JPH0636072A - Handwritten character input device - Google Patents

Abstract

PURPOSE:To lighten the burden on a user by easily and surely narrowing down the character kinds of recognition candidates. CONSTITUTION:The handwritten character input device is equipped with an input part 3a where coordinate values corresponding to a position optionally indicated on a coordinate input surface are inputted, a character recognition processing part 3b which inputs the coordinate sequence inputted at the input part 3a as character input stroke data, performs character recognition as to all character kinds prepared in a dictionary for recognition, and obtains a constant number of candidate characters for each of the character kinds, a candidate character kind limiting part 3c which narrows down the candidate characters obtained by the character recognition of the character recognizing process part 3b according to the limitations of the character kinds indicated before or after the character recognition is performed by the character recognizing process part 3b, and a display control part 3e which displays characters according to the candidate characters narrowed down by the candidate character kind limiting part 3c.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a handwritten character input device for recognizing characters handwritten on a position input device.

[0002]

2. Description of the Related Art In recent years, handwritten character input devices for recognizing and inputting characters written on a position input device for inputting coordinate values by online handwritten character recognition have been widely used. In this type of handwritten character input device, in the handwritten character recognition, the type of character to be recognized and the recognition candidates are narrowed down for efficient character recognition.

The following methods are used for narrowing down the recognition target character type and recognition candidates. The first method is
The target character type is limited in advance by internally limiting the recognition target character type before performing handwritten character recognition, or by allowing the user to select the character type before performing character input. In the first method, since the target character type is limited in advance, the candidate character obtained by handwritten character recognition does not include a character of an incorrect character type, and efficient recognition processing can be performed.

The second method is to perform character recognition with all character types as recognition target character types, and after the character recognition, let the user select a character type as a recognition candidate. The second method can deal with the case where the character type to be input cannot be specified in advance.

[0005]

As described above, the above-described first and second methods are used for the method of narrowing down the recognized character types or recognition candidates in the conventional handwriting input device. However, the first method is effective in that it does not include a character type that is not a recognition target, but if there is an error in the selection of the recognition target character type especially when the user selects it, re-input is required. did not become. In this case, the operation becomes complicated and the efficiency of character input is reduced. Therefore, it does not give a good impression to the user and cannot be said to be a good human interface.

Further, in the second method, since all the character types are targeted for recognition, the number of character type candidates obtained by character recognition is reduced. That is, in the character recognition process, since a predetermined number of recognition candidates that are expected to include correct characters are obtained in descending order of recognition rate, it is not always possible to obtain a desired character type candidate among the recognition candidates. Not exclusively. Therefore, it may be different from the recognition candidate when the target character type is limited, and the correct character may not be obtained.

The present invention has been made in view of the above points, and provides a handwritten character input device capable of reducing the burden on the user by easily and surely narrowing down the character types of recognition candidates. The purpose is to do.

[0008]

SUMMARY OF THE INVENTION According to the present invention, input means for inputting coordinate values corresponding to a position arbitrarily designated on a coordinate input surface, and a coordinate value string input by the input means are used as character input strokes. Character recognition processing means for capturing as data, character recognition for all character types prepared in the recognition dictionary, and obtaining a fixed number of candidate characters for each character type, and character recognition by the character recognition processing means Depending on the limitation of the character type instructed before or after execution, the candidate character type limiting means for narrowing down the candidate characters obtained by the character recognition by the character recognition processing means, and the candidate character narrowed down by the candidate character type limiting means Accordingly, the display control means for displaying characters is provided. Further, the coordinate value detected by the position input means in which the coordinate input surface and the display screen are integrated is input by the input means.

[0009]

With this configuration, the character types to be recognized when performing handwritten character recognition are not limited, and recognition candidates are individually obtained for all character types, so that the user can select character types before or after recognition. In other words, even if the recognition candidates are limited, the recognition candidates can be limited without reducing the number of recognition candidates for each character type itself.

Therefore, by adopting a method of allowing the user to select the character type to be recognized, even if the user makes a mistake in selecting the recognized character type, there is no need for the user to re-input the character type. You can make a choice. The recognition candidate characters obtained by this method can be effectively used as data for post-recognition processing for improving other recognition rates.

