JPS61138382A - Recognizing device of hand-written character - Google Patents

Abstract

PURPOSE:To read properly an input character even after character correction by fetching a hand-written character from an input tablet so as to recognize it and providing a printer for printing it result an area except a character writing area on a writing form. CONSTITUTION:When an input pen 10 is brought into contact with the square of the writing form 9 on the input tablet 2 to write a character, the number of the lattice 8 is set in a square number setting area, and simultaneously coordinates at a pen position are written in a coordinate memory area. Then, the dictionary collation processing is applied to an input character pattern composed of a directional code by referring to a dictionary area. When the input character pattern is equal to any of standard pattern, a recognized character is displayed on a CRT5, and simultaneously stored in its corresponding square number area. When a tablet input termination command is inputted, stored character contents are printed in the printing area of the paper 9 by the printer 11, and moreover character data in a character memory area is outputted to an external information processor.

Description

[Detailed Description of the Invention] <Technical Field of the Invention> The present invention relates to characters handwritten on an input tablet.
The present invention relates to a handwritten character recognition device using an online processing method that captures the input characters and executes a dictionary matching process to recognize input characters.

<Summary of the Invention> The handwritten character recognition device of the present invention includes a printer in an input tablet, and when a character is handwritten input on a form on the input tablet, the recognition result of the character is printed on the same form by the printer. It is configured to print, thereby making it possible to read the input characters even if the entry form is soiled due to character corrections, etc., thereby increasing its value as an archival document.

<Background of the Invention> Conventional handwritten character recognition devices include a method in which characters are input by handwriting directly onto an input tablet, and a method in which a predetermined entry form is placed on the input tablet and handwritten characters are input from there. In particular, the latter method is effective in applications where it is necessary to leave handwritten manuscripts, such as in the case of election voting.

In a typical character recognition device, a person writing letters sequentially handwrites them on a form placed on an input tablet, and checks each time on a display device whether or not the written characters have been correctly recognized. It is supposed to be done. When checking the display contents, if a misreading by the device or a typo by the person filling in the characters is discovered, the characters are corrected by overwriting the correct characters over the misread or miswritten characters. Therefore, in the areas where characters are corrected on the entry form, the characters before and after the correction overlap, making it difficult to decipher the characters, which causes this type of form to lose its value as an archival document.

<Object of the Invention> An object of the present invention is to provide a new handwritten character recognition device that does not easily lose its value as an archival document even if the entry form has undergone character correction.

<Configuration and Effects of the Invention> In order to achieve the above object, the handwritten character recognