JP2804660B2 - Document processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document processing apparatus having a function of recognizing handwritten information.

[0002]

2. Description of the Related Art A well-known keyboard and mouse are mainly used for input of a conventional document processing apparatus. However, these operations are unnatural to human sensations and require effort to get used to. . For this reason, a pen-type interface for designating a position with a pen or writing and inputting characters by a character recognition function on a tablet in which display and input are integrated has attracted attention.

[0003] In this type of character input, a character is written in a specific area (character entry frame), and a recognition result (character) is input at a cursor position.
Although there is a type in which characters are input at the writing position, inputting characters at the writing position is more suitable for human senses and excellent in operability. In this type of document processing apparatus, when a character is input at a writing position, the input position (line, digit) is determined based on the center of the writing character or the center of the left side of a rectangle surrounding the writing character, and the reference position ( x, y) is determined by determining at which line or column in the document the character string is input. Similarly, when a character string is input, the input position (x, y) is determined based on the center position of a rectangle surrounding the writing character string. In general, the input position of the first character of the character string is obtained.

However, when the character input position is obtained in this manner, if the size of the written character is equal to or smaller than the character input and displayed in the document (hereinafter, display character), In the case where the displayed characters and the line spacing are very large, the input position calculated by the above method and the input position intended by the writer match, but the written character is larger than the displayed character, and the character spacing and line spacing are larger. If it is narrow, there is a problem that the input position calculated by the above method and the input position intended by the writer do not always match.

The deviation between the calculated input position and the input position intended by the writer becomes more remarkable as the difference between the size of the character to be written and the size of the displayed character becomes larger, and as the space between characters and lines becomes narrower. . The size of display characters, the space between characters, and the space between lines cannot be made too large in view of the amount of information as a sentence displayed on a display screen of a limited size. Also, if the character size that can be written is limited to a small size, it becomes difficult to write.

For this reason, in a conventional document processing apparatus using pen input, a character is rarely input at a position different from an intended position by writing over a display range of two or more display characters. Absent. FIG. 2 shows an example in which a character is input at a position different from the intended position. FIGS. 3A, 3C and 3E show a character string "Today is fine" displayed on the display device. In order to input a character string "Tenari" following this character string, characters are displayed. This shows a state where column B is written. The result of inputting "Tenari" at the character input position obtained based on the written position is shown. The character input position is a line and a digit position Q corresponding to the reference position with reference to the center position P of the left side of the rectangle surrounding the written character string. In FIGS. 6A, 6C and 6E, the reference position P of the written character string is one character (half-width or full-width) from the line immediately after the already input character string (right) and the digit position R. Right, left,
Since it is shifted down by one line, the character string C of the character recognition result "heaven" is input at that position, and FIG.
(D) and (f) result.

[0007] As described above, in order to input data following the already-input character string, the intention is to write immediately after (right) the already-input character string. In some cases, a blank space is opened, the character at the end of the input character string is input, and the character at the end of the input character string is erased by mistake, or input is performed at a position shifted by one line.
In addition, it is extremely nervous to write a character at an appropriate position assuming the input position, which is extremely inconvenient for the writer.

[0008]

SUMMARY OF THE INVENTION A document processing apparatus according to the present invention realizes a document processing apparatus capable of inputting a written character at an intended position even if a display character size, a character bitch, and a line pitch are small. .

[0009]

A document processing apparatus according to the present invention comprises: coordinate input means for inputting handwriting coordinates with a pen;
Recognition means for recognizing a character from the handwriting coordinate data input by the coordinate input means; character input position determination means for determining an input position of the character recognized by the recognition means; Display means for displaying the handwriting,
A character information storage unit configured to store character information such as a character code recognized by the recognition unit and position information of the character, wherein the character input position determination unit includes the character information storage unit Character position information reading means for reading out the character position information of the already input character set stored in the storage unit, and the positional relationship between the newly input character set based on the character position information read by the character position information reading means. A determination unit for determining is provided, and an input position of a newly input character set is determined based on a determination result of the determination unit.

