JPH10143315A - Data input device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide such a natural use feeling as to write on paper at the time of a data input. SOLUTION: A reference point and a desired format on a form P which is placed on a tablet 3 are set with point inputs by using a pen, the stroke data of a character which is written on the form P is detected at coordinate positions by the tablet 3 and is undergone coordinate correction based on the reference point by a CPU 10, character recognition is carried out based on the stroke data which is undergone coordinate correction by a character recognizing engine 19, the candidate characters of a recognition result are shown on a displaying part 5 and the selection of the candidates characters is performed by selecting a prescribed icon on the form P.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data input device for inputting data by writing using an input device such as a tablet.

[0002]

2. Description of the Related Art Conventionally, this type of data input device includes, for example, a pen personal computer (personal computer) using a tablet as a substitute for a keyboard and a PDA (Persian computer).
onal Didital Assistants) and LCD (Liquid Cr
This is a configuration in which a display unit and a tablet are integrally formed. The handwriting written on the tablet is displayed as an image on the LCD display unit.

In these data input devices, an LCD is used.
An entry frame is fixedly arranged on the display unit, and the user inputs data by writing in the entry frame. When data is input to this entry frame, recognition processing is started for the handwriting of the input data. For example, if it is a handwriting of a character, a candidate character is displayed on the LCD display unit as a recognition result. When one character is selected from the displayed candidate characters by pen input, the selected character is displayed as a print image at the cursor position.

[0004]

However, in the data input device according to the conventional example described above, since the LCD display unit and the tablet are integrated and a glass material is used for the surface portion thereof, it is difficult to write with a pen. In addition, the pen tip slipped on the glass material, and the natural feeling of writing on paper could not be obtained, and the unnatural feeling of operating a computer was felt.

Also, due to the structure, a writing surface on which a pen is written and L
Since there is a gap between the display screen of the CD display unit and the display screen of the user, parallax is generated depending on the viewing angle of the user. In this case, a natural feeling of writing on paper cannot be obtained.

Further, since the handwriting is displayed as an image on the LCD display unit, the roughness of the definition (dot pitch) of the LCD causes the handwriting display to feel unnatural and the display of small characters to be unclear. Was.

In addition, since the recognition result is displayed at the character input position on the LCD display unit, the LCD display unit is enlarged when the size of the display area is increased. In this case, the device becomes a large size that goes against the reduction in thickness and size, and the reduction in weight is limited.

An object of the present invention is to make it possible to input writing data with a natural feeling of use, such as writing on paper with a pen.

[0009]

According to the present invention, there is provided a field defining means for defining a field divided into a plurality of entry frames on a writing area, and a plurality of entry fields defined by the field definition means in each of the entry fields defined by the field definition means. Character recognition means for recognizing written handwritten data for each entry frame; display means for displaying a list of a plurality of recognition candidates recognized by the character recognition means in association with the entry frames; Selecting means for selecting a desired candidate from the recognition candidates listed on the display means for each entry frame.

That is, according to the present invention, the field defining means defines a field divided into a plurality of entry frames on the writing area, and the character recognizing means defines each of the fields defined by the field defining means. The handwriting data written in each entry frame is character-recognized for each entry frame, and the display means displays a list of a plurality of recognition candidates recognized by the character recognition means in association with each entry frame, The selection means selects a desired candidate from the recognition candidates listed on the display means for each entry frame.

Therefore, it is possible to input writing data with a natural feeling of use as if writing with a pen.

[0012]

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is an external perspective view showing a shape of a data input device according to an embodiment of the present invention, and FIG. 2 is a side sectional view of a main part of the data input device shown in FIG.

The data input device 1 shown in FIG. 1 has a rectangular base panel 2. A paper clip 4 is attached to the rear end 2a of the base panel 2. A clip restraining spring 8 is interposed between the rear end 2a and the paper clip 4, and the elasticity of the paper clip 4 in the S direction is used. Enables opening and closing operations. This paper clip 4
2A and 2B, the paper P set on the base panel 2 is clipped by the bias of the clip restraining spring 8, as shown in FIGS. A paper holding / clip cushioning material 6 is attached to the base panel 2 as a cushioning material when the paper P is clipped, and the cushioning force of the paper clip 4 is reduced by this cushioning. The sheet presser / clip cushioning material 6 holds the sheet P by urging the sheet clip 4 in the S1 direction.
Has a function of absorbing the force applied to the tip P1 of the first. The leading end P1 of the sheet P is positioned by a sheet position determining unit 7 forming the inner wall surface of the frame of the base panel 2 (FIG. 2).
(B)).

