JP3177045B2 - Online character recognition device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an online character recognition device capable of inputting, recognizing and editing a plurality of characters via a coordinate input device such as a tablet.

[0002]

2. Description of the Related Art In recent years, an online character recognition device that recognizes a character input by handwriting via a coordinate input device such as a tablet and uses it for document creation or the like has attracted attention. In the online character recognition device, when a character is written on a tablet, the character stroke is displayed as a handwriting and the character stroke sequence is recognized. The result obtained by character recognition is displayed in a font of an appropriate size of the system instead of a character stroke.

In the conventional online character recognition device, the following procedure is required when a character string is to be inserted later into the already input text displayed as described above.

(1) The device is notified of the insertion operation using a button operation or the like. (2) Specify a sentence into which a character string is to be inserted. (3) Specify the position where the character string is to be inserted in the input text. (4) Write the character string to be inserted.

[0005] However, such a device has a problem in that the above operations (1) to (4) need to be performed every time the insertion operation is performed. In other words, extra operations that have nothing to do with writing originally prevent the writer from thinking about writing. Such a problem hinders the spread of online character recognition devices.

[0006]

As described above, the conventional on-line character recognition device requires an operation for notifying the device of the insertion operation / insertion position when a character string is inserted into a sentence. This hinders the writer's thinking of writing sentences, and the large number of operations hinders efficient writing of sentences.

[0007] The present invention has been made in view of the above points, and has as its object to provide an online character recognition device capable of easily inserting a character string into an input sentence.

[0008]

SUMMARY OF THE INVENTION According to the present invention, there is provided an online character recognition apparatus for recognizing a character string input by handwriting from a coordinate input device.
A sentence coordinate storage unit for storing sentence coordinate data indicating a position of a character string input by handwriting from the coordinate input device; a position of a character string input by handwriting from the coordinate input device; and the sentence coordinate storage unit. A mode determining means for determining whether or not an input mode is an insertion mode in which the input character string is to be inserted into another character string, based on the stored sentence data of the input character string; The character recognition result of the extracted character string is extracted by a phrase unit to detect a phrase position in the character string, and obtained by the phrase extraction means when the mode determination unit determines that the insertion mode is set. Character string inserting means for inserting a character string being input at a phrase position of the inputted character string and a predetermined phrase position determined based on the position of the inputted character string. To.

[0009]

According to such a configuration, when a character string is written near a sentence that has already been input, it is determined that the writer intends the insertion operation. In addition, the inserted character string is determined as the insertion position, and the phrase position closest to the written character string in the phrase delimiter position of the already-input sentence is determined as the insertion position and inserted. Therefore, when the user wants to insert a character string into a previously written sentence, the desired sentence can be obtained simply by writing and inputting the character string to be inserted, without performing extra work such as button operation. It is possible to freely write sentences continuously without disturbing the thoughts during document creation.

[0010]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the online character recognition device according to the present embodiment. As shown in FIG.
The online character recognition device includes a display-integrated coordinate input device 1, a character input unit 2, a mode determination unit 3, a sentence coordinate storage unit 4,
Character extraction unit 5, character recognition unit 6, reference pattern storage unit 7, phrase extraction unit 8, phrase extraction Japanese dictionary 9,
It is composed of a character string insertion unit 10 and a display control unit 11.

The display-integrated coordinate input device 1 has both an input function 1a for writing a character string as a series of coordinate values and an output function 1b for displaying a writing / recognition result.
In the display-integrated coordinate input device 1, a coordinate input surface and a display surface are overlapped and integrated, and display and coordinate input can be performed on the same surface. The character input unit 2 controls writing input and inputs a character string as a series of coordinate values (stroke sequence).

The mode determination unit 3 compares the coordinates input by handwriting with the already input sentence coordinates stored in the sentence coordinate storage unit 4 described later, and based on the comparison result, performs a normal input operation or an insertion operation. Is determined. The mode determination unit 3 sets a mode flag (mode flag) (not shown) to an on state “1” at the time of an insertion operation and clears (sets it to an off state “0”) at an input operation.

