JPH04195670A - Handwritten character recognizing japanese syllabary to chinese character conversion system - Google Patents

Abstract

PURPOSE:To improve the efficiency of document creation by handwritten character input by providing an input area for each character kind, performing character recognition and KANA (Japanese syllabary)-KANJI (Chinese character) conversion for each input area, and writing input characters distinctively in the respective input areas. CONSTITUTION:A CPU 21 controls the whole information processing unit, which is provided with a character recognizing function for recognizing character patterns entered into a specific input frame by referring to a recognition dictionary, a KANA-KANJI converting function which performs the KANA-KANJI conversion of a character string generated by the character recognizing function by referring to a conversion dictionary, etc. The input areas are set by the kinds of characters to be inputted and the KANA-KANJI conversion of the recognition result of a character recognizing means corresponding to the set input areas is performed according to coordinate sequences inputted in the set input areas. Consequently, the KANA-KANJI conversion using handwritten character recognition can be performed through easy operation and the burden on the character recognition is lightened.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a handwritten character recognition kana-kanji conversion method in an information processing device such as a slip creation device, an electronic notebook, or a handwritten word processor.

(Prior Art) Conventionally, a character input method has been considered in which characters written by hand on a coordinate input tablet are recognized by a machine and directly input, and the input characters are converted into kana-kanji. As a device for performing such kana-kanji conversion, there is known a character recognition and kana-kanji conversion device described in Japanese Patent Publication No. 49264/1983. Special Public Service 1986-4926
No. 4 has a kana-kanji conversion designation symbol (
By writing the underline), this character string part is converted into kanji. For example, to input the kanji ``device'', type ``souchi'' by handwriting in katakana, which is easy to recognize, touch the specified position on the tablet with the stylus pen, and then underline the character string ``souchi.'' give On the device side, there is a functional block that detects underlines, and when this functional block detects an underline, it performs kana-kanji conversion processing on the corresponding character string.
The kanji ``device'' is obtained.

(Problem to be solved by the invention) As described above, in kana-kanji conversion using conventional handwritten character recognition, it is necessary to write a kana-kanji conversion designation symbol (underline) for the character to be converted to kanji. Ta.

The present invention has been made in view of the above-mentioned points, and it is an object of the present invention to provide a handwritten character recognition/kana/kanji conversion method that is capable of performing kana/kanji conversion using handwritten character recognition with simpler operations. .

[Structure of the Invention] (Means for Solving the Problems) The present invention has an integrated display human power device configured by stacking and integrating a transparent coordinate input tablet on the display surface of a display device, and In the information processing apparatus, the information processing apparatus displays on the display device, as necessary, a manual area indicating a specific area on the tablet for indicating the processing content and instructing the execution of the processing content. an area setting means for setting an input area, a character recognition processing means for performing character recognition processing based on a coordinate series input in the input area set by the area setting means, and recognition by the character recognition processing means. The apparatus further comprises a kana-kanji conversion means for converting the results into kana-kanji according to the input area set by the area setting means.

The invention further includes a setting means for arbitrarily setting a character type to be input into the manual area set by the area setting means, and the character recognition means performs character recognition processing on the character type set by the setting means. It is configured to perform the following.

(Function) According to such a configuration, simply by inputting different characters into the input area set by the area setting means, the characters are converted into the type of character corresponding to the input area. Furthermore, by arbitrarily setting the type of character to be input, the burden of character recognition processing is reduced.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings. FIG. 2 is a perspective view showing the external configuration of an information processing device equipped with an integrated display and input device according to the same embodiment. In FIG. 2, an integrated display/input device 2 is provided on the upper surface of the device housing 1. As shown in FIG. - The body shape display input device 2 is constructed by integrating an LCD (liquid crystal display) and a transparent tablet for coordinate input into a laminated type, and displays on the same coordinate plane.
It is possible to perform data and human power.

3 in the figure is a stylus pen, which is used to indicate an arbitrary coordinate position on the tablet of the integrated display human power device 2. The stylus pen 3 has a built-in switch that detects contact with the tablet surface.

The detection result of this switch is notified to a processing function section provided inside the device housing 1 via the code 4. The integrated display/input device 2 and the stylus pen 3 are controlled by a processing function unit (such as CP021, which will be described later).

FIG. 3 is a plan view of the information processing apparatus in the same embodiment, and FIG. 4 is a side sectional view. As shown in FIGS. 3 and 4, the integrated display human power device 2 is constructed by integrally laminating a transparent tablet 6 and an LCD 7. In addition,
FIG. 3 shows a state in which menu icons are displayed on the LCD 7.

FIG. 5 is a sectional view showing the structure of the switch built in the stylus pen 3. As shown in FIG. As shown in FIG. 5, in a steady state, the electrode 13 of the pen tip 11 in the stylus pen shaft 10 is in a non-contact state (switched off) by the spring 12. When using the stylus pen 3 to indicate (click) the first position of a character,
Pushed against the force of the pen tip 11 or spring 12,
The electrode 13 portion is brought into contact (switched on). Based on this contact (switch-on) and non-contact (switch-off) information (switch information), it is detected that the writing of one character of the icon instruction is being performed within the coordinate detection effective range of the transparent tablet 6. Note that the on and off states are notified to the CPU 2L via the tablet controller 27, which will be described later, regardless of whether or not the transparent tablet 6 is within the effective coordinate detection range.