On the base panel 2, there is attached a tablet 3 as an underlay of the paper P. The tablet 3 detects a coordinate position from a handwriting written on the paper P, and transmits the data of the coordinate position to the CPU 10. Output to

The paper clip 4 has a display unit 5 such as an LCD and an EL, and displays a recognition result of a handwriting (character) described later and displays a candidate character.

FIG. 3 is a block diagram showing an example of the internal configuration of the data input device 1, and FIG. 4 is a diagram showing an example of the memory configuration of the candidate memory 15. The data input device 1 includes, for example, a format file unit 11, a data file unit 12, a ROM 13, a work memory 14, a candidate memory 15, a buzzer 16, a serial interface 17, a flash memory, in addition to the tablet 3 and the display unit 5 described above. 18, a character recognition engine 19, and a bus 20.

The CPU 10 controls the overall operation of the data input device 1. The units shown in FIG. 3 are connected by a bus 20, and output control signals and data to control data read / write. I do.

The format file section 11 is composed of a nonvolatile memory such as a flash memory, for example, and stores a plurality of different format files received from a host computer (not shown) via the serial interface 16.

The data file section 12 is composed of a nonvolatile memory such as a flash memory, for example, and stores and saves a data file obtained by character recognition of handwriting input according to each format.

The ROM 13 stores various programs for operating the CPU 10. The various programs include, for example, a main program according to the flowchart of FIG. 8, an initialization processing program according to the flowchart of FIG. A stroke input program according to the flowchart of FIG. 17, a calibration program according to the flowchart of FIG. 18, a character recognition program according to the flowchart of FIG. 19, a candidate selection program according to the flowchart of FIG. 20, a re-input program according to the flowchart of FIG. , And a command receiving program according to the flowchart of FIG. 23 are included.

The work memory 14 is a memory used as a work area when the CPU 10 operates according to various programs, and stores, for example, a data file being created.

The candidate memory 15 is a memory for storing a recognition result (character data of a candidate character) obtained based on a stroke serving as a handwriting of a character, and can be cleared under the control of the CPU 10. As shown in FIG. 4, the candidate memory 15 stores the number of fields (field numbers # 1 to #m) and the number of entry frames (entry frame number # 1) set according to a desired format.
To #n) and the number of candidates (candidate numbers # 1 to # 1) in a three-dimensional matrix memory configuration. No. of each of these fields, entry frames, and candidate characters. Determines the address where the character data is stored.

The buzzer 16 generates a notification sound under the control of the CPU 1. The serial interface 17
This unit is connected to a host computer (not shown) by a cable (not shown) and transmits and receives data. The flash memory 18 is a memory for temporarily storing a data file stored in the data file unit 12 or the work memory 14 when transmitting the data file to the host computer.
The character recognition engine 19 converts the handwriting data input to the frame in the data file into the attributes of the prepared characters (kanji, hiragana, katakana, alphabet, numbers,
Character recognition is performed using each dictionary corresponding to the symbols. In this character recognition, for each field in the above-described format, a dictionary that matches the attribute of the character written in that field is used.

The display unit 5 displays characters based on the character data of the candidates stored in the candidate memory 15. The tablet 3 detects the coordinate position of the stroke from the handwriting and outputs it to the CPU 10. Things. The tablet 3 employs a pressure-sensitive method as an example, and detects the coordinate position from the pen pressure of the handwriting written on the paper P.