The sentence coordinate storage unit 4 stores coordinate values indicating the positions of character strings input by handwriting from the display-integrated coordinate input device 1. The character cutout unit 5 cuts out a stroke string for each character from a stroke string representing the character string input by the character input unit 2.

The character recognizing unit 6 performs character recognition by comparing the stroke sequence of one character cut out by the character cutting unit 5 with the reference pattern stored in the reference pattern storage unit 7.

The reference pattern storage unit 7 stores a reference pattern relating to a character to be recognized for character recognition. The phrase extraction unit 8 refers to the Japanese dictionary stored in the phrase extraction Japanese dictionary 9 from the character string composed of the characters obtained by the character recognition unit 6 and extracts the phrase based on the connection relation between words. Detect cuts.

The phrase extraction Japanese dictionary 9 is referred to by the phrase extraction unit 8 and registers words (character strings) for each part of speech. The character string insertion unit 10 inserts the input character string at a predetermined position in the input sentence. The display control unit 11 displays a handwriting and a character recognition result (character string). Next, the operation of the present embodiment will be described. In this embodiment, a case in which a horizontally written sentence is input by handwriting from the display-integrated coordinate input device 1 will be described as an example.

The on-line character recognition apparatus of the present embodiment has an X-coordinate indicating a character string range for an input character string, that is, for all character strings displayed by the output function 1b of the display-integrated coordinate input device 1. The minimum value Xmin, the maximum value Xmax (character string length) of the X coordinate, and the reference value Yno of the Y coordinate
(Character string position) is stored in the sentence coordinate storage unit 4. If nothing is displayed on the coordinate input device 1, the sentence coordinate storage unit 4
There is no coordinate data inside.

It is assumed that the character input unit 2 operates in parallel with the other parts so that the coordinate data input by handwriting can be input. First, as shown in FIG. 2, it is assumed that a sentence 21 is written on a display input surface 20 of the display-integrated tablet 10 in a place where nothing is written. The input function 1a on the display-integrated coordinate input device 1 sequentially inputs stroke strings (coordinate data strings) constituting characters. Here, it is assumed that the sentence "I am inputting the first sentence now" is written. The output function 1b displays handwriting as shown in FIG. 2 according to the input coordinate data.

The mode determining unit 3 determines an operation mode (input operation, insertion operation) based on the input character string. The processing operation of the mode determination unit 3 is shown in the flowchart of FIG.

First, as shown in FIG. 4, the mode determination section 3 sets a circumscribed rectangle for a coordinate data sequence representing a character string input by handwriting (step A1). Then, a maximum value X'max and a minimum value X'min of the X coordinate indicating the range of the circumscribed rectangle, and a maximum value Y'max and a minimum value Y'min of the Y coordinate are obtained (step A2). Further, the reference value Y ′ of the Y coordinate
No, for example, (maximum value of Y coordinate + minimum value of Y coordinate) /
2 (step A3). The range of the character string input by handwriting is the maximum value X'max and the minimum value X'min of the X coordinate.
, And a reference value Y′no of the Y coordinate. Hereinafter, the coordinate data indicating the character string range is referred to as sentence coordinate data.

Here, the mode determination unit 3 refers to the sentence coordinate data stored in the sentence coordinate storage unit 4 (step A).
4). Here, since the sentence coordinate data does not exist in the sentence coordinate storage unit 4 because it is a character string written first, Xmax (i), Xmin (i), and Yno (i) are stored in the sentence coordinate storage unit 4. I do.
i is a counter indicating the storage position of the sentence coordinate data in the sentence coordinate storage unit 4. The initial value is 1, and is incremented every time the sentence coordinate data is stored. Mode determination unit 3
Clears the mode flag (sets it to "0"), indicating that it is not an insertion operation (step A7).

