device of the small electronic device 11. The details will be described below.

FIG. 2 is a block diagram showing a schematic configuration of a character input / recognition system of a small electronic device. As shown in the figure, a control unit 100 is provided in the small electronic device, and the control unit 100 controls the small electronic device, for example, controls the ten keys 12, the display unit 13, and the like. Further, the control unit 100 is provided with a recognition unit 101, and the recognition unit 101 performs recognition by inputting English characters drawn directly from the numeric keypad 12.

The recognition unit 101 has an input data conversion unit 102
Is provided, and the input data conversion unit 102
It decomposes input data of English characters directly drawn with the fingers of the hand and converts them into vectors. Further, the recognition unit 101 is provided with a reference data storage unit 103, and the reference data storage unit 1
Reference numeral 03 stores reference data for converting the input data of the input data conversion unit 102 into a vector, and a plurality of reference data for grouping the English characters by the characteristics of the lines constituting the English characters.

Further, the recognition unit 101 includes a character estimation unit 10.
4 is provided, and the character guessing unit 104 directly draws an English character drawn on the numeric keypad 12 from the reference data stored in the reference data storage unit 103 with the finger of the hand based on the result of conversion of the input data conversion unit 102 into a vector. Guess the character. Recognition unit 10
1 displays the English characters guessed by the character guessing unit 104 on the display unit 13.

The control unit 100 is provided with a Japanese conversion unit 105. The Japanese conversion unit 105 converts the English characters A to Z obtained from the character estimation unit 104 of the recognition unit 101 into Japanese by Roman conversion. Then, it is displayed on the display unit 13. Note that the English characters may be displayed as an English sentence without passing through the Japanese conversion unit 105. FIG. 3 is a diagram illustrating the operation of the input data conversion unit 102. As shown in this figure, the input data conversion unit 102
With the key (here “5”) as the base point
Four types of straight line vectors, in which keys in the direction (2, 4, 8, 6) are pressed, and four types of diagonal vectors, in which keys in (1, 7, 9, 3) are pressed, are shortened to a total of eight types. As a (basic) vector, one vector of the key press locus is obtained from a plurality of combinations of these short vectors and the relationship of time.

In the following description, the key numbers 1 to 9 of the numeric keypad 12 are left as they are, the key number of * is 10, the key number of 0 is 11, and the key number of # is 12.
And the processing is performed. The direction of the short vector is determined as follows based on the difference Δ between the adjacent key number and the base key number.

FIG. 4 is a diagram for explaining short vectors converted by the input data conversion unit 102. As shown in this figure,
If the difference Δ = −3 between the adjacent key number and the base key number,
The short vector created by the adjacent key and the base point key is a straight line vector and is upward (upward and downward).

When Δ = 3, the short vector is a straight line vector and points downward (short downward). If Δ = -1, then
The short vector is a straight line vector and is directed leftward (short left).
When Δ = 1, the short vector is a straight line vector and is directed rightward (short right). In the case of Δ = −2, the short vector is an oblique vector and is directed obliquely upward (right obliquely upward short).

When Δ = 2, the short vector is a diagonal vector and is directed downward and left diagonally (left diagonally downward short). Δ = -4
In the case of (1), the short vector is an oblique vector and is directed obliquely upward to the left (left obliquely upward short). When Δ = 4, the short vector is an oblique vector and is directed obliquely downward and to the right (obliquely downward short).

FIG. 5 is a diagram for explaining a plurality of combined vectors related to the short vector. As shown in this figure,
An upward straight line (upper) is obtained by a combination of 2-4 upward short vectors (upper and lower). Downward straight line (bottom)
Is obtained by a combination of 2 to 4 downward short vectors (lower short).

A rightward straight line (right) is obtained by combining a few rightward short vectors (right short). The straight line to the left (left) is obtained by a combination of a few short left vectors (short left). The oblique line upward diagonally right (diagonally upper right) is obtained by a combination of 2 to 6 short diagonally upward right vectors (diagonally right upper diagonal), short upward vector (upper diagonal short), and right vector (right diagonal short).

The diagonal line obliquely downward to the left (diagonally below left) is 2 to
6 is obtained by a combination of the left diagonally downward short vector (diagonally left lower short), the downward short vector (lower short), and the leftward vector (left short). Diagonal line diagonally upward to the left (diagonally upward to the left)
Is the left diagonally upward short vector of 2 to 6 (left diagonally upward short),
It is obtained by a combination of an upward short vector (upper short) and a leftward vector (left short).