Next, the paper P will be described. FIG. 5 is a diagram showing an example of a format of a sheet P used in this embodiment. The paper P shown in FIG. 5 is an example of an address book creation sheet, and its areas are a clip margin 41, an icon area 42, and an address book creation area 4.
No. 3 The clip margin 41 is an area for the paper clip 4 to clip. In the icon area 42, for example, a numeric icon 42a for selecting a number from "1" to "8", a re-input icon 42b for selecting a data re-input process, and a candidate icon for selecting a candidate selection process A candidate icon 42c, an OK icon 42d for selecting OK (OK), a NO icon 42e for selecting NO (no), and an end icon 42f for selecting end (END) are recorded in advance (for example, printing). ) Has been. Icon 42 described above
When one of the areas a, 42b, 42c, 42d, 42e, and 42f is selected by pen input, the coordinate position of the selected position is detected by the tablet 3, and the coordinate position data is sent from the tablet 3 to the CPU 10. Sent. The CPU 10 executes a process of recognizing which icon has been selected from the coordinate position data and a format code indicating the type of format described later.

Reference point markers 44A and 44B for deriving a reference point of the paper P are recorded at the left end of the icon area 42 and the lower right end of the address book creation area 43, respectively. The reference point markers 44A and 44B are provided. Are coordinate position information for recognizing the position and inclination of the sheet P. When a point is input by a pen to any of the reference point markers 44A and 44B, the coordinate position input by the pen is sent from the tablet 3 to the CPU 10 and recognized as a reference point designation from the coordinate position and format.

In the address book preparation area 43, for example, a name entry field 46, an address entry field 48, a zip code entry field 49, a telephone number entry field 50,
Comment entry field 51, map entry field 5
2, and the format specification field 53 is recorded (printed) in advance.
Have been. In the name entry field 46, the address entry field 48, the postcode entry field 49, the telephone number entry field 50, and the comment field 51, entry frames W for entering letters and numbers are arranged. The handwriting entered in the entry frame W is a character recognition target. The name entry field 46 is divided into areas 46a, 46b, 46c, and 46d for entering kanji and katakana as their readings for "last name" and "first name", respectively. The attributes of the characters entered in the name entry field 46 include, for example, kanji, hiragana, katakana, and the alphabet. Address entry area 48
Is a first address entry area 48a for entering "address 1"
And a second address entry area 48b for entering "address 2". For example, kanji, hiragana, katakana, alphabets, and symbols correspond to the attributes of characters written in the address entry area 48. The attribute of the character entered in the postal code entry field 49 corresponds to, for example, a numeral. The attributes of the characters to be entered in the telephone number entry field 50 include, for example, numbers and symbols. For example, kanji, hiragana, katakana, alphabets, and symbols correspond to the attributes of the characters written in the comment field 51. The map entry field 52 is used as an area for writing a figure that is not a character recognition target.
0 is recognized as image data (dot data). The format specification field 53 is an area for inputting a plurality of points prepared in advance, and the position and number of the input points indicate a format code indicating the type of format. This format code is recognized by the CPU 10 based on the position and number of input points.

Next, the operation will be described. FIG. 6 is a diagram showing a display example of candidate characters at the time of address entry, FIG. 7 is a diagram showing an example of correction at the time of address entry, FIG. 8 is a flowchart illustrating main processing, and FIG. 9 is a description of initialization processing. Flowchart, FIG. 10 is a diagram showing an example of an initial screen of the display unit 5, FIG. 11 is a diagram showing an entry example and a display example at the time of address entry, FIG.
FIG. 17 is a flowchart illustrating stroke input;
18 is a flowchart illustrating a calibration process, FIG. 19 is a flowchart illustrating a character recognition process, FIG. 20 is a flowchart illustrating a candidate selection process,
FIG. 21 is a flowchart for explaining the re-input processing.
FIG. 23 is a flowchart for explaining data storage processing, and FIG. 23 is a flowchart for explaining command reception processing.

Before using the data input device 1, the user sets the paper P on the tablet 3 as an initial setting. At this time, the user lifts the paper clip 4 and abuts the vertical tip P1 of the paper P against the paper position determination unit 7 as shown in FIG. 2B, and adjusts the left and right positions appropriately. Then, lower the paper clip 4. At this time, the paper pressing surface 4 a of the paper clip 4 presses the paper P by the bias of the clip restraining spring 8, but receives the buffer of the paper pressing and clip cushioning material 6 provided on the tablet 3, and Since it does not reach the surface, it is not necessary to damage the surface of the tablet 3.