FIG. 6 is a block diagram showing the internal configuration of the information processing device. Transparent tablet 6 constituting integrated display input device 2
Transparent electrodes are wired in a matrix in the X and Y directions perpendicular to each other in a glass plate. Pulse voltages are sequentially applied to these electrodes, and the XY coordinate position indicated by the stylus pen 3 is detected by electrostatic coupling.

Details will be described later. The transparent tablet 6 and stylus pen 3 are controlled by a tablet controller 27 (described later), and the LCD 7 is controlled by an LCD controller 25 (described later).

21 is a CPU, which controls the entire information processing device. Reference numeral 22 denotes a RAM, which is a readable/writable memory and is used to store programs that define the operation of the CPU 21 and various data. RAM22
is for storing various information such as information regarding the software keyboard displayed on the LCD 7 (display information, position information, etc.), patterns entered by handwriting on the transparent tablet 6, documents created by the document creation function, etc. It is something. 23 is a ROM, which is a read-only memory that stores a diagnostic program after the information processing device is powered on;
A boot program for starting up the information processing device, processing programs for various processing functions, etc. are stored. The processing functions of the information processing device of the same embodiment include:
A character recognition processing function that recognizes characters by referring to a recognition dictionary for patterns written within a predetermined manual frame, and a Kana-Kanji conversion function that converts character strings created by the character recognition processing function to kana-kanji by referring to a conversion dictionary. etc. are provided. As a result, it is possible to create a form in which "name", "address", etc. are entered manually in an input area set in a form format, or to create a general document containing a mixture of kana, kanji, alphanumeric characters, etc. 24 is a VRAM, which is a bitmap memory for storing display information. 25 is LCD
The controller is responsible for display control such as display information transfer from the VRAM 24 to the LCD 7 and display information transfer from the CPU 21 to the VRAM 24. 27 is a tablet controller that scans the X and Y electrodes of the transparent tablet 6, detects the voltage at the position indicated by the stylus pen 3 (capacitive coupling method), generates XY coordinate data,
It also controls a series of tablet controls, such as sending XY coordinate data to the CPU 21. Reference numeral 29 denotes an external interface, which is used for exchanging data between the present apparatus and an external device (a printer, an external storage device, etc.) as necessary.

FIG. 7 is a schematic configuration diagram for explaining the operation of an information processing apparatus equipped with an integrated display and input device.

The information processing device in this embodiment adopts an online character recognition method, and when a character is handwritten on a transparent tablet 6 with a stylus pen 3, coordinate data indicating the stroke of the character is input, and based on this coordinate data, Performs character recognition processing.

Coordinate data indicating strokes of characters, etc. is input to the coordinate input section 3.
0 (including the transparent tablet 6, stylus pen 3, and tablet controller 27). The coordinate human power unit 30 generates coordinate data based on a position indicated by a pen (not shown) on a flat tablet. The coordinate human power unit 30 includes a processing control unit 31 (CPU 21.
(including RAM 22, ROM 23, and external interface 29) are connected thereto. The processing control unit 31 includes:
Display unit 32 (LCD 7, LCD controller 25, VRA
M24), and an external device 33 (a printer device, an external storage device for storing dictionary data, etc. necessary for character recognition processing, kanji conversion, etc.) are connected.

The transparent tablet 6 of the coordinate input section 30 and the LC of the display section 32
As shown in FIG. 8, D7 has the same dimensions and the same coordinate plane. The transparent tablet 6 and the LCD γ are integrally formed in a laminated manner on the processing control unit 31 . Therefore, the information displayed on the display section 32 can be visually recognized through the coordinate input section 30 from the transparent tablet.

Here, details of the coordinate input section 30 will be explained.

The capacitively coupled transparent tablet 6 used in the coordinate input section 30 is configured as shown in FIG. 9, for example. That is, in the transparent tablet 6, a pair of transparent glass plates 6a and flib are joined via an insulating protective film (not shown). A plurality of transparent electrodes 6c and 6d are arranged in parallel at predetermined intervals on the transparent glass plates 8a and 6b, respectively. Transparent glass plate 6a.

The transparent electrodes Be and 8d are connected to each other so as to be perpendicular to each other. The transparent electrodes 6c and 6d are connected to a shift controller 27, respectively, as shown in FIG.
A shift register 27f operating under the control of e,
Connected to 27g.

The specific operation is as follows.