Next, the input character string is separated into a stroke string for each character in the character extracting section 5. The character recognition unit 6 performs character recognition for each character based on the reference pattern stored in the reference pattern storage unit 7. This method uses, for example, the technology described in “Online Character Recognition of Borderless Handwritten Character Strings by Candidate Character Lattice Method” (Transactions of the Institute of Electronics, Information and Communication Engineers, Vol.J68-D No.4, p765 / 772,1985). Can be.

The character recognition result obtained by the character recognition unit 6 is sent to the phrase extraction unit 8. Phrase extraction section 8
, Pn, Pn,..., Pn are detected from the character string obtained by character recognition based on the connection relation of words, etc., while referring to the Japanese dictionary 9 for phrase extraction. It is stored in the storage unit 4.

The phrase extraction processing in the phrase extraction unit 8 includes, for example, "Japanese analysis in a sentence proofreading support function" (Information Processing Society of Japan, Natural Language Processing Vol.88-N).
L-69 No. 2, p1 / 8, 1988).

The display control unit 11 replaces the character pattern input by handwriting with a character obtained by the character recognition process and converts the character into a font (for example, 1
(2 points Mincho). FIG. 5 shows a display screen displayed by the output function of the display-integrated coordinate input device 1 when the input of the first sentence is completed.

Next, a case where a character string is inserted into an input sentence will be described. First, it is assumed that the writer wrote a character string to be inserted in the vicinity of a sentence as shown in FIG. The example shown in FIG. 6 is a case where a character string of “handwriting” is to be inserted between “the first sentence now” and “I am inputting”.

The mode determination unit 3 sets a circumscribed rectangle of the written character string as in the case of the character string initially input (FIG. 3, step A1), and inputs the character string "handwritten".
The sentence coordinate data X'max, X'min, and Y'no are calculated (steps A2 and A3).

Here, assuming that the number of sentence coordinate data stored in the sentence coordinate storage unit 4 is n,
For all Yno (i) in the range of 1 ≦ i ≦ n, Yno (k) satisfying | Yno (i) −Y′no | ≦ α is obtained.

Here, α is a constant for determining whether or not the character string input by handwriting is a character string to be inserted. For example, α = 20. If the sentence Yno (k) exists (step A5), the mode determination unit 3 determines that the writer has input a character string with an intention of insertion, and sets the mode flag to "1" (step A8). ). The corresponding sentence is 2
If there is more than a character string, the character string with the shortest distance (the value of | Yno (i) -Y'no |) is selected as the character string Yno (k) to be inserted.

In the case shown in FIG. 7, as a result of determining the Y coordinate reference value Yno (1) of the first sentence and the Y coordinate reference value Yno (2) of the second sentence, Yno (1) close to Y'no The character string (k = 1) is determined as the character string to be inserted. As described above, the character string “handwritten” to be inserted, which is written and input,
The character extracting unit 5 and the character recognizing unit 6 perform character recognition for each character.

After the character recognition processing, when the mode determination flag is set to "1" (step B1), the character string insertion unit 10 performs the insertion processing according to the flowchart shown in FIG.

As shown in FIG. 9, the character string insertion unit 10
X coordinate Xins of the first character “hand” 62 of the inserted character string “handwritten” 61 and Y of the character string selected by the mode determination unit 3
The reference value Yno (k) is obtained (steps B2 and B3).

Here, the character string insertion unit 10 determines the character string of the Y reference value Yno (k) selected by the mode determination unit 3
The phrase cutout position stored in the sentence coordinate storage unit 4 is read (step B4). Here, as shown in FIG. 9, it is assumed that there are four segment break positions P0, P1, P2, and P3.