When the paper P is arbitrarily fixed on the tablet 3 as described above, the power is turned on by operating a power switch (not shown) in order to start pen input from the paper P. This power-on may be performed before the paper P is set. Note that the setting of the paper P does not need to be accurate, and correction on the data can be performed in accordance with the later calibration processing.

After the power is turned on, initial settings are first performed in the main processing (FIG. 8; step S1). This initialization is performed as shown in FIG. 9 by initializing the hardware in step S101 (serial interface 17).
, The opening process of various files in step S102, the initialization process of the candidate memory 15 in step S103, and the formation of the input mode screen of the display unit 5 in step S104.

When the initialization process is completed in step S1, the stroke (coordinate) input of the tablet 3 is sensed in the following step S2. This step S
In the stroke input of No. 2, as shown in FIG. 12, if the input stroke is unprocessed, that is, unrecognized in the currently input field of the current format, the stroke data (in the work memory 14) of the stroke is stored. The process is transferred to the character recognition engine 19, and the process returns to step S2 of the main process. On the other hand, if there is no unprocessed stroke (step S201), the status of the tablet 3 is checked in step S203, and a new stroke input is sensed in the subsequent step S204.

In steps S204 to S211,
When the coordinate position is detected according to the stroke input,
Icon input, character input, image input, field cancellation, based on the coordinate position data indicating the coordinate position,
It is determined whether the input is a reference point input or a reference point input.

As a result, if the input is an icon (step S207), this is a selection of one icon in the icon area 42, and the process proceeds to step S301 (see FIG. 13).
Then, the icon code of the icon corresponding to the input coordinate position is generated. If it is a character input (step S208), this is the address book creation area 4
3 until one character is input and the stroke data of the input character is input for one character (FIG. 14; step S311), in step S312, each stroke data is transferred to the character recognition engine 312, or When the input of one character is completed (entry in another entry frame), the entire stroke data of the one character is saved in the work memory 14 in step S313. If the input is an image (step S20)
9) This is the map entry area 5 of the address book creation area 43
2 is entered in the map, and the process proceeds to step S321.
The process proceeds to (see FIG. 15), and the input stroke data of the map is converted into image data. If the input is a field cancellation (step S21)
0), which means that all the input strokes of one field are cleared, and the process proceeds to step S331 (FIG. 16).
Reference), and the data of the field to be canceled is cleared from the candidate memory 15. If it is a reference point input (step S211), this specifies a reference for accurately inputting a stroke input in correspondence with a format, and the processing is performed in step S341 (FIG. 17).
Reference), and the first reference point, the reference point marker 4
4A coordinate position data is stored in the work memory 14.

As described above, when the stroke data is transferred to the character recognition engine 19 in step S202, when there is no new stroke input in step S204, when an icon code is generated in step S301, in step 312 or S313, If the stroke data has been output to the character recognition engine 19 or the work memory 14, the process proceeds to step S321.
If the image conversion is performed in step S33,
1 if the field is cleared, or in step S341, the first reference point (reference point marker 44
When A) is specified, the processing is performed in step S of the main processing.
Move to 3.

Here, the stroke detected by the tablet 3 is a coordinate position, and corresponds to a writing position on the paper P fixed on the tablet 3. The coordinate position is fetched from the tablet 3 to the CPU 10 as coordinate position data. Note that a pen such as a ball-point pen and a pencil is used for writing on the paper P as usual for recording on the paper. This pen is used in the address book creation area 43 according to the format of the paper P.
This is used for writing characters and graphics into the area and for specifying a point for the icon area 42.

What is determined from the stroke input in step S2 is one of a reference point (step S3), an icon (step S5), a candidate (step S10), a re-input (step S12), and an end (step S15). Become. However, this main processing is performed as shown in FIG.
First, a flow of sensing a reference point input, and sensing an icon input when the reference point input is not detected,
If there is a reference point input, the process proceeds to step S4. If neither the reference point input nor the icon input is performed, the process proceeds to step S6. If there is an icon input, the process proceeds to step S10 and subsequent steps.

Therefore, in the stroke input in step S2, the input of the first reference point is sensed first (step S211), and when the input of the first reference point (reference point marker 44A) is detected. The processing shifts from the step S3 to the calibration processing of the step S4.