First, when detecting coordinates in the horizontal direction, a predetermined voltage is sequentially applied to the transparent electrode 6C connected to the shift register 27r from the end. When the electrode to which voltage is applied approaches the stylus pen 3, an electrostatic induced voltage is induced in the stylus pen 3 due to the capacitance formed between the transparent electrode 6c and the stylus pen 3. This electrostatically induced voltage is amplified via an amplifier 27j and a high frequency component is removed via a filter 27k to provide a stable indicated position detection signal, which is then applied to a counter 27h. counter 2
7h starts counting when a predetermined voltage is applied to the electrode located closest to the origin among the transparent electrodes 6c, and ends counting when a designated position detection signal of a certain value or more is manually input. By performing such an operation, the coordinates of the stylus pen 3 are detected in terms of time. When the horizontal coordinate measurement is completed, the vertical coordinate measurement is started. In the vertical coordinate measurement, similarly to the horizontal coordinate measurement, a predetermined voltage is sequentially applied to the transparent electrode 6d connected to the shift register 27g from the end, and from the electrostatic induced voltage detected by the stylus pen 3, the amplifier 27j1 A designated position detection signal is generated by the filter 27k, and the coordinates are detected by the counter 27h.

When characters, etc. are written on a transparent tablet using the stylus pen 3, the information indicating the writing stroke or the connection of the positions indicated by the stylus pen 3, that is, the connection of the indicated position coordinates (series of position coordinates). It will be entered.

In this way, the information on the characters input by hand via the coordinate input unit 30, that is, the sequence of position coordinates, is transferred to the processing control unit 31.
is provided to the display section 32 via the. Then, according to the series of position coordinates, the display unit 32 is driven to display the same coordinates, as will be described later. With this display control, the characters written on the transparent tablet surface are simultaneously displayed as the trajectory of the characters, and character information can be input as if writing characters on paper using the stylus pen 3. There is.

Next, the memory section (RAM22, ROM
23) is shown in FIG. As shown in FIG. 11, it is comprised of processing units 40a to 40r1 that store programs for performing various processes, and data buffers 40g to 40m that hold various data. That is, the processing units 40a to 40f include the coordinate input unit 3.
0, a recognition processing section 40b that performs character recognition processing with reference to a recognition dictionary, a kana-kanji conversion processing section 4001 that performs kana-kanji conversion processing with reference to a conversion dictionary; Area discrimination processing unit 40d1 that determines which area (described later) the coordinate data indicates indicates; acquires data for displaying guides, soft keys, icons, etc. corresponding to the specified area; Guide display data extraction processing section 40e
, and a display control section 40 that controls display on the display section 32.
In addition, the data buffers 40g to 401 that hold various data include a position data buffer 40g that stores the coordinates of the specified position specified in the coordinate input section 30, and a pen state (on, off). ) A pen status data buffer 40h1 that stores data for displaying guides, soft keys, icons, etc. A guide data buffer 401 that stores data for displaying guides, soft keys, icons, etc. Dictionary data for recognizing handwritten characters and converting kana-kanji characters Dictionary data buffer 40j that temporarily holds dictionary data of
1 Area data buffer 4 that stores information regarding each area
Output buffer 401 that holds data that is output as 1 to 0
, a recognition frame data buffer 402 in which data relating to an area where character recognition processing is to be performed is stored, and a numerical data buffer 40n in which numerical data is stored.

Further, the area data buffer 40k is provided with an area data buffer for independently storing data regarding each input area when inputting data.

Here, we will explain the four areas A-D when creating a slip.
1 to 4 in the area 40 for storing various information about
4 is shown at 0. Each area data buffer 40kl ~
40 and 4 are, as shown in FIG. 12, an area coordinate X starting point data buffer 41a, an area coordinate Y starting point data buffer 41b, in which coordinate data indicating the position of the area are stored;
A area coordinate X end point data buffer 41c, pi area coordinate Y end point data buffer 41d, and area attribute data buffer 41 for storing attribute data regarding the area. e is provided. Area attribute data buffer 41. e is an area processing attribute for storing data specifying whether or not to perform handwriting input processing on the coordinate data input to the area (including data specifying whether or not to perform character recognition processing). Data buffer 41el, recognition attribute data buffer 41e2 for storing data specifying dictionary data used when executing character recognition processing in handwritten input processing, whether or not item contents for the area have been input. A manual completion flag 41e3 indicating this is provided.

Furthermore, the recognition frame data buffer 40m includes area data buffers 40m1 to 40m for independently storing data regarding each area in which character recognition processing is performed based on coordinate data handwritten on the tablet.
3 is provided. Two areas are provided for storing various information regarding areas A and B to be subjected to character recognition processing, and area C to which numerical data is input. Each area data buffer 40m1 to 40m3 is the first
As shown in FIG. 3, a frame number data buffer 42a stores data indicating the number of frames provided within the area;
Area coordinates
b1~42bn, area coordinate Y starting point data buffer 420
1 to 42cn, area coordinates X starting point data buffer 42d1
~42dn, area coordinate Y starting point data buffer y 42el
~42en.

There are also recognition result data buffers 42rl to 42f'n in which results of character recognition processing or image patterns (which vary depending on the attributes of the area) are stored.

Note that when creating a general document, the next candidate display area 52 is displayed, for example as shown in the display screen shown in FIG.
, function area 51, and handwritten character input area 5
3 input areas are set. At this time, data buffers for each of the regions 51 to 53 are set in the region data buffer 40k. Further, data for each input frame in each row of the handwritten character input area 53 is also set for the recognition frame delta buffer 740m.