[0010]

In the document processing apparatus according to the present invention, a character is recognized by the recognition means from the handwriting coordinates of the pen input by the coordinate input means, and the input position of the recognized character is already input by the character input position determination means. By determining the character set based on the positional relationship or the linguistic relationship between the written character set and the written character set, the character can be input at the position intended by the writer.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment according to the present invention will be described below in detail with reference to FIGS. FIG. 1 is a block diagram showing the configuration of an embodiment of the document processing apparatus of the present invention. In the figure,
The display-integrated tablet 1 has a display unit 1 composed of an LCD.
a, a pen coordinate detecting unit 1b composed of a tablet, and a coordinate of the display unit 1a and a pen coordinate detecting unit 1b.
Have a one-to-one correspondence.

When a character string is written on the display-integrated tablet 1 with an input pen, the pen coordinate detector 1b detects the coordinates (x, y) of the pen at regular time intervals, and displays a time series indicating the pen's handwriting. (Hereinafter referred to as handwriting coordinate data), and the time during which the pen is separated from the tablet 1 (hereinafter referred to as pen-up time) between handwritten strokes is sent to the document editing apparatus 2.

The document editing apparatus 2 comprises a CPU and has a document memory 2a comprising a RAM for managing documents. Pen coordinate detector 1b of display-integrated tablet 1
Tablet 1 based on handwriting coordinate data received from
Is displayed on the display unit 1a. Next, the handwriting coordinate data and the pen-up time are sent to the character recognition device 5 and the handwriting coordinate data is sent to the character input position determination device 3.

The character recognition device 5 comprises a character recognition processor, and the handwriting coordinate data inputted from the document editing device 2;
Character extraction and character recognition are performed based on the pen-up time, and the recognition result is returned to the document editing device 2. That is,
First, the vertical / horizontal ratio of a rectangular area surrounding a stroke sequence assumed to be equivalent to one character is obtained, and characters are cut out based on the result and the pen-up time between strokes. Then, the character is recognized with the cut-out character degree. The character recognition dictionary 5a attached to the character recognition device 5 stores, for each character to be recognized (Kanji, Hiragana, English alphabet, number, symbol), the normal stroke number of the character, the shape of the stroke constituting the character, the positional relationship of the stroke, Character features such as the distribution of the distance from each coordinate of the four sides of the rectangular area surrounding each writing character collected as the standard character pattern to the stroke are registered. The character recognition device 5 calculates the characteristic amount of the character from the handwriting coordinate data of each character, and compares the characteristic amount with the characteristic amount of each character in the character recognition dictionary 5a. The character having the highest matching degree is set as the recognition result, and the character recognition result (character string recognition result) for all cut out characters is sent to the document editing device 2.

The character input position determining device 3 is composed of a CPU and determines the input position of the written character string (display unit 1b) based on the positional relationship between the written character string and the already-input character string and the language relationship.
Upper row, column position). That is, the apparatus 3 sets the predetermined area W1 near the head (head of the character string) of the rectangular area surrounding the handwritten character string and the tail (the tail of the character string) from the handwriting coordinate data of the character string received from the document editing apparatus 2. Is set in the vicinity of. This predetermined area may be variable in size and position depending on the size of writing of the character, or may be fixedly provided.

If the predetermined area W1 or the predetermined area Z1 contains a part of an already input character string (a character string in the document memory and displayed on the display unit 1b), the character string is continued from the already input character string. Input position (line,
Digit). If neither the predetermined area W1 nor the predetermined area Z1 contains a part of the already input character string, the character input position (line, line, or the like) is determined based on the center position of the left side (head) of the rectangular area surrounding the writing character string. Digit). The character input position determining device 3 refers to the information on the contents and position of the already-input character string, sends an instruction for referring to the information to the document editing device 2, and refers to the document memory 2a by the document editing device 2. This is done by obtaining the results.

Further, when the device 3 determines the input position of the handwritten character string, it returns the result to the document editing device 2. Thereby, the document editing device 2 inputs the character string recognition result (character string) received from the document character recognition device 5 to the input position of the determined writing character string, and updates the contents of the document memory 2a. The updated contents of the document memory 2a are displayed on the display unit 1b.