The details of the calibration process are shown in the flowchart of FIG. In the calibration process shown in FIG.
A message (for example, “Tap the second reference point”) instructing the input of the second reference point (reference point marker 44B) is displayed on the display unit 5. Next step S402
Then, a stroke input from the user is sensed according to the message displayed in step S401. In this case, the stroke input is performed by specifying a point, and the positional relationship (distance, etc.) between the first reference point and the second reference point is the same regardless of the format of the paper used in the data input device 1. Therefore, if the positional relationship is not largely shifted, the input first and second reference point markers 44
A and 44B are used to calculate the vertical and horizontal deviations and inclinations between the arrangement serving as the reference of the format and the arrangement of the paper P set by the user, and the deviations and inclinations are input as stroke data. Is stored in the work memory 14 as coordinate correction data at the time of (step S403).

As described above, when the reference for coordinate correction is determined in step S403, in step S404,
A message instructing input of the format code (for example, “Touch the format code”) is displayed on the display unit 5. In the next step S405, input of a format code from the user is sensed according to the message displayed in step S404. The input of the format code is performed by specifying a point in the format specification field 53 of the paper P, and the CPU 10 recognizes the format code (format type) from the position of the specified point (step S406). As a result of this recognition, the type of the format, that is, the format name is displayed as “ADDR”, for example, as shown in FIG. At this point, the display screen of the display unit 5 is in a state in which there is no display based on the determination of the field name and no display of candidate characters in the field.

When the format code is recognized in step S406, a format file corresponding to the format code is opened in the format file unit 11 in step S407. In the following step S408, fields and icon areas are provided in the candidate memory 15 based on the opened format file, that is, the format contents of the paper P (initialization). Thereafter, the processing shifts to the stroke input processing (step S2) of the main processing.

As described above, when the calibration process in step S4 is completed, the format for the stroke input (the format of the paper P) is determined, and the input stroke can be subjected to coordinate correction to create a data data file. Become.

Thereafter, for example, a case where a stroke is entered in the address entry field 48 in the address book creation area 43 will be described.

In this case, in the main processing, first, it is determined that the inputted stroke data is the first stroke of the character, and in which field the stroke data is written based on the opened format file and the calculated coordinate correction data. Is determined (step S20).
8, S311), and the stroke data is transferred to the character recognition engine 19 (steps S311, S31).
2). Since the target field is the address entry field 48, the discriminated field has the attributes of kanji, hiragana, katakana, alphabets, numbers, and symbols. The attribute of this character is set as a field attribute.

When the above stroke input is completed, the main processing shifts to step S8. In step S8, when the end of one character is confirmed, the process proceeds to step S9, and the character recognition process shown in FIG. 19 is executed. In this character recognition process, first, the field attribute determined in step S901 is
And the process of setting the recognition target is performed. In the following step S902, the character recognition engine 19 is instructed to start character recognition. Character recognition engine 19
Receives the start instruction, refers to the dictionary of the attribute to be recognized, and starts character recognition based on the stroke data already received.

As described above, after the instruction to start character recognition is given, a process of sensing the end of recognition to the character recognition engine 19 is executed. If the recognition end is detected (step S903), the character recognition engine 19
The candidate character data that is the recognition result of each character is sent,
The candidate character data is sequentially stored in the candidate memory 15 (step S904).

In the sheet P shown in FIG. 11A, the entry frames W1 to W6 of the first address entry area 48a include, for example, "East", "Kyoto", "Tokyo", "Wings", "village",
Six characters of "city" are written. In this case, in step S905, the display unit 5 lists the first candidate character of each of the entry boxes W1 to W6 following the format name "ADDR" and the field name "address 1" as a result of the character recognition. At this time, if even one character is displayed differently from the entered character, a candidate selection process described later (see FIG. 20) is performed.

After displaying the first candidate character of each entry frame in step S905, the process returns to step S2 of the main process.

The details of this candidate selection processing are shown in the flowchart of FIG. In step S1101, the format name “ADD” is displayed on the display unit 5 shown in FIG.
R ", field name" address 1 ", and then each entry box W1-
For W6, the second and subsequent candidate characters are also listed and displayed. The candidate characters listed in each of the entry boxes W1 to W6 are:
It is displayed in correspondence with the candidate number, and the user can specify the character by selecting the candidate number.
At that time, first, the entry frame 1 is highlighted and selection of candidate characters is started from the entry frame 1 (step S110).
2).