Thus, by reducing the operation of re-inputting by the user, handwriting input can be made smooth, and characters can be input at intervals of writing on paper, and the recognition candidate is used for post-processing. By providing the user with processing that cannot be done,
The human interface can be improved considerably.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 2 is a block diagram showing the configuration of the handwriting input device according to the embodiment. As shown in FIG.
The handwriting input device includes a position input device 1, a processing control device 2,
The display device 3 and the external storage device 4 are included. The position input device 1 inputs a position input coordinate value which is handwritten using a stylus pen 7i described later.

The processing control device 2 controls the entire handwriting input device, and loads various programs and data stored in the external storage device 4 to execute various processes. For example, processing such as recognition (online handwritten character recognition) of a character written and input from a predetermined area (handwritten character input area) provided on the input surface of the position input device 1 having a transparent tablet is performed. In online handwritten character recognition, recognition candidates are obtained by individually recognizing all the character types for each recognition target character type (the target range of the recognition dictionary is each character type), and the acquired recognition candidates are selected. The detailed configuration of the processing control device 2 will be described later.

The display device 3 is composed of a liquid crystal display or the like, and is obtained by a coordinate value sequence (character handwriting data) input from the position input device 1 or a process (character recognition, candidate selection, etc.) in the process control device 2. Displayed characters are displayed.

The external storage device 4 stores a dictionary used for recognizing characters input from the position input device 1 in the processing control device 2, a system program for starting this device, an application program, and the like.

The position input device 1 is provided with, for example, a transparent tablet (details will be described later) for the coordinate input device. The display device 3 is composed of, for example, a liquid crystal display. In this embodiment, as shown in FIG. 3, the transparent tablet of the position input device 1 and the display screen of the display device 3 have the same size and the same coordinate plane, and the position input device 1 is superimposed on the display device 3. And a processing control device (apparatus main body).

Next, a detailed configuration of the processing control device 2 will be described. As shown in FIG. 1, the processing control device 2 includes an input unit 3a, a character recognition processing unit 3b, and a candidate character type limiting unit 3
c, an area discrimination processing section 3d, a display control section 3e, a position data buffer 3f, an area data buffer 3i, an input frame data buffer 3j, a recognition candidate data buffer 3k, and an output data buffer 3l.

The input unit 3a inputs the coordinate data and the like input from the position input device 1. Character recognition processing unit 3b
Performs character recognition processing on the coordinate data series (character writing pattern) input by the input unit 3a with reference to the recognition dictionary.

The candidate character type limiting unit 3c is obtained by the character recognition processing by the character recognition processing unit 3b according to the character type designated by using the recognized character type setting button provided on the display screen (coordinate input surface). Narrow down the recognition candidates.

The area discrimination processing section 3d discriminates which area (described later) of the handwritten character input area or the like provided on the input surface the coordinate data from the position input device 1 is included in. The display control unit 3e controls the display of the screen displayed on the display device 3.

The position data buffer 3f is for storing the coordinate data input via the input unit 3a. The recognition data buffer 3g includes a character recognition processing unit 3b.
Is for storing a coordinate data series which is a target of character recognition, and predetermined data is transferred from the position data buffer 3f. The pen data buffer 3h is for storing data indicating the state (on, off) of a pen switch provided in the stylus pen 7i used for writing input. The area data buffer 3i is for storing data (area position, processing attribute, data indicating a recognized character type, etc.) regarding various areas (corresponding to the display screen) provided on the coordinate input surface. The input frame data buffer 3j is for storing data (the number of input frames, frame positions, etc.) regarding a handwritten character input area in which character recognition processing is performed. The recognition candidate data buffer 3k is
This is for storing data indicating character recognition candidates obtained by the character recognition processing in the character recognition processing unit 3b. The output data buffer 3l is for storing the data and the like output from the processing control device 2 and displayed on the display device 3.

The area data buffer 3i described above is shown in FIG.
As shown in, the start point x coordinate data 4a, the start point y coordinate data 4b, the end point x coordinate data 4c, the end point y coordinate data 4
1 for the area for storing d and the processing attribute data 4e
It is configured to include data for an area and the processing area for which the data is set. The processing attribute data includes data designating whether or not to perform character recognition processing on the coordinate data input in the area.