Next, the first step is to explain the processing procedure when inputting character information.
This will be explained with reference to the flowchart shown in FIG.

First, when the device is powered on, an initial screen as shown in FIG. 15 is displayed on the display unit 32 (
Step Al). On this initial screen, a plurality of item names in the slip format and entry frames corresponding to each item name are displayed.

In other words, the item name includes "Name", which indicates a field for entering the name,
There is also an ``Odokoro'' that indicates ``age'' and address.

Furthermore, the name entry field includes fields for writing in kanji and furigana. Here, the name entry column in kanji is area A, the name entry column in furigana is area B, the age entry column is area C1, and the address entry column is area A. In addition, for area A, handwritten character recognition processing for kanji, for area 2, handwritten character recognition processing for katakana, for area C, input using the soft numeric keypad, which will be described later, and for area 2, handwriting display (the image pattern of characters etc. remains). It is assumed that the settings are set to perform processing.

Such information about each area A to D is set in advance in the area data buffer 40k and the recognition frame data buffer 40m within the processing control unit 31.

That is, the data of the coordinates indicating the area position, the processing attribute, and the recognition attribute are stored in the area data buffer 40 corresponding to each area in the area data buffer 40 from 1 to 40.
It is stored in 4. Also, area A.

The number of frames provided in B and the coordinates indicating the frame positions are:
It is stored in the area data buffer 40m1.4(]++2 of each area in the recognition frame data buffer 40s.

The display control unit 40f controls the display of the initial screen based on the data stored in each area data buffer.

When the initial screen is displayed in this way, the input section 40
a enters the input waiting state. That is, it is determined whether coordinate data is detected by pointing the stylus pen 3 on the transparent tablet surface of the coordinate input unit 30 (step A2). When the input unit 40a detects that coordinate data has been input, the input unit 40a transfers the coordinate data to the position data buffer 4.
0g, and data indicating the state of the stylus pen 3 (on or off, etc.) is stored in the pen state data buffer 4.
It is stored in 0h (steps A3 and A4). Coordinate data and data indicating the pen status or a predetermined buffer y40g.

40h, the area determination processing unit 40d moves the stylus based on the coordinate data stored in the position data buffer 40g and the coordinate data indicating the position of the area stored in the area data buffer 4.0k. It is determined in which area the position specified by the pen 3 is included. If it is determined that the input coordinate data is included in one of the areas A to D, information regarding that area is extracted from the area attribute data buffer 41e (step A5). The area determination processing unit 40d determines whether or not the area indicated by the stylus pen 3 is a target for manual handwriting processing based on the processing attribute data stored in the area processing attribute data buffer 41e1 (step A6). ). The area discrimination processing unit 40d uses the stylus pen 3
The input unit 40a is notified of the determination result as to whether or not the area designated by is a target for handwriting input processing. That is, if the designated area is not a target of handwriting input processing, the handwriting input state is turned off, and if it is a target of handwriting input processing, it is set to a handwriting input waiting state (on) (steps A7 and A8). In the same example, area A
, B, and D are specified, the handwriting input processing state is set. Further, if the specified area is to be subjected to handwritten input processing, the area determination processing unit 40d determines whether handwritten character recognition processing is to be performed based on the area processing attribute data (step A9).

If the area is a target area for character recognition processing, the area discrimination processing unit 40c determines the attributes (kanji, katakana, etc.) stored in the dictionary data buffer 40j based on the recognition attribute data stored in the attribute data buffer 41e2. The dictionary data corresponding to the data is taken out and transferred to the recognition processing section 40b (step A10).

Then, the guide display data extraction processing unit 40e determines whether the manual completion flag 41e3 for the designated area is on or off (step A11). If no data has been entered, of course the input completion flag 4
1e3 is off and continues to wait for data input corresponding to each area. Here, the manual completion flag 41e
3 is on and it is determined that data has been input for the designated area, guide data for performing guide display stored in the guide data buffer 40i is retrieved. Guide display data extraction processing section 4
The guide data retrieved by 0e is transferred to the output buffer 401. The display control unit 40f displays a guide message such as "Please press" as shown in FIG. 16(a), for example, based on the ID data in the output buffer 401 (step A12). The example of the guide display shown in FIG. Here, if the stylus pen 3 is on (
In step A13), as shown in FIG. 16(b), the data in the output buffer 7401 regarding this area is cleared (step A14), and the guide display is also cleared.

Then, the input completion flag 41e3 is turned off (step A15). In step A13, if the pen state is off and the manual data is not to be cleared, the process returns to step A2 and enters an input waiting state.

Then, the region determination processing unit 40d determines in which region A to D the designated position is located (steps A1.8 to
A19), processes A-D for areas A-D are executed (steps A, 20-23).

Here, the processing for area A will be explained with reference to the flowchart shown in FIG.