The diagonal lines obliquely downward and to the right (obliquely downward to the right) are 2 to
6 is obtained by a combination of the right diagonally downward short vector (left diagonally lower short), the downward short vector (lower short), and the rightward vector (left short). The vector of the curve is obtained by a combination of an upward short vector (upper short) and a downward short vector (lower short).

Further, a curve is obtained by a combination of a short right vector (right short) and a left short vector (left short). In addition, the curve is an upward short vector (upper and lower),
It is obtained by a combination of a downward short vector (downward short), a rightward short vector (right short), and a leftward short vector (left short).

FIG. 6 is a diagram for explaining the relationship between the combination of short vectors and time. As shown in the figure, it is possible to determine an oblique vector or a straight line vector from the relationship between the point at which the key is pressed and the time. In addition, the relationship between the time from when the key is released once and the time when the next key is pressed allows the vectors to be distinguished from each other.
It is possible to determine the number of strokes of a character.

When the above curve is drawn, the ten key 12
Since there are only twelve, from the relationship between this and time, it starts in a certain direction, and the direction of the opposite phase appears immediately. Thus, it can be expected that the input is drawing a semicircle or a circle. FIG. 7 is a diagram illustrating reference data stored in the reference data storage unit 103 in FIG. 2 and illustrating an example of reference data for converting input data into a vector.

As shown in the figure, the reference data storage 10
Reference numeral 3 designates 16 types having eight directions and two sizes used to decompose characters prepared in groups and prepared in advance for comparison with handwritten characters input to the numeric keypad 12 by vectors. This shows the vector of. In the reference data storage unit 103, "1" and "2" for rightward and leftward vectors, "3" and "4" for rightward and leftward short vectors, and "5" for rightward oblique upward and right oblique downward vectors, "6", right diagonally upward, left diagonally downward short vector "7", "8", left diagonally upward, right diagonally downward vector "9", "10", left diagonally upward, right diagonally downward short vector Contains "11", "12", upwards,
"13" and "14" for the downward short vector, "15" and "16" for the upward and downward vector, and "1" for the curve
7 "is assigned.

Fig. 8 is a view for explaining an example of inputting the English character "A" from the numeric keypad 12 of Fig. 1. As shown in this figure, as an example, contact of a finger of the hand with the numeric keypad 12 of a small electronic device is shown. As a result, the hatch portion 22 indicated by the arrow 23
If the English character "A" is drawn like this, the numeric keypad 12 becomes "* → 7 → 4 → 2, 2 → 6 → 9 → #, 7 → 8 →
9 ”in that order. Here, “,” indicates a short pause time until the next numeric key 12 is pressed.

Information generated by this pressing is transmitted to the control unit 100.
Is output to the input data converting unit 102 of the recognizing unit 101. FIG. 9 is a flowchart illustrating the operation of the input data conversion unit 102 for the English character input of FIG. In steps S11 and S12, the input data conversion unit 102 sequentially inputs information of * key and 7 key pressed.

In step S13, the input data conversion unit 102 obtains the difference Δ = −3 between the number of the 7 key and the * key number (10) from the information of the * key and the 7 key pressed, and refers to the reference data storage unit 103. Then, the short vector formed by pressing the * key and the 7 key is determined as the upward short vector (13). In step S14, next, the user inputs the 4 key press information.

In step S15, the difference Δ = −3 between the 4 key number and the 7 key number is obtained, and the reference data storage unit 1
With reference to 03, the short vector formed by pressing the 7th key and the 4th key is determined as the upward short vector (13). In step S16, the user inputs the key press information. In step S17, the difference Δ = −2 between the number of the 2 key and the number of the 4 key is obtained, and by referring to the reference data storage unit 103,
The short vector formed by pressing the 4 key and the 2 key is determined to be a right diagonally upward short vector (7).

In step S18, the input data converter 102 confirms that the finger has been released from the numeric keypad 12 based on the short pause time until the next numeric keypad 12 is pressed. In step S19, the vector formed by pressing the * key to the two keys is formed by a combination of two upward short vectors (13) and one diagonally right upward short vector (7). Conversion unit 10
2 judges this vector to be a diagonally rightward upward vector (5).