The input frame data buffer 3j is shown in FIG.
As shown in FIG. 5, frame number data 5a, frame start point x coordinate data 5b, frame start point y coordinate data 5c, frame end point x coordinate data 5
d, the area for storing the frame end point y coordinate data 5e is 1
These pieces of input frame data are provided as the number of frames as the frame data for each piece.

As shown in FIG. 6, the recognition candidate data buffer 3k has an input character upper left x coordinate data 6a, an input character upper left y coordinate data 6b, and an input character size x data 6.
c, input character size y data 6d, recognition target character type data 6e, and candidate character data for all character types (kanji, hiragana, katakana, numbers, uppercase letters, lowercase letters, symbols in the example) to be recognized (the same). In the embodiment, areas of the first to fourth candidate character data 6f to 6i) are provided for each character type, and these areas are provided as recognition candidate data for one frame for the number of frames.

Next, the transparent tablet provided in the position input device 1 will be described. The position input device 1 is realized, for example, as an electrostatic transparent tablet as shown in FIGS.

In the electrostatic transparent tablet, as shown in FIG. 7A, a pair of transparent glass plates 7a and 7b are joined together via an insulating protection layer (not shown). Transparent glass plate 7
A plurality of transparent electrodes 7c and 7d are arranged in parallel on a and 7b at a predetermined interval, respectively.
The a and 7b are joined so that the transparent electrodes 7c and 7d are orthogonal to each other. The transparent electrodes 7c and 7d are shown in FIG.
As shown in (b), each shift controller 7e
Shift registers 7f and 7g to operate under the control of
Connected to.

The specific operation is as follows. First, when measuring coordinates in the horizontal direction, a predetermined voltage is sequentially applied to the transparent electrode 7c connected to the shift register 7f from the left end (or the right end). When the stylus pen 7i approaches the electrode to which the voltage is applied, the transparent electrode 7c
The electrostatic capacitance formed between the stylus pen 7i and the stylus pen 7i induces an electrostatic induction voltage in the stylus pen 7i. This electrostatic induction voltage is amplified by the amplifier 7j, and the high frequency waveform is removed by the filter 7k.
It is given to the counter 7h as a stable pointing position detection signal. The counter 7h starts counting when a predetermined voltage is applied to the electrode located closest to the origin (the side where the voltage application is started) of the transparent electrodes 7c, and indicates a designated position detection signal equal to or more than a predetermined value. The count is ended when is detected. By performing such an operation, the coordinate value designated by the stylus pen 7i is converted from time and detected.

When the horizontal coordinate measurement is completed, the vertical coordinate measurement is started. In the coordinate measurement in the vertical direction, similarly to the coordinate measurement in the horizontal direction, a predetermined voltage is sequentially applied from the upper end (or the lower end) of the transparent electrode 7d connected to the shift register 7g, and the electrostatic induction detected by the stylus pen 7i. The pointing position detection signal is generated from the voltage by the amplifier 7j and the filter 7k, and the coordinate value is detected by the counter 7h.

When a character or other pattern is written on the transparent tablet using the stylus pen 7i, it is input as a connection (position coordinate data series) of instructed position coordinate values. The stylus pen 7i in the present embodiment is provided with a pen switch, and is in a switch state (ON state) if it is within a certain distance on the coordinate detection surface.
Is detected.

Thus, based on the coordinate data series indicating characters or other patterns input from the position input device 1, the processing control device 2 sets the same coordinates as the position designated on the transparent tablet of the display device 3. Is driven for display. By this display control, the pattern input by handwriting on the transparent tablet surface is simultaneously displayed as the handwriting input trajectory, and characters or other patterns are input as if writing on the paper surface using a pen. .

Here, the processing procedure of the handwritten character input processing, the character recognition processing, and the recognized character type selection processing will be described using an example of an application including handwritten character input shown in FIG.

When an application including handwritten character input is started on this apparatus, the display control section 3e causes the handwritten character input area (a plurality of handwritten input frames) as shown in FIG.
An initial screen including the above is displayed on the screen of the display device 3 (step A1).

First, on the initial screen display of this device, the user
That is, when the stylus pen 7i is brought closer to the transparent tablet surface of the position input device 1 for writing, the position input device 1 detects the coordinate value of the point pointed by the stylus pen 7i and the switch state.

In the case of the present embodiment, the detected pen status is stored in the pen data buffer 3h as switch status data (on or off) acquired as a result of interrupting the driver of the tablet every certain time. It refers to what was done.