If the position specified by the stylus pen 3 is within area A, handwriting is performed according to the procedure described above (step B1), and the input coordinate position (frame inside), the handwriting is displayed as if it were written on paper (step B2). Based on the coordinate data of the input handwritten character, the recognition processing unit 4Qb extracts the position data (starting point of the frame coordinates, (end point) is taken out (step B3). The recognition 1 processing unit 40b uses the set dictionary data to perform a handwritten character recognition process (described later) on the coordinate data of the input handwritten character (step B4). The recognition processing unit 40b stores the result (character data) of the handwritten character recognition process in the area A data buffer 40Il.
The recognition target in l.

(Step B)
5), character data is stored in the output sofa 401. That is, when a character is written down as shown in FIG. 18(a), the character data of the kanji character "yama" is stored in the recognition data () rough y42fi. In this way, the display control unit 40r displays the result of the handwritten character recognition process, as shown in FIG. 18(b), based on the character data stored in the output buffer 401 (step B6).

Note that the processing for area B is performed in the same manner as the processing for area A described above, or a dictionary regarding katakana is performed in accordance with the recognition attribute data stored in the recognition attribute data buffer 41e2 in the area B data buffer 40. The data is set in the recognition processing section 40b, and handwritten character recognition processing is performed. FIG. 19 shows an example of a display screen when processing for area B is being performed.

In addition, the area processing displays the coordinate data (image pattern) of the input handwritten characters as is,
Does not perform handwritten character recognition processing.

FIG. 20 shows an example of a display screen when processing for this area is performed.

Here, processing for area C in which guide display other than the above is performed will be explained with reference to the flowchart shown in FIG.

First, the area determination processing unit 40d determines whether the stylus pen 3
When it is determined that the position specified by is within area C, the pen state is determined based on data indicating the pen state stored in the pen state data buffer 40h (step C1). Here, if it is determined that the pen state is off, the guide display data extraction processing unit 40e extracts the guide data corresponding to the area C stored in the guide data buffer 401 and transfers it to the output buffer 401.

The display control unit 40 displays a guide message such as ``Please press'' as shown in FIG. 22(a) in accordance with the guide data in the output buffer 401 (step C2).
). In step C, if the pen state is off,
No guide display is performed.

Here, when the pen state is turned on to input data, the guide display data extraction processing unit 40e extracts the guide data stored in the guide data buffer 40+, which indicates the soft numeric keypad used for manually inputting the data. is read out and transferred to the output sofa 401. The display control unit 40f displays a soft numeric keypad at a predetermined position at the bottom of area C, as shown in FIG. 22(b) (step C4).
). Here, when the stylus pen 3 is in the ON state (step C5) and the numeric area portion of the soft numeric keypad is designated, numerical data is input (step C6). That is, when the area of the soft numeric keypad is specified with the stylus pen 3, the area determination processing unit 40d specifies the area of the stylus pen based on the area data indicating the area of each number of the soft numeric keypad stored in the guide data buffer 40i. 3
It is determined whether the coordinates of the indicated position are specified by or which number is specified. When the area determination processing unit 40C determines whether the number is the designated number, the numerical data is stored in the numerical data buffer 4On. Further, character data corresponding to the numerical data stored in the numerical data buffer 40n is stored in the output buffer 7401, and the first of three frames provided in the area indicating "age" by the display control unit 40r is stored in the output buffer 7401.
It is displayed in the digit (step C7). In this way, when an area indicating a number is designated with the stylus pen 3, the numerical data is sequentially stored in the numerical data buffer 40n as the least significant digit (first digit), and the Display in the first digit. That is, the display of the numerical data input earlier is carried up in sequence. here,
If the input numerical data exceeds 3 digits (Step C)
8) clears the numbers displayed in each frame of the numerical data buffer 40n1 and area C (step C9).

When the area indicating the end of the soft numeric keypad is designated by the stylus pen 3, the display of the soft numeric keypad is cleared and data input for area C is ended (steps C1, 0 to C11).

Next, the processing procedure from the coordinate input unit 30 to displaying characters input by writing them on the transparent tablet 6 will be explained with reference to the flowchart shown in FIG. 23.

First, when inputting data, the processing control unit 31 initializes each buffer for storing human data (step DI). The processing control unit 31 checks whether the stylus pen 3 is on (step D2). When the state of the stylus pen 3 is on, the coordinate data of the position indicated by the stylus pen 3 inputted from the coordinate input section 30 is taken into the human power section 40a, and the position data buffer 40
g (step D3). This storage of coordinate data is continued until the position data buffer 40g is notified of delimiter data indicating that all coordinate data sequences for one character have been input. That is, the delimiter data is set in the off state of the stylus pen 3, and when this off state is determined in step D2, the position data buffer 40g is notified and stored (step D4). In addition to the processing method described above, the end of inputting coordinate data for one character is determined based on whether the input coordinate data indicates a position outside the frame area that is currently being input. . For example, if you change the coordinate data input target from the current input target frame to another frame among the multiple frames indicating character entry positions provided on the initial screen, the coordinate data input target is changed from the current input target frame to another frame. It is determined that the manual input of the coordinate data for one character in the selected frame has been completed. In this way, the recognition processing time of character recognition processing based on coordinate data is shortened. In this way, the input 1
The character coordinate data is stored in the position data buffer 40g and transferred to the output buffer 401. Then, the display control unit 40 simultaneously displays the information at a predetermined position on the display unit 32 as writing pattern information.