In steps S20 and S21,
Information on the pressing of the 2 key and the pressing of the 6 key is sequentially input. In step S22, the input data conversion unit 102 calculates the difference Δ = −4 between the number of the 6 key and the 2 key number from the information of the 2 key and 6 key presses, refers to the reference data storage unit 103, and It is determined that the short vector formed by pressing the 6 key is a short obliquely downward short vector (12).

In step S23, next, the key press information is input. In step S24, the difference Δ = 3 between the number of the 9th key and the number of the 6th key is obtained, and the short vector formed by pressing the 6th key and the 9th key is referred to the reference data storage unit 103, and the downward short vector (14 ).
In step S25, # key press information is input.

In step S26, the difference Δ = 3 between the # key number (12) and the 9 key number is determined, and the short vector formed by pressing the 9 key and the # key is referred to the reference data storage unit 103. Is determined as a downward short vector (14). In step S27, the input data conversion unit 102
Confirms that the finger has been released from the numeric keypad 12 based on the pressing time until the next numeric keypad 12 is pressed.

In step S28, the vectors formed by pressing the 2 key to the # key are one diagonally downward right short vector (12) and two short downward vectors (1
4), the input data conversion unit 102 determines that the vector is a diagonally rightward downward vector (10). Next, in steps S29 and S30, information on the depression of the 7 key and the depression of the 8 key are sequentially input.

In step S31, the input data conversion section 102 obtains the difference Δ = 1 between the number of the 8 key and the 7 key number from the information on the press of the 7 key and the 8 key, and
With reference to 03, a short vector formed by pressing the 7th key and the 8th key is determined to be a rightward short vector (3). Next, in step S32, 9-key press information is input.

In step S33, the difference Δ = 1 between the number of the 9th key and the number of the 8th key is obtained, and the reference data storage 10
With reference to 3, the short vector formed by pressing the 8th and 9th keys is determined to be a rightward short vector (3). In step S34, the input data conversion unit 102 confirms that the finger has been released from the numeric keypad 12.

In step S35, since the vector formed by pressing the keys 7 to 9 is formed by a combination of two rightward short vectors (3), the input data conversion unit 102 Judge as the diagonally downward vector (1). FIG. 10 is a diagram illustrating an example of conversion of the English character “A” into a vector by the conversion operation of the input data conversion unit 102 in FIG. 9.

The English character "A" is designated by reference numeral 3-1 in the figure.
As shown in (3), it consists of a diagonally rightward upward vector, a diagonally downward right vector, and a rightward vector, and as represented by reference numeral 3-2, is represented by (5, 10, 1) when it is represented by an assigned number. As described above, the English character “A” is converted into a plurality of vectors, and in the above case, the number of strokes of the English character “A” is 3
It is a picture.

FIG. 11 is a diagram for explaining an example of conversion of the English character "B" into a vector. The input data of the English character "B" to the numeric keypad 12 is (1 → 4 → 7 → *, 1 → 2).
→ 6 → 8 → 7,4 → 5 → 9 → 0 → *)
As shown in this figure, when this is represented by a vector, it is composed of a downward vector, a curve, and a curve, and when represented by an assigned number,
(15, 17, 17).

Further, another example of the English character "B" will be described. When the input data of the English character “B” to the numeric keypad 12 is (1 → 4 → 7 → *, 1 → 2 → 3, 3 →
6, 6 → 5, 8 → 9, 9 → #, # → 0 → *), as shown in the figure, if this is represented by a vector, a downward vector, a rightward vector, a downward short vector, a leftward vector It consists of a short vector, a short right vector, a short short vector, and a left vector.
(16, 1, 14, 4, 3, 14, 2).

Further, another example of the English character "B" will be described. When the input data of the English character “B” to the numeric keypad 12 is (1 → 4 → 7 → *, 1 → 2 → 3, 3 →
5, 8 → #, # → 0 → *), as shown in this figure, when this is represented by a vector, a downward vector, a rightward vector, a left oblique downward short vector, a right oblique downward short vector, a leftward upward It is composed of vectors, and is represented by (16, 1, 8, 12, 2) when represented by the allocated number.

Next, a table is formed in the reference data storage unit 103 by decomposing all 24 English characters to be prepared into lines and dividing them into the following groups according to the characteristics of the lines. FIG. 12 is a diagram illustrating reference data stored in the reference data storage unit 103 in FIG. 2 and illustrating a table example of reference data in which characters correspond to lines.