In this device, the switch state is changed from the off state to the on state by pressing the pen tip of the stylus pen 7i on the transparent tablet surface. Coordinates can be detected. As a result, at the time when the coordinate value designated by the sillus pen 7i is detected (step A
2), the area discrimination processing unit 3d discriminates, on the basis of the data in the area data buffer 3i, the position on the display screen to be instructed before the user's input (step A3).
When the processing attribute data 4e of the determined area is data for performing handwritten character recognition processing (step A
4) Set initialization processing for recognizing handwritten characters (step A5). In the present embodiment, the character recognition initialization processing includes initialization of the character recognition driver, setting of display attributes of handwriting at the time of character input, processing of handwriting data after recognition, or
A process for expanding the input box data of the target character input area according to the data stored in the input box data buffer 3j is performed. When the initialization process of handwritten character recognition is performed, the handwritten character recognition processing unit 3b is activated and the character recognition process is performed. If there is valid data, that is, a recognition result from the character recognition processing (step A6), the recognition result is displayed.

If the detected coordinate value is outside the character recognition area, and if there is data that is being input to the character input area, the character is forcibly cut off and recognized, and the initial character recognition is performed. Resets the processing (steps A7, A
8). Then, the processing for each processing area is called (step A9). The processing for each processing area includes the above-described recognition result display, recognition candidate selection processing described later, and the like.

Here, the character recognition processing for recognizing characters by the above-described processing, the character display processing for displaying the characters acquired by the character recognition processing (recognition result display processing), and the recognition candidate selection processing by the recognition character type selection are performed. Details will be described. (1) Character recognition processing

The operation procedure of the character recognition process is shown in the flowchart of FIG. First, when the user inputs a character by handwriting in the input frame in the character input area, the pen switch is determined from the coordinate value of the point pointed by the stylus pen 7i detected by the position input device 1 and the pen state information of the switch state. It is determined whether it is in the on state (step B1).
As described above, the detection of the pen status information refers to the pen status data stored in the pen data buffer that is acquired as a result of interrupting the tablet driver at regular intervals. Is.

If the switch state is on and the first stroke data starting point to the input frame is set, the coordinate data is stored in the position data buffer 3f, but the following processing is performed. That is, at this time, if the character input data is already stored in the position data buffer 3f, it is regarded as the data of the character previously input to another input frame, and the input of that character is terminated ( Frame inspection processing). Then, the stored coordinate data is transferred to the recognition data buffer 3g, the contents of the position data buffer 3f are cleared, and the coordinate data just input is stored (steps B2 and B4).

If the switch state is ON but the first stroke start point to the input frame is not the stroke data input after the start point data of the first stroke data, the coordinate data is stored in the position data buffer 3f. ,
The process returns to the beginning (step B3).

If the switch state of the stylus pen 7i is off and the switch state of the previous data is on, it is considered that the stroke data has been input, and the stroke delimiter data is stored in the position data buffer 3f. Then, the process returns to the beginning (steps B1, B5, B
6).

When the switch state of the stylus pen 7i is off and the switch state of the previous data is also off,
The gap between input strokes. If this state is over a certain period of time, it is considered that one character has been input (time cutout process). Then, the data stored in the position data buffer 3f is transferred to a recognition data buffer (not shown), and the position data buffer 3f is cleared (step B).
1, B5, B7, B8).

The input stroke data is transferred to the output data buffer 3l at the same time as it is stored in the position data buffer 3f and is sequentially displayed on the display device 3 as input handwriting pattern information in accordance with the handwriting input.

In this way, the character recognition processing unit 3b analyzes the coordinate data series with respect to the stroke data stored in the recognition data buffer 3g after being frame-censored or time-censored by storing the input stroke data. (Step B9), character recognition processing is performed (steps B10 to B13).

In this character recognition processing, the stored coordinate data series is divided into writing strokes by stroke delimiter data, and a direction string showing the characteristic is obtained for each writing stroke. For example, the pattern shown in FIG.
It is given by being quantized in eight directions (direction values) as shown in 1 (b). Then, the information obtained by the direction value conversion of the writing stroke and the position code conversion processing is used for the recognition target range of the character recognition dictionary in units of character types (Kanji, Hiragana, Katakana, numbers, uppercase letters, lowercase letters, symbols, etc.). It is narrowed down and referred to for recognition. The character recognition dictionary is, for example, as shown in FIG. 11C, classified according to the number of strokes forming a character to be recognized, associated with each character code, and configured for each character book.