Next, the character input position determination operation of the character input position determination device 3 will be described with reference to FIGS. FIG.
(A), the already input character string A “Today is fine” is displayed on the display unit 1b, and the character string B is displayed near the rear of the character string A.
This shows a state where "Tenari" is written. Since the character input position determination device 3 includes a part of the input character string A in the predetermined area W1 set at the head of the writing character string B, the character input position (the head position of the character string) A position Ra immediately after the column A is set. As a result, the recognition result C "Tenari" of the handwritten character string B is input at the position of Ra, and the content of the display unit 1b is as shown in FIG.

FIG. 4A shows a state where the input character string F "recognize" is displayed on the display section 1b and the character string D "character" is written near the front of the character string F. I have.
The character input position determination device 3 has already input the character input position (the tail position of the character string D) because a part of the input character string F is included in the predetermined area Z1 set at the tail of the written character string D. Let it be the position Rf immediately before the character string F. As a result, the recognition result E "character" is input such that the trailing position of the recognition result E "character" of the written character string D is Rf, and the content of the display unit 1b is as shown in FIG.

As shown in FIGS. 3 and 4, only one of the predetermined areas W1 and Z1 contains a part of the input character string.
In addition, when the input character string included in W1 or Z1 is one character string, the input character string continuous to the writing character string is limited to one, so that the input position of the writing character string can be specified. . However, when a part of the input character string is included in both the predetermined areas W1 and Z1, or when a part of each of two or more input character strings is included in either the W1 or Z1, the handwriting is performed. Since there are two or more input character strings that are assumed to be continuous with the character string, the input position of the handwritten character string cannot be specified. For example, as shown in FIG. 5 (a), in the predetermined area W1 for the written character string B "Tenari", each of the input character string A "Today is fine" and the input character string F "Recognize" If a part of the input character string G “tomorrow is” is included in the predetermined area Z1, there are three candidates for the input position of the writing character string B, Ra, Rf, and Rg. Cannot be identified in one place.

When the input position of the handwritten character string cannot be specified in one place by the predetermined areas W1 and Z1 as described above,
The character input position determination device 3 narrows down the predetermined regions W1 and Z1 to narrow down the candidates for the input position of the handwritten character string. The predetermined areas W2 and Z2 in FIG.
Is an area where W1 and Z1 are smaller. Thereby, since a part of the input character string A is included only in the predetermined area W2, the input position of the handwritten character string can be specified to one position of Ra. As a result, when the recognition result C of the handwritten character string B is input, the content of the display unit 1b becomes as shown in FIG.

When the input position of the handwritten character string cannot be specified by the above-described processing, for example, as shown in FIG. 6A, the input character string A “Today is fine” is written in the predetermined area W2 (narrow area). Part of each of the input character strings F "recognize" is entered,
If there are two candidates for the input position of the written character string, Ra and Rf, the character input position determination device 3 determines the candidate based on the language processing relationship between the character string recognition result of the written character string and the already input character string. squeeze.

The language processing device 4 receives an arbitrary character string and refers to the attached grammar dictionary 4a and word dictionary 4b to extract a phrase, an independent word, a verb / adjective from the input character string. This is a device for judging the correctness of adjunct words such as endings, particles, and auxiliary verbs, and referring to the frequency of appearance of independent words.

The character input position determining device 3 converts a character string obtained by combining the recognition result (character string) of the written character string and the already input character string assumed to be continuous with the character string into a language by the language processing device 4. Is calculated, and a combination of the already-input character and the written character string having the highest evaluation score is obtained, and a position that is continuous with the already-input character string is determined as the input position of the written character string. In the example of FIG. 6A, FIG.
As shown in the figure, a character string obtained by combining the recognition result C of the written character string B "Tenari", the input character string A "Today is fine", and the input character string F "Recognize" Are obtained as evaluation scores Ha and He. Input character string A
The final character of "Today is fine""Haru" and the written character string "Tenari"
Is combined with the first character "ten", the word "clear" is obtained. However, a new word cannot be generated even when the already-input character string F "recognize" and the written character string "tenari" are combined, so that Ha> He It is clear that Therefore, the input position of the handwritten character string is determined to be Ra. As a result, when the recognition result C of the handwritten character string B is input, the content of the display unit 1b becomes as shown in FIG.