As shown in the figure, the reference data storage 10
3 is composed of only vertical, horizontal, and diagonal straight lines.
The English characters 6 are: In particular, English characters composed only of vertical lines and horizontal lines, E, F, H, I, L, T, B,.

V, W, X,..., Especially English characters consisting only of diagonal lines + horizontal lines, A, Z,..., Especially English characters consisting only of diagonal lines + vertical lines, K, M, N, Y, ...

In particular, B, D, G, J, P, U,... English composed of straight lines + curves + diagonal lines, as English characters composed of diagonal lines + vertical lines + horizontal lines B,. As characters

R,... Are stored as reference data grouped as C, O, Q, S. It should be noted that, for example, the English character “B” including a straight line and a curve includes a curve as shown in FIG.
When inputting, there are conversion examples into several lines. For example, the reference data storage unit 103 stores not only straight lines + curves but also vertical lines + horizontal lines and oblique lines + vertical lines + horizontal lines as reference data. Is stored in In this way, errors due to input can be minimized.

FIG. 13 shows reference data stored in the reference data storage unit 103 of FIG. 2, and is a view for explaining another example of a table of reference data for making characters correspond to lines. As shown in this figure, the number of vertical lines of English characters “0”, “1”,
English characters are grouped into a group of “2”. FIG.
Reference numeral 4 denotes reference data stored in the reference data storage unit 103 in FIG. 2, and is a diagram illustrating another example of a table of reference data that associates characters with lines.

As shown in this figure, English characters are grouped into groups of the number of horizontal lines "0", "1", "2" and "3" of English characters. FIG. 15 is a diagram illustrating reference data stored in the reference data storage unit 103 in FIG. 2 and illustrating another example of a table of reference data that associates characters with lines. As shown in the figure, the number of diagonal lines of English characters “0”,
English characters are grouped into groups of “1”, “2”, and “4”.

FIG. 16 shows reference data stored in the reference data storage section 103 of FIG. 2, and is a view for explaining another example of a table of reference data for making characters correspond to lines. As shown in this figure, the number of strokes of English characters “1”, “2”, “3”,
English characters are grouped into a group of “4”. FIG.
Reference numeral 7 denotes reference data stored in the reference data storage unit 103 in FIG. 2, and is a diagram illustrating another example of a table of reference data that associates characters with lines.

As shown in this figure, for each key of the ten keys 12, English characters to be pressed first in the writing order when inputting English characters are grouped. Similarly, the numeric keypad 12
For each key, the English character pressed last in the writing order when the English character is input is grouped. In this table, when inputting English characters, it is possible to predict the probable candidate of the key of the ten key 12 to be pressed first in the writing order and the probable candidate of the key of the ten key 12 to be pressed last. It is possible to make full use of a limited number of twelve numeric keys 12 which are only 12 by grouping these influential candidates into pairs.

As described above, the character information is provided in a table of a plurality of reference data having different concepts so that accurate character recognition can be performed with only a total of 12 key input devices. FIG. 18 shows the reference data storage unit 103 of FIG.
FIG. 4 is a diagram illustrating reference data stored in the storage device and illustrating an example of reference data for converting input data into lines.

As shown in this figure, one English letter “A”
There is a candidate 1 and a candidate 2 in the writing order of, for example, a key (*, 7, 8) adjacent to the key of the numeric keypad 12 expected to be pressed first in the writing order related to the candidate 1.
Key) is one first group. The group of keys (keys 7 and 8) expected to be pressed second is the second group.

Subsequently, a third, fourth,... Group is formed.
Weights 10, 9, and 8 are assigned to the *, 7, and 8 keys of the first group in consideration of the direction in which characters are drawn in the groups. Furthermore, weights 10 and 9 are assigned to the 7 and 8 keys of the second group. By storing the input key sequence of the numeric keypad 12, it is possible to compare with the reference data in the writing order.

Here, a simple comparison between the key pressed first and the key pressed last and the reference data becomes possible. Next, from the reference data stored in the reference data storage unit 103 based on the result of conversion of the input data to the vector by the input data conversion unit 102, an English character directly drawn on the numeric keypad 12 with the finger of the hand, for example, “A” For an example to guess,
Give an explanation.