In the entry box 1, as shown in FIG. 11B, candidate characters "east", "car", and "bundle" are listed in correspondence with candidate numbers 1 to 3. In selecting one of the candidate characters, the character can be selected from eight numbers of the number icon 42a. The eight numerals of the numeral icon 42a correspond to the candidate numbers 1 to 8, respectively. For example, when "1" of the numeral icon 42a is selected, the candidate number of the currently highlighted entry frame is displayed. This means that 1 has been selected. After step S1102, input of stroke data is awaited. Here, selection of any of the No icon 42e, the number icon 42a, and the re-input icon 42b becomes valid.

If the stroke data in step S1103 has selected the No icon 42e, the process proceeds to step S1105, where the first candidate is blank. As described above, when the No icon 42 e is selected at the time of selecting a candidate, all candidate characters listed corresponding to the currently highlighted entry frame are invalidated and cleared from the candidate memory 15. Accordingly, the No icon 42e invalidates the entry frame candidate that is being displayed in reverse, and thus the character can be re-input to the entry frame.

If the stroke data in step S1103 has selected any one of the numeric icons 42a, the process proceeds to step S1107, and the selected candidate character is set as the current first candidate. The character is displayed in place of the character displayed at the position of candidate number 1.
In the following step S1108, only the replaced first candidate character is displayed. By this exchange, the character is fixed in the currently highlighted entry frame,
Therefore, other candidate characters are cleared from the display unit 5.

Further, in step S1109, it is determined whether or not the character of the entry frame is determined to be the last entry frame. If the final entry frame candidate is not selected, the process proceeds to step S11.
Then, the process returns to step S1103, and if the last entry frame candidate is selected, the process proceeds to step S1111 to switch the display unit 5 to the input mode screen and set the main mode. The process returns to step S2.

For example, in FIG.
The characters corresponding to candidate number 1 are "East" and "
“Tei”, “To”, “Ju”, “Village”, and “City”, and the characters actually entered in the entry boxes 1 to 6 of the sheet P are “East”, “Kyo”, “To”, Since they are "feather", "village", and "city", there are two entry frames, W2 and W4, in which the first candidates need to be replaced by the numeric icons 42a. The candidate number “3” is correct for the entry frame W2, and the candidate number “2” is correct for the entry frame W4. Therefore, in the case of FIG.
In 1103, the numerical icons 42a are displayed as “1”, “3”, “1”, “2”,
Selection of “1” or “1” (stroke data input) is ideal.

When the numeral icon 42a is selected in accordance with this ideal, the characters that have become the first candidates of the entry frames W1 to W6 are "East", as shown in the display section 5 shown in FIG.
"K", "To", "Feather", "Village", and "City" are listed and displayed (step S1108).

If there is no character corresponding to the character listed as a candidate in the candidate selection processing, the character can be re-input from the paper P in this candidate selection to list the candidates again.

Therefore, the icon is not the No icon 42e or the number icon 42a, but the re-input icon 42b.
If it is (step S1112), the process proceeds to step S1113 to execute a re-input process.

The details of the re-input processing are shown in the flowchart of FIG. In FIG. 21, when the re-input icon 42b is selected, first, in step S1301, a message indicating the re-input mode is displayed on the display unit 5.
Will be displayed. Then, steps S1302 to S130
In step 4, the stroke data is transferred to the character recognition engine 19 for each stroke data, and the process proceeds to step S1305 with the end of one character. The end of this one character is determined by measuring a certain time and elapse of the certain time. In this re-input process, a single character written in any size and on the paper P is to be recognized irrespective of a field or an entry frame. As described above, the character "feather" can be rewritten in a place where the character "wing" (entry frame 4) for which no correct answer was obtained can be easily removed, for example, by removing the previous entry. in this way,
The same operation as actually writing on paper can be obtained.

In step S1305, control for activating the character recognition engine 19 is performed.
In step 6, a process of fetching the character data of the candidate character, which is the recognition result of one character, from the character recognition engine 19 and storing it in the candidate memory 15 is executed. This step S13
After 06, the process proceeds to step S1114 of the candidate selection process.
Move to

In this step S1114, the candidate characters obtained by the character recognition engine 19 are enumerated and displayed in correspondence with the candidate numbers of the entry frames to be re-input. Also in this case, the entry frame that has become the candidate re-display target is highlighted.