The character code and recognition value of each recognition candidate of the character thus recognized, the upper left coordinate value of the character at the time of input, the character size, etc. are stored in the recognition candidate data buffer 3k (step B14). The process returns to the main routine (step B15). (2) Character Display Processing (Recognition Result Display Processing) The operation procedure of the character display processing is shown in the flowchart of FIG.

When the handwritten character recognition processing described above recognizes a character handwritten in the input frame by the user,
Returning to the main routine, the process corresponding to each area is called, but when character recognition is performed, the candidate character obtained as a result of recognition is displayed.

In this process, first, the position of the input character is determined (step C1). The display position of the recognition result is determined in which input frame based on the upper left coordinate data of the input character stored when the recognition candidate was stored in the recognition candidate data buffer 3k in the character recognition process and the character size value. To determine.

Then, the first candidate of the recognition candidates is displayed in the determined input frame, and the next candidate character is displayed in the next candidate display area. However, if any of the recognition target character type setting flags is turned on, that is, if the recognition target character type is selected before the handwritten character recognition processing is performed (step C2), within the selected character type according to the character type. The first candidate is displayed in the input frame, and the subsequent candidates are displayed in the next candidate display area (step C4). The selection of the recognition target character type will be described in the next candidate selection process.

If all the flags for setting the character type to be recognized are off, all the flags are turned on to display the first candidate character and the next candidate character (steps C3 and C4). In the embodiment, the candidate having the best recognition value among all the candidates is displayed as the first candidate in the corresponding input frame, and the candidate character including the first candidate is displayed in the next candidate display area for each character type. . After the recognized result is displayed as described above, the process returns to the main routine (step C5). (3) Next candidate selection process FIG. 1 is a flowchart showing the processing procedure of the next candidate selection process.
3 shows.

When the next candidate is selected from the candidates recognized and displayed by the handwritten character recognition process and the character display process described above, the candidate character type limiting unit 3c narrows down the recognition candidate characters (character types) and re-selects the candidates. Display etc.

First, the switch state of the stylus pen 7i is determined (step D1). If the switch state of the stylus pen 7i is changed from off to on, it is determined that the stylus pen 7i is used to instruct a button on the display screen (an input area on the tablet surface corresponding to the display position indicating the button). , The processing is sorted according to which button is pressed. The area, item, button, etc. are discriminated by the area discrimination processing of the main routine. The next candidate selection process can be divided into a recognized character type setting operation, a next candidate button selection operation, and a next candidate selection operation from the user's operation.

A button for setting the recognized character type in the display screen shown in FIG. 9 by the user ([A] indicates kanji. Similarly, [A] Hiragana, [A] Katakana, [1] number,
When an arbitrary character type is selected from [A] uppercase letters, [a] lowercase letters, and [◆] symbols), the main routine discriminates the recognized character type setting item, and this process is called. As a result, if the character type flag corresponding to the designated recognized character type setting button is off, it is turned on, and if it is turned on, it is turned off and the selection flag is set / reset (step D).
1-D5).

Then, the first candidate character and the next candidate are displayed from the character type flag and the candidate data stored in the recognition candidate data buffer 3k (step D6). Here, the candidate character with the best recognition value among the character types with the character type flag turned on is displayed as the first candidate in the corresponding input frame, and the candidate character corresponding to the selected character type is displayed in the next candidate display area for each character type. To do.

The corresponding input frame is set according to the data of the corresponding position data buffer 3f secured inside the process, and is changed by an instruction from the next candidate display button. Normally, if there is no instruction from the next candidate display button, the input frame located on the rightmost side of the already input frame is set. These candidates are displayed and the process returns to the main routine (step D11).

If the user selects the next candidate display button (provided under each input frame in the embodiment) (step D7), the corresponding data is stored in the recognition candidate data buffer 3k. According to the character type to be recognized, which is set as the candidate data of the input frame,
The candidates and the next candidates are displayed (step D8). Then, the process returns to the main routine (step D11).