When the sequence of coordinate data for one character is thus stored in the position data buffer 40g, the recognition processing unit 40b analyzes the coordinate data sequence and performs character recognition processing (
Step D6. D7).

This character recognition process, for example, separates a series of coordinate data into military writing strokes by determining stroke breaks based on the distance between each positional coordinate data, and converts each written stroke into a series of values indicating the direction representing its characteristics. It is done as follows. Here, the value indicating the direction is, for example, as shown in FIG.
It is quantized and given in eight directions (direction values). Then, find out what kind of direction value series represents the series of coordinate data indicating each writing stroke, and if the same direction (direction value) continues, combine those direction values into one direction value. The data is compressed by representing each writing stroke, and the position coordinate data representing each writing stroke is converted into a direction value. According to this quantization, for example, the number "3" is the second
As shown in Figure 5, it is converted as a direction value r252357J.

Through this processing, coordinate data indicating a character manually written via the coordinate input unit 30 is expressed as a direction value for each written stroke.

The recognition processing unit 40b compares the information on the direction value sequence obtained in this manner with dictionary data for performing character recognition processing provided via the dictionary data buffer 40j stored in the external storage device 29b. to recognize characters. The dictionary data used for the character recognition process, for example, as shown in FIG. It is categorized into 1 and associated with each character code. The recognition 1 processing unit 40b refers to dictionary data such as 2 and sets a character code corresponding to the stroke characteristics of the coordinate data indicating the handwritten input character as a recognition result (step D8).

Here, if the input coordinate data is for the first character, the character size for character display is set. That is, the size of the written character in the X and Y directions is detected based on the series of manually entered coordinate data (step D10). Then, the X size and Y size are compared (step DLL), and depending on the result,
Set the size in the X and Y directions as the font size according to either the size or the Y size (
Ste Tsubu D1. L~D14). Then, the display control section 4
0e displays characters according to the set character size (step D15). Note that if the input coordinate data is for the second character or later, the process of determining the character size is not performed, and the characters are displayed according to the character size set by the data for the first character.

In this way, as a result of manually inputting character information to the initial screen shown in FIG. 15 using a method corresponding to each area A to D, a display screen as shown in FIG. 27, for example, is obtained.

Next, a method for handwritten character recognition and kana-kanji conversion will be explained.

Here, character recognition is performed by handwriting characters in a predetermined area, rather than inputting them into the character input format described above, and the results are converted into kana-kanji to create a general kana-kanji mixture. It is possible to create written sentences.

FIG. 28 shows an example of a display screen when a document is created by handwritten character recognition and kana-kanji conversion. This document creation screen is roughly divided into four areas.

That is, a created document display area 50 for displaying a created document, a function area 51 for displaying various function keys, a next candidate display area 52 for displaying the next candidate for character recognition or kana-kanji conversion,
and handwritten character manual area 5 for manually creating handwritten characters
It is 3. The function area 51 includes, for example, a mode key 54 for instructing the character type of human characters (katakana, numbers, alphabets, etc.), a document display area for displaying character strings (for example, part of phrases) created in the handwritten character input area. 5
An rMOVEJ key 55 for transferring the characters to 0 and a "kanji conversion" key 56 for prompting execution of kanji conversion when the handwritten character input area is not filled with characters are provided.

In the handwritten character input area 53, in FIG.
A line is provided. In the figure, a manual frame is provided so that 12 characters can be entered in each line, but this number of characters is not particularly limited. For the characters written in the first line of the handwritten character human power area 53, recognition dictionaries for ``hirakana'', ``katakana'', and ``alphanumeric characters'' are used, and for the characters written in the second line, ``hiragana'' and the number of strokes are used. A recognition dictionary for kanji with few characters is used. Therefore, in the second line, any kanji character with a small number of strokes can be entered as is. For example, if you are trying to get the word "Japanese" in kanji, write "Japan" in kanji on the second line, which has a small number of strokes, and write "go" for the kanji "go", which has a large number of strokes and is difficult to recognize. Fill out. As a result, the character string ``Nihongo'' is recognized and subjected to kana-kanji conversion.

Note that the kanji that can be recognized in the second line is, for example, the number of strokes N.
A recognition dictionary is prepared for kanji within a certain range,
This allows reliable character recognition. The number of strokes N takes an appropriate value depending on the ability of the character recognition processing function.

Next, the procedure for creating a document using the display screen shown in FIG. 28 will be explained with reference to the flowchart shown in FIG.

First, in the handwritten character area of the first and second lines,
Characters are entered separately into parts that should be written as kanji and other parts (step El). Recognition of handwritten characters written in each rectangular frame requires three stylus pens for each character.
After a certain period of time has elapsed after the input is stopped, it is assumed that the manual input has been completed, and character recognition is started based on the coordinate data series input manually up to that point. Alternatively, even if writing is stopped and characters are started in another writing frame even within a certain period of time, recognition of characters written in the previous writing frame is started.