In the main process of FIG. 8, even when the icon input is the re-input icon 42b, the same process as in step S1113 performed in the above-described candidate selection is executed. In the case of the re-input processing from the main processing, in step S1306, the candidate memory 1
After the character data of the candidate character is captured in 5, the process proceeds to step S14, and the first candidate character is displayed.

In order to perform re-input, in the above-described candidate selection processing, the entry frame in which the first candidate was blank by the selection of the No icon 42e was targeted for re-input. It is possible to set a target for re-input. In this case, the target of clearing and re-entering is 1
It is an area for the field.

In step S6, when the user draws a line in one stroke from the left end to the right end of the field, the CPU 10 interprets the field as an instruction to cancel the field and clears the data of the field from the candidate memory 15 ( Step S7). Thereafter, when the re-input icon 42b is selected, a character re-input process is executed on the canceled field in the flowchart shown in FIG.

When the user has completed all the entries on the sheet P, the end icon 42f is selected. When the selection of the end icon 42f is detected (step S1)
5), the processing is step S for performing data storage processing
Move to 16.

The details of the data storage process are shown in the flowchart of FIG. The data saving processing shown in FIG. 22 is processing for finally saving as a data file after the candidate selection processing (see FIG. 20) has already been completed. In step S1601, the character is recognized and the first
Character data stored as candidates are to be saved and are stored in the data file unit 12 as one data file. After storing this data file, step S
At 1602, all the contents stored in the candidate memory 15 are cleared, and the process returns to step S2 of the main process.

In the main processing, if a host computer is connected to the serial interface 17,
Command reception is sensed from the host computer.
Details of the command receiving process are shown in the flowchart of FIG.

In FIG. 23, when a command is received from the host computer, the instruction content of the command is determined. As a result, if the command instructs upload, the process proceeds to step S1702.
In step S1702, the data file unit 12
Is read and transmitted to the host computer, the data file section 12 is cleared in the following step S1703. On the other hand, if the command has instructed download, the process moves to step S1704. This step S170
In step S1705, when the format file data is received from the host computer in step S4, the received format file data is stored in the format file unit 1.
1 is stored. At this time, when the candidate icon 42c in the icon area 42 is selected with the pen, the icon code is generated (steps S207 and S301), and the process proceeds to step S11 for candidate selection in subsequent step S10. Is done.

As described above, according to this embodiment, characters and figures are written from the paper P in accordance with the format and data input is performed, the input data is subjected to character recognition in accordance with the format, and characters are recognized in accordance with the format. Since the data file is created and stored in the data file unit 12 or transferred to an external computer, a natural feeling of use such as writing on paper at the time of data input can be realized.

In addition, since the candidate characters which are the result of character recognition for the data input from the paper P are displayed on the display unit 5 and the characters are selected, the display target is the candidate characters. Display information can be minimized. Thus, the size of the display unit 5 can be reduced, and the entire device can be reduced in thickness, size, and weight.

Also, when inputting data for the first time, character recognition is performed by using a predetermined format, but when data is re-input, character recognition is performed without using the predetermined format. When data is re-entered by character correction or the like, characters can be written at distant positions without overwriting characters once written on the paper P. As a result, it is possible to realize a natural feeling of use such as writing on paper even when re-inputting data.

Further, since re-input is selected using the re-input icon 42b in the icon area 42 of the paper P, troublesome operations and designations for re-input become unnecessary.
It is easy to provide a function like an icon on the paper P, whereby it is possible to reduce the size of the display unit 5 while minimizing the display information to a necessary minimum. By this re-input, it is possible to sufficiently recover from unrecognition or erroneous entry (wrong writing, etc.).

In addition, since character recognition of input data is performed for each entry frame in the form, continuity can be obtained in writing characters (data input) to each entry frame, thereby actually writing on paper. Such a natural feeling of use can be realized. In particular, when selecting candidates, selecting multiple characters while displaying candidate characters collectively for multiple entry frames provides continuity of operation, so that users who are unfamiliar with operation can get used to it immediately. Can be.