FIG. 19 is a diagram for explaining the outline of the character estimating unit 104 of the recognizing unit 101 in FIG. The conversion result of the input data conversion unit 102 shown in FIG.
Consists of horizontal and oblique lines. The character guessing unit 104 finds that the English characters “A” and “Z” are closest to the reference data in the reference data storage unit 103 shown in FIG.
Give them the best 10 points. The next closest English letter "H" is given 8 points. The lowest letter 0 is given to the English letter "C" containing the curve. In this way, a group that moves away from the closest group is given a lower score as the distance increases.

Next, according to the conversion result of the input data conversion unit 102, the number of horizontal lines of the English character "A" is "1", the number of vertical lines is "0", and the number of oblique lines is "2". It is. The character guessing unit 104 obtains the guess result 2 from the reference data in the reference data storage unit 103 shown in FIG. 13, FIG. 14, and FIG.
The English letter "A" is closest, so give 10 points. The next closest English letter "H" is given 8 points.

Next, in the conversion result of the input data conversion unit 102, the number of strokes of the English character "A" is "3". The character guessing unit 104 obtains, from the reference data in the reference data storage unit 103 shown in FIG.
These are given 10 points because "H" is closest.

Next, according to the conversion result of the input data conversion unit 102, the ten key 12
Is “2”. The character guessing unit 104 stores the reference data storage unit 103 shown in FIG.
, The English characters "A" and "I" are closest, and thus are given 10 points.

Next, the character guessing unit 104 determines that the English character “A” is closest in writing order from the reference data in the reference data storage unit 103 shown in FIG.
This is awarded 10 points. The character guessing unit 104 adds the guessed results 1 to 5, and guesses and confirms that the English character having the largest added value is the input English character.

As described above, once conversion into vector information rounds off errors such as simultaneous depression of the numeric keypad 12 at a plurality of times and key depression of an unintended location, it is necessary to accurately identify expected English characters. Is possible. In the above description, input and recognition of English characters A to Z have been described, but input and recognition of katakana characters can be similarly performed. This makes it possible to adapt to user preferences.

The ten key 12 has been described as being composed of a total of twelve keys in three columns and four rows, but the present invention is realized by inputting and recognizing vertical, horizontal and oblique straight lines. , Two columns vertically and two rows horizontally, a total of four keys can be similarly realized. Thereby, it can be applied to various input devices.

[0070]

As described above, according to the present invention,
With only the numeric keypad as an input device, characters drawn with the fingers of the hand on the numeric keypad are converted into vectors and characters are recognized from the converted vector, so simply drawing characters on the numeric keypad is inexpensive Many characters can be input intuitively and easily.

Further, character information is provided in a plurality of reference tables having different concepts, and the character is guessed and confirmed from the converted vector by referring to the reference table. Draw on
It becomes possible to input.

[Brief description of the drawings]

FIG. 1 is a view schematically showing a small electronic apparatus according to the present invention.

FIG. 2 is a block diagram showing a schematic configuration of a character input / recognition system of the small electronic device.

FIG. 3 is a diagram illustrating an operation of an input data conversion unit 102.

FIG. 4 is a diagram illustrating a short vector converted by an input data conversion unit 102.

FIG. 5 is a diagram illustrating a plurality of combined vectors related to a short vector.

FIG. 6 is a diagram illustrating a relationship between a combination of short vectors and time.

7 is a diagram illustrating reference data stored in the reference data storage unit 103 in FIG. 2 and illustrating an example of reference data for converting input data into a vector.

FIG. 8 is a diagram illustrating an example of inputting an English character “A” from the numeric keypad 12 in FIG. 1;

9 is a flowchart illustrating the operation of the input data conversion unit 102 for inputting English characters in FIG. 8;

10 is a diagram illustrating an example of conversion of an English character “A” into a vector by the conversion operation of the input data conversion unit 102 in FIG. 9;

FIG. 11 is a diagram illustrating an example of conversion of an English character “B” into a vector.

12 is a diagram illustrating reference data stored in the reference data storage unit 103 in FIG. 2 and illustrating a table example of reference data in which characters correspond to lines.

13 is a diagram illustrating another example of a table of reference data which is reference data stored in the reference data storage unit 103 of FIG. 2 and which associates a character with a line.