As shown in FIG. 28, when a character is written in the handwritten character area 53 and the character recognition result is displayed (step E
2) If there is an error in the recognition result, the characters are corrected (step E3). The character is corrected, for example, by re-entering the character within the rectangular frame of the erroneous character (step E4). When the results of the character recognition process no longer require correction, execution of the kana-kanji conversion process (step E5) or transfer to the created document display area 50 is instructed (step E6). Here, rMOVE
The J key 55 is designated, for example, when the input character string only includes "kana" or when it includes only simple kanji characters with a small number of strokes.

Here, when the "Kanji conversion" key 56 is designated with the stylus pen 3, the kana-kanji conversion process is executed for the characters entered manually on the second line (step E7). In the kana-kanji conversion process, the characters input in the second line of the handwritten character manual area 53 are converted into kanji by referring to a dictionary for kana-kanji conversion.

In FIG. 28, it is assumed that the character string ``Eikoku'' is converted to ``UK'' and the character string ``Kishoji'' is converted to ``Owakeji''. The number of characters may change due to kana-kanji conversion, and blanks may appear in the input frame, but when the input frame is transferred to the created document display area 50, the characters are filled in and the input frame is no longer blank.

If an erroneous conversion result such as "wake-up character" is obtained, a correct character is selected from the next candidates displayed in the next candidate display area 52 and replaced (step E8). The display of the next candidate in the next candidate display area 52 is such that, for example, when a conversion result of "Owakaseji" is displayed for "Kiyoshi", "
This is done for the ``character'' by touching the input frame on the first line directly above the ``character'' with the stylus pen 3. The next candidate is selected by touching the frame in which the desired character is displayed in the next candidate display area 52.

For example, if the entire word "Meteorology" is incorrect, you can touch the frame in the first line just above "Ki" to change the entire word "Kishoshi" to the next one. The candidates are displayed in the next candidate display area 52. If the next candidate is "Wake up character", select this candidate and display it in the input frame, then touch directly above "Character" to select the next candidate with only "Character". Ru.

Note that when performing kana-kanji conversion on characters entered in the handwritten character input area 53, the "kanji conversion" key 56
, but you may also do the following. In other words, stylus pen 3 that was writing characters on the second line
However, when the first line is written, it immediately performs character recognition processing on the characters written on the second line, then performs kana-kanji conversion on the word containing that character, and displays the result in the character entry frame. It is something. Also,
If you are writing characters in the second line and stop writing for a predetermined time, even if the writing does not move to the first line, character recognition will be performed for the characters written in the second line, and the character will be included. It is also possible to perform kana-kanji conversion on the word and display the result in the character witness frame.

In this way, the desired kana-kanji conversion result is obtained, and rM
When the OVEJ key 55 is touched, the character string displayed in the handwritten character input area 53 is transferred to the created document display area 50.

In the above, the number of lines in the handwritten character input area 53 is two lines, and the characters to be entered in the first line are ``hiragana'', ``katakana'', and ``alphanumeric characters'', but the role of the first line is subdivided into numbers. It may be divided into columns and provide a plurality of lines for "alphanumeric characters", "katakana", etc. FIG. 29 shows an example of a display screen with a separate line for inputting ``katakana''. Regarding the characters written in the "Katakana" input line shown in Figure 29,
A recognition dictionary for "Katakana" is used.

Further, the mode key 54 provided in the function area 51 is used to specify the type of character to be input into the input line of the handwritten character input area 53, which facilitates the character recognition process. In other words, in Japanese sentences, kanji and hiragana appear overwhelmingly frequently, so the handwritten character input area 53 is made up of two lines, and the second line is a line for entering character strings that should be converted into kanji. Now let's use the first line as the line for inputting hiragana. When the mode key 54 is touched, characters of the specified mode can be entered in the first line, thereby reducing the burden of character recognition processing. The flow chart of FIG. 30 shows the procedure for specifying this mode and inputting characters (in the case of specifying the "Katakana" mode). For example, when creating the sentence ``Morning Tea'', the ``Katakana'' key of the mode key 54 is first touched (step F1), and the user is instructed to enter katakana characters (``Katakana'' mode). or on).

Here, when "Morning Tea" is written in the first line, character recognition results are performed for each character (step F2
．． F3). If there is an error in the character recognition result, correction is made by re-inputting (step F5). When the desired character string is obtained by modification, the "Katakana" key is touched and "Katakana" mode is turned off (
Step F6). Further, when the rMOVEJ key 55 is pressed, the input characters are transferred to the created document display area 50 (step F8). As a result, the "katakana" mode is canceled and input in kana and kanji becomes possible (step F8), and the characters "toi" are written separately on each line. FIG. 31 shows the display screen when the desired sentence has been input in this manner.

When a "number" or "alphabet" of the mode key 54 is touched, the same process is performed depending on the key. Furthermore, although three types of mode keys 53 are provided in the embodiment, more types (character types) may be provided.

In addition, in FIG. 29, there is a row for inputting "katakana", but input rows for "alphabetic characters", "numbers", etc. may also be provided, for example.