In addition to this, when the user selects a candidate character from the next candidate display area (step D9), the candidate character corresponding to the designated coordinate value is searched and the selected candidate character is displayed in the corresponding input frame. (Step D10). And
The process returns to the main routine (step D11).

The above processing will be described using a specific example. As shown in FIG. 14, when a character is written and written in the input frame using the stylus pen 7i, it is assumed that the candidate characters as shown in FIG. 15 are obtained by the handwritten character recognition process. If the recognition target character type is not set before the character input, the all-character-type flag is turned on in the character display process, and the first candidate is displayed in the character-input input frame as shown in FIG. The next candidate is displayed in the next candidate display area.

After that, as shown in FIG. 17, when the recognition target character type is narrowed down by an instruction from the recognition target character type setting button (in FIG. 17, the character type is designated as katakana [A]), the next candidate display area is displayed. Only the Katakana candidates are displayed in accordance with the Katakana candidate character data for the corresponding input frame of the recognition candidate data buffer 3k.

Further, as shown in FIG. 18, when the target character type and the corresponding input frame are designated, the first candidate is displayed in the designated input frame and designated in the next candidate display area. Next candidates of the selected character type (Kanji [A] and Katakana [A] in FIG. 18) are displayed. In addition, as shown in FIG. 19, characters (instruction candidates) displayed in the next candidate display area
When is specified, it is displayed in the corresponding input box. As described above, the processing of the present invention is performed based on the processing described above.

In this way, when inputting characters by handwritten character recognition, the selection of the recognition candidate character type for handwritten character recognition can be arbitrarily set either before or after the handwritten character recognition. Then, the handwritten character recognition is performed so as to obtain a fixed number of recognition candidates for each character type regardless of the set recognition candidate character type, and the display shows candidates according to the preset recognition candidate character type. Characters can be displayed, and candidates for all character types can be internally stored and held in the recognition candidate data buffer 3k.

By this handwritten character recognition processing, even if the user selects or changes the recognition candidate character type after character recognition, the candidate character of the set character type is acquired without inputting the character again in the handwritten character input frame. can do.

Further, since a fixed number of recognition candidates are acquired for each character type for all character types, the number of candidates for each character type does not decrease due to narrowing down of the recognition candidate character types, which may adversely affect the recognition rate. Absent. That is, the probability that a character to be input is included in the next candidate increases, the recognition rate can be improved, and the handwritten character input device is good in terms of human interface.

The present invention is not limited to the above embodiment. For example, the coordinate detection method of the position input device 1 is not limited to the electrostatic capacitance method, and other methods such as an electromagnetic induction method and a pressure-sensitive method may be used, and the stylus pen 7i may have a cord according to these methods. It may be cordless instead. Also, the tablet may be an opaque tablet. The display device 3 may be configured by a plasma display or the like in addition to the liquid crystal display.

With respect to the character recognition processing, in the above-described embodiment, the stroke of the first stroke of the stroke data, which becomes the character data, is input to the input box provided with the handwritten character input area, or the stroke is moved outside the character input area. When the start point data is input, the stroke data (coordinate value string) previously input in the input frame is regarded as the input data for one character, and character recognition is performed, but it is recognized by the input area. The driving method may not be used, and a recognition driving area or a button may be provided so that the recognition driving is performed when the user gives an instruction. Further, in the embodiment, the configuration in which four candidates of characters are acquired for each character type has been described, but the number of candidates or the configuration of candidate data is not limited to the embodiment.

Although the recognition result display processing and the next candidate selection processing have been described in the embodiment using the screen format shown in FIG. 13, the present invention is not limited to this format. Although the number is set individually, other numbers may be set and the frames may not be adjacent to each other. Regarding the display method of handwriting data or recognition candidates, it is not the method of deleting the handwriting data after recognition and displaying the first candidate, but the method of displaying the first candidate in a position other than the input frame and not deleting the handwriting data. May be. Furthermore,
The method of selecting the next candidate character by tiling and displaying the next candidate character by performing a button instruction may be used instead of the method of listing the next candidate and instructing from the list. In short, the present invention can be variously modified without departing from the gist thereof.

[0067]

As described above, according to the present invention, the recognition target character type when performing handwritten character recognition is not limited, and all the character types are individually recognized, so that before or after recognition. The user can select the character type, that is, limit the recognition candidates. Moreover, even if the recognition candidate character types are limited by individually recognizing all the recognition target character types during handwritten character recognition. Since the number of candidates for the character type itself is not reduced, the character types of the recognition candidates can be easily and reliably narrowed down, and the burden on the user can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a processing control device 2 shown in FIG.