14 is a diagram illustrating reference data stored in the reference data storage unit 103 in FIG. 2 and illustrating another example of a table of reference data that associates characters with lines. FIG.

15 is a diagram illustrating another example of a table of reference data which is reference data stored in the reference data storage unit 103 of FIG. 2 and which associates characters with lines.

16 is a diagram illustrating another example of a table of reference data that is reference data stored in the reference data storage unit 103 of FIG. 2 and that associates a character with a line.

17 is a diagram illustrating reference data stored in the reference data storage unit 103 in FIG. 2 and illustrating another example of a table of reference data that associates a character with a line.

18 is a diagram illustrating reference data stored in the reference data storage unit 103 in FIG. 2 and illustrating an example of reference data for converting input data into lines.

FIG. 19 is a character estimating unit 10 of the recognizing unit 101 in FIG.
FIG. 4 is a diagram illustrating an outline of No. 4;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 ... Small electronic equipment 12 ... Numeric keypad 13 ... Display part 100 ... Control part 101 ... Recognition part 102 ... Input data conversion part 103 ... Reference data storage part 104 ... Character estimation part 105 ... Japanese conversion part

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) H04M 1/02 G06F 3/023 310K 1/23

Claims (13)

[Claims] 1. A character input / recognition system for a small electronic device having an input device composed of a plurality of keys. An input data conversion unit for converting, a reference data storage unit for storing reference data in which characters are grouped by the characteristics of lines constituting the characters as a plurality of tables, and a reference to the table of the reference data storage unit. A character estimating unit for estimating and recognizing a character from a line corresponding to the vector converted by the input data converting unit. 2. The small electronic apparatus according to claim 1, wherein the input data conversion unit sets a direction of up to eight keys adjacent to an arbitrary key of the input device as a basic vector. Character input and recognition system. 3. The basic vector includes an upward basic vector, a downward basic vector, a left basic vector, a right basic vector, a right oblique upward basic vector, a right oblique downward basic vector, a left oblique upward basic vector, and a left oblique downward basic vector. 3. The method of claim 2, wherein
2. A character input / recognition system for a small electronic device according to claim 1. 4. A key press trajectory according to a relationship between a combination of the basic vectors and a key press time of the input device.
3. The method according to claim 2, wherein two vectors are determined.
2. A character input / recognition system for a small electronic device according to claim 1. 5. The method according to claim 4, wherein a curve is obtained as one of key press trajectories from a combination of an upward basic vector and a downward basic vector, and a combination of a left basic vector and a right basic vector. Character input and recognition system for small electronic equipment. 6. The character input / output device according to claim 1, wherein the reference data storage unit prepares a plurality of features of a line constituting the character for the same character. Recognition system. 7. The reference data storage unit stores first, second, third, and fourth line types, such as vertical lines, horizontal lines, diagonal lines, and curved lines, vertical lines, horizontal lines, diagonal lines, strokes, and strokes that form characters. The character input / recognition system for a small electronic device according to claim 1, wherein the characters are grouped by using at least one of the first pressed key and the last pressed key. 8. The reference data storage unit further groups keys adjacent to each pressed key with respect to a key order of the input device that is pressed according to a writing order of each character, and stores a key within the group. The character estimating unit refers to the reference data stored in the reference data storage unit from the input data of the press key input from the input data conversion unit, and presses the press key. The character input / recognition system for a small electronic device according to claim 1, wherein the character is recognized by inferring the following. 9. The character input / recognition system for a small electronic device according to claim 1, wherein the input device comprises at least two rows and two rows of keys. 10. The character input / recognition system for a small electronic device according to claim 1, wherein the input device comprises three columns and four columns of keys. 11. The character input device according to claim 1, wherein the input data conversion unit converts input data of English characters drawn by hand on a plurality of keys into a vector.・ Recognition system. 12. The character input device according to claim 1, wherein the input data converter converts input data of katakana characters drawn by hand on a plurality of keys into a vector.・ Recognition system. 13. A character input / recognition method for a small electronic device having an input device including a plurality of keys, wherein input data of keys sequentially pressed when a character is drawn by hand on the plurality of keys is converted into a vector. Converting, and storing, as a plurality of tables, reference data in which characters are grouped by the characteristics of the lines constituting the characters; and referring to the tables, converting characters from the lines corresponding to the converted vectors. And a method for inputting and recognizing characters in a small electronic device.