The character string insertion unit 10 determines a segment break position P
Among the 0, P1, P2, and P3, the phrase position at which the distance from the X coordinate Xins of the first character "hand" of the insertion character string "handwriting" is minimum is determined as the insertion position (ins pos) (step B).
5). In the case shown in FIG. 9, since the phrase position P3 is closest to the X coordinate Xins, it is determined as the insertion position.

The character string insertion unit 10 is provided at the insertion position (ins pos).
Insert the character string "handwriting" into P3 (step B
6). Furthermore, the insertion of the character string changes the coordinate data of the sentence, so that Xmax (k), Xmin (k), Yno (k) (here K =
1) Update the phrase segmentation position (step B)
7), reset of the mode determination flag (mode flat)
g = "0") (step B8).

The result obtained as described above is output to the display controller 11.
And is displayed by the output function 1b of the display-integrated coordinate input device 1. As shown in FIG. 10, when character string insertion processing 10 is performed, the previously displayed character string is deleted, and a new character string with the character string inserted is displayed.

The present invention is not limited to the embodiment described above. For example, a writing underline may be drawn on the handwriting input surface of the display-integrated coordinate input device 1 so that the writer can easily write a character string. Further, the character cutout processing, the character recognition processing, and the phrase cutout processing may use other methods than those described above.

In the above embodiment, the phrase cutout position is stored in the sentence coordinate storage unit 4. However, the phrase cutout position may be calculated each time a sentence to be inserted is selected without storing the phrase cutout position. .

[0039]

As described above, according to the present invention, when a writer wants to perform an insertion operation while writing a sentence, he or she can perform an operation such as a button operation which is completely unrelated to writing. Just by writing a character string near the input text, the character string is automatically inserted into the text. As a result, the burden on the writer during the document input is reduced, and a great effect in practical use is obtained, which enables efficient text input.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an online character recognition device according to one embodiment of the present invention.

FIG. 2 is a diagram showing an example of a screen on which handwriting of a character string input by handwriting is displayed.

FIG. 3 is a flowchart for explaining a processing operation of a mode determination unit 3;

FIG. 4 is a view for explaining detection of sentence coordinate data.

FIG. 5 is a diagram showing an example of a screen on which a character string after character recognition is displayed.

FIG. 6 is a diagram illustrating an example of a screen when a character string to be inserted is written near a sentence that has been input;

FIG. 7 is a view for explaining the relationship between the position of a sentence and the position of a character string to be inserted;

FIG. 8 is a flowchart for explaining the processing operation of the character string insertion unit 10;

FIG. 9 is a view for explaining processing for specifying the insertion position of a character string.

FIG. 10 is a diagram showing an example of a screen on which a new character string in which a character string has been inserted is displayed.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Display integrated type coordinate input device, 2 ... Character input part, 3 ... Mode determination part, 4 ... Sentence coordinate storage part, 5 ... Character cutout part,
6: character recognition unit, 7: reference pattern for character recognition, 8: phrase extraction unit, 9: Japanese dictionary for phrase extraction, 10 ...
Character string insertion unit, 11... Display control unit.

Continuation of the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G06K 9/62 G06F 3/03 380 G06F 17/22 506 G06F 17/24 554 JICST file (JOIS)

Claims (1)

(57) [Claims] 1. An online character recognition device for recognizing a character string input by handwriting from a coordinate input device, comprising: sentence coordinates for storing sentence coordinate data indicating a position of the character string input by handwriting from the coordinate input device. Storage means, based on the sentence coordinate data of the input character string stored in the sentence coordinate storage means, and the position of the character string handwritten from the coordinate input device, Mode determination means for determining whether or not the insertion mode should be inserted into a character string, and phrase extraction means for extracting a character recognition result of a character string input by writing in units of a phrase and detecting a phrase position in the character string When the mode determination unit determines that the insertion mode is set, the phrase position of the input character string obtained by the phrase cutout unit is input. The predetermined phrase positions determined based on the position of the character string, line character recognition apparatus characterized by comprising a character string insertion means for inserting a character string in the input.