Further, since data recognition is performed with different attributes for each entry frame in the format, useless processing such as character recognition for all attributes for one entry frame can be omitted. Thus, the efficiency of character recognition, the recognition rate, and the recognition speed can be improved.

Since the reference point and the format of the paper P are set or switched only by specifying the position with respect to the paper P, the data written on the paper P in accordance with the reference point can be accurately determined according to the desired format. And the format can be easily switched to the desired type.

In the above-described embodiment, a desired format is switched by inputting a format code from the format entry field 53. At this time, the type of format and the format code corresponding to the type are changed. The corresponding format list may be displayed on the display unit 5 so that the user can select another format by referring to the format list without knowing the format code.

In the above-described embodiment, the serial interface 17 is used for connection with the host computer. However, the present invention can be applied to a parallel interface.

[0078]

According to the present invention, one field of handwritten data is entered in a field divided into a plurality of entry frames, and then the character recognition candidates are determined collectively for one field. Accordingly, the writing operation is not interrupted by the determination of the character recognition candidate for each entry frame, and the writing data can be input with a natural feeling of writing on paper with a pen.

[Brief description of the drawings]

FIG. 1 is an external perspective view showing a shape according to an embodiment of a data input device according to the present invention.

FIG. 2 is a side sectional view of a main part of the data input device shown in FIG.

FIG. 3 is a block diagram showing an example of the internal configuration of the data input device shown in FIG.

FIG. 4 is a diagram illustrating a memory configuration example of a candidate memory;

FIG. 5 is a diagram illustrating a format example of a sheet.

FIG. 6 is a diagram showing a display example of candidate characters at the time of address entry.

FIG. 7 is a diagram showing an example of correction at the time of address entry.

FIG. 8 is a flowchart illustrating main processing.

FIG. 9 is a flowchart illustrating an initialization process.

FIG. 10 is a diagram illustrating an example of an initial screen of a display unit.

FIG. 11 is a diagram showing a display example and an entry example when an address is entered.

FIG. 12 is a flowchart illustrating stroke input.

FIG. 13 is a flowchart illustrating stroke input.

FIG. 14 is a flowchart illustrating stroke input.

FIG. 15 is a flowchart illustrating stroke input.

FIG. 16 is a flowchart illustrating stroke input.

FIG. 17 is a flowchart illustrating stroke input.

FIG. 18 is a flowchart illustrating a calibration process.

FIG. 19 is a flowchart illustrating a character recognition process.

FIG. 20 is a flowchart illustrating a candidate selection process.

FIG. 21 is a flowchart illustrating a re-input process.

FIG. 22 is a flowchart illustrating data storage processing.

FIG. 23 is a flowchart illustrating a command receiving process.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Data input device 2 Base panel 3 Tablet 4 Paper clip 5 Display part 6 Paper press and clip cushioning material 7 Paper position determination part 8 Clip restraining spring 10 CPU 11 Format file part 12 Data file part 13 ROM 14 Work memory 15 Candidate memory 16 Buzzer 17 Serial interface 18 Flash memory 19 Character recognition engine 20 Bus

Claims (4)

[Claims] 1. A field defining means for defining a field divided into a plurality of entry boxes on a writing area, and writing data written in each entry box of the field defined by the field definition means is written in each of the above-mentioned fields. Character recognition means for recognizing a character for each entry frame; display means for displaying a list of a plurality of recognition candidates recognized by the model recognition means in association with each of the entry frames; Selecting means for selecting a desired candidate from the candidates for each entry frame, a data input device comprising: 2. When new writing data is re-input to a desired entry frame in a state where a plurality of recognition candidates associated with each entry frame are displayed, the writing data is re-input to the entry frame. 2. The data input device according to claim 1, wherein characters are recognized as input characters and displayed as recognition candidates for the entry frame. 3. When a specific symbol is written on a desired entry box for correcting the entry data, the entry data written on the entry box is erased, and a new entry of the entry data in the entry box becomes possible. The data input device according to claim 1, wherein 4. When a line is written in one stroke over both ends of a field on a field where the written data is to be corrected, the written data written on the field is erased, and new written data for the field is deleted. 2. The data input device according to claim 1, wherein the data input device enters an input enabled state.