[0025]

According to the document processing apparatus of the present invention, a character string is already input, and a character string is written near the already input character string in order to input a character string following the character string. In this case, even if the display character size, character bitch, and line pitch are small, the character is input at a position intended by the character writer. This reduces the burden of specifying the writing position in character input, and improves operability.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a document processing apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing an operation example of a conventional document processing apparatus.

FIG. 3 is a view showing an operation example (setting of an input position based on a positional relationship) of the document processing apparatus according to the embodiment of the present invention;

FIG. 4 is a view showing an operation example (setting of an input position based on a positional relationship) of the document processing apparatus according to the embodiment of the present invention;

FIG. 5 is a view showing an operation example (setting of an input position based on a positional relationship) of the document processing apparatus according to the embodiment of the present invention;

FIG. 6 is a diagram showing an operation example (setting of an input position based on a language relation) of the document processing apparatus according to the embodiment of the present invention;

[Explanation of symbols]

 Reference Signs List 1 display / input integrated tablet 1a LCD display unit 1b pen coordinate detection unit 2 document editing device 2a document memory 3 character input position determination device 4 language processing device 4a grammar dictionary 4b word dictionary 5 character recognition device 5a character recognition dictionary

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-3-28987 (JP, A) JP-A-63-269266 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G06F 17/20-17/26

Claims (2)

(57) [Claims] 1. Coordinate input for inputting handwriting coordinates with a pen
Force means and handwriting coordinate data input by the coordinate input means
Recognition means for recognizing a character from
Character input position determining means for determining the input position of the character,
A table displaying the characters and the handwriting recognized by the recognition means;
Indicating means, the character code recognized by the recognizing means,
Character information storage for storing character information such as position information of written characters
And a character input position determining means, wherein the character information storage means
A sentence for reading out the stored character position information of the already input character set
Character position information reading means, and character position information reading means.
Newly entered character based on the issued character position information
A determination unit configured to determine a positional relationship with the set; and the determination unit may be written by the coordinate input unit.
From the handwriting data of the character string
An area is provided, and a sentence at the end of the already input character set is set in the predetermined area
If a part of the character is included, the written string is
As the character string continuing to the set, the end of the written character string
The recognition result of the first character is continuous with the last character of the character set
The character string input position is set so as to be the position, and a part of the last character of the already input character set is entered in the predetermined area.
If not, it is based on the handwriting position of the written character string.
The character string input position is set, and each trailing character of two or more input character sets is set in the predetermined area.
Of the recognition result of the written character string.
When the force position is not uniquely determined, the written character
Column recognition results and the existing positions at each of the input position candidates
Language processing means that performs language processing together with
Selecting a candidate for the input position based on the processing result.
Characteristic document processing device. 2. A coordinate input for inputting handwriting coordinates with a pen.
Force means and handwriting coordinate data input by the coordinate input means
Recognition means for recognizing a character from
Character input position determining means for determining the input position of the character,
A table displaying the characters and the handwriting recognized by the recognition means;
Indicating means, the character code recognized by the recognizing means,
Character information storage for storing character information such as position information of written characters
And a character input position determining means, wherein the character information storage means
A sentence for reading out the stored character position information of the already input character set
Character position information reading means, and character position information reading means.
Newly entered character based on the issued character position information
A determination unit configured to determine a positional relationship with the set; and the determination unit may be written by the coordinate input unit.
From the handwriting data of the character string
An area is provided, and the first sentence of the already input character set is provided in the predetermined area.
If any part of the character is included, the written character string
The tail of the written string as a continuous set
Is one character before the first character of the character set
Set the string input position so as to position a portion of the first character already input character set the predetermined region is input
If not, it is based on the handwriting position of the written character string.
To set the character string input position, and set the first character of each of two or more input sets in the predetermined area.
Partial input of the recognition result of the written character string
When the position is not uniquely determined, the written string
Recognition result and input already at the position of each of the input position candidates
Processing of language processing means that performs language processing together with character set
Select the input position candidate based on the result.
Document processing device.