FIG. 2 is a block diagram showing a configuration of a handwritten character input device according to an embodiment of the present invention.

FIG. 3 is a diagram for explaining structural features of a handwriting input device.

FIG. 4 is a diagram for explaining a configuration of a region data buffer 3i provided in the processing control device 2.

FIG. 5 is a diagram for explaining the configuration of an input frame data buffer 3j provided in the processing control device 2.

FIG. 6 is a diagram for explaining the configuration of a recognition candidate data buffer 3k provided in the processing control device 2.

FIG. 7 is a diagram showing a configuration of a transparent tablet provided with the position input device 1.

FIG. 8 is a flowchart showing a procedure for inputting handwritten characters.

FIG. 9 is a diagram showing an example of an initial screen including a handwritten character input area.

FIG. 10 is a flowchart showing a procedure of character recognition processing.

FIG. 11 is a diagram showing a direction example representing a stroke direction at the time of character recognition and a configuration example of a recognition dictionary.

FIG. 12 is a flowchart showing a processing procedure of character display processing (recognition result display processing).

FIG. 13 is a flowchart showing a processing procedure of next candidate selection processing.

FIG. 14 is a diagram showing an example of a display screen when a character is written and input in the input frame.

FIG. 15 is a diagram showing recognition candidates for each character type obtained by character recognition.

FIG. 16 is a diagram showing an example of a display screen on which a first candidate and a next candidate by character recognition are displayed.

FIG. 17 is a diagram showing an example of a display screen when an instruction to narrow down recognition target character types is instructed.

FIG. 18 is a diagram showing an example of a display screen when a target input frame for designating a target character type and displaying a next candidate is displayed.

FIG. 19 is a diagram showing an example of a display screen when the next candidate character is selected.

[Explanation of symbols]

1 ... Position input device, 2 ... Processing control device, 3 ... Display device,
3a ... Input unit, 3b ... Character recognition processing unit, 3c ... Candidate character type limiting unit, 3d ... Region discrimination processing unit, 3e ... Display control unit,
3f ... Position data buffer, 3g ... Recognition data buffer, 3h ... Pen data buffer, 3i ... Area data buffer, 3j ... Input frame data buffer, 3k ... Recognition candidate data buffer, 3l ... Output data buffer, 4 ... External storage device 4a ... Start point x coordinate data area, 4b ... Start point y coordinate data area, 4c ... End point x coordinate data area, 4d ... End point y coordinate data area, 4e ... Processing attribute data area, 4f
... recognized character type data area, 5a ... frame number data area, 5
b ... frame start point x coordinate data area, 5c ... frame start point y coordinate data area, 5d ... frame end point x coordinate data area, 5e ... frame end point y coordinate data area, 6a ... input character upper left x coordinate data area, 6b ... input character Upper left y-coordinate data area, 6c ... input character size x data area, 6d ... input character size y data area, 6e ... recognition target character type data area, 6f ... first
Candidate character data area, 6g ... Second candidate character data area,
6h ... Third candidate character data area, 6i ... Fourth candidate character data area, 7a, 7b ... Transparent glass plate, 7c, 7d ... Transparent electrode, 7e ... Shift controller, 7f, 7g ... Shift register, 7h ... Counter, 7i ... a stylus pen,
7j ... amplifier, 7k ... filter.

Claims (2)

[Claims] 1. An input means for inputting coordinate values corresponding to a position arbitrarily designated on a coordinate input surface, and a coordinate value string input by said input means is fetched as character input stroke data and stored in a recognition dictionary. Character recognition is performed for all prepared character types, and a character recognition processing unit that acquires a fixed number of candidate characters for each character type, and a character type designated before or after execution of character recognition by the character recognition processing unit Depending on the limitation of the above, the character recognition processing means narrows down the candidate characters obtained by the character recognition, and the character display is performed according to the candidate character narrowed down by the candidate character type limiting means. A handwritten character input device comprising: display control means. 2. The handwritten character input device according to claim 1, wherein the input means inputs the coordinate value detected by the position input means in which the coordinate input surface and the display screen are integrated.