Also, whether the integrated display/input device 2 is configured using an electric field coupling method.1. In addition, electromagnetic induction methods and pressure-sensitive methods can also be used.

In this way, lines corresponding to the type of characters to be input are provided in the handwritten character manual area 53, and by writing and writing the characters separately on the predetermined lines, character recognition is performed reliably, and even kana-kanji A conversion is performed. In other words, it becomes possible to create documents manually by handwriting, without having to enter characters or specify the parts to be converted into kanji.

[Effects of the Invention] As described above, according to the present invention, an input area is provided for each character type, and character recognition processing and kana/asaaku conversion are performed for each human power area, so that it is possible to write the character human power separately for each human power area. This makes it possible to reliably obtain the desired kana-kanji conversion result without any other instructions such as the conversion range. Therefore, it is possible to improve the efficiency of document creation by inputting handwritten characters.

[Brief explanation of the drawing]

1 to 31 are diagrams relating to an information processing apparatus according to an embodiment of the present invention, and FIG. 1 is a flowchart showing a processing procedure when performing kana-kanji conversion of a handwritten literary magazine 1;
FIG. 2 is a perspective view showing the external configuration of an information processing device equipped with an integrated display input device, FIG. 3 is a plan view of the information processing device,
FIG. 4 is a side sectional view of the information processing device, FIG. 5 is a sectional view showing the configuration of the switch built into the stylus pen, and FIG.
The figure is a block diagram showing the internal configuration of the information processing device, FIG. 7 is a schematic configuration diagram for explaining the operation of the information processing device, FIG. 8 is a diagram showing the correspondence between the transparent tablet and the LCD, and FIG. FIG. 10 is a block diagram for explaining the operation of the coordinate input section; FIG. 11 is a block diagram showing the contents of the memory section of the processing control section; FIG.
The figure is a block diagram showing the configuration of the area data buffer.
3 is a block diagram showing the structure of the area data buffer in the recognition frame data buffer, FIG. 14 is a flowchart showing the processing procedure when inputting character information, FIG. 15 is a diagram showing an example of the initial screen display, and FIG. FIG. 16 is a diagram for explaining an example of guide display in area A, FIG. 17 is a flow chart showing the data processing procedure for area A, and FIG.
Figure 8 shows an example of the display when inputting handwritten characters to area A, Figure 19 shows an example of the display when inputting handwritten characters to area B, and Figure 20 shows an example of the display when inputting handwritten characters to area B. Figure 21 showing an example of display when performing
The figure is a flowchart showing the data processing procedure using a soft numeric keypad for area C, Figure 22 is a diagram for explaining an example of a guide display in area C, and Figure 23 is a flowchart showing the procedure for manually inputting data using a soft numeric keypad. FIG. 24 is a diagram for explaining the direction of a written stroke. FIG. 25 is a diagram showing an example of converting position coordinate data indicating a written stroke into a direction value. FIG. 26 is a diagram for character recognition. Figure 27 is a diagram showing an example of the structure of dictionary data used for processing, Figure 27 is a diagram showing an example of a display screen with character information input for each area, and Figure 28 is a diagram showing an example of a display screen when character information has been input for each area. Figure 29 is a diagram showing an example of a display screen in which a line for inputting "Katakana" is provided in the handwritten character manual area, Figure 30 is a diagram showing the processing procedure for character input in Katakana Mote. The flowchart, FIG. 31, is a diagram showing an example of a display screen when characters are entered manually in katakana mode. 2... Integrated display input device, 3... Stylus pen, 6... Transparent tablet, 7... LCD (M product display), 31... Processing control section, 40a... Input section, 40b . . . recognition processing section, 40c . . . area discrimination processing unit, 40e . . . display control unit, 40j . . . area data buffer, 401 . Applicant's representative Patent attorney Takehiko Suzue No. 11! liQ Za” 4th test on the actual day] Menu icon Seida Omoteki 15th Figure 3 M 8 Figure 15 Figure 16 Figure 11 Figure 12 Figure 13 Figure 17 Figure 182 Figure 19 Figure 20 Figure 21 Figure 22 Figure 23 Figure 24 Figure 26 Figure 27 Figure 30

Claims (2)

[Claims] (1) To have an integrated display input device configured by stacking and integrating a transparent coordinate input tablet on the display surface of the display device, and to indicate specific processing contents and instruct execution of the processing contents. An information processing device that displays an input area indicating a specific area on the tablet on the display device as needed, comprising: area setting means for setting a plurality of input areas for each character type to be input; and the area setting means. a character recognition processing means that performs character recognition processing based on the coordinate series input in the input area set by; and a character recognition processing means that performs character recognition processing based on the coordinate series input in the input area set by the character recognition processing means, and a recognition result by the character recognition processing means according to the input area set by the area setting means. A handwritten character recognition kana-kanji conversion method characterized by comprising: a kana-kanji conversion means for performing a kana-kanji conversion. (2) Setting means for arbitrarily setting a character type to be input into the input area set by the area setting means; The handwritten character recognition kana-kanji conversion method according to claim 1, characterized in that a recognition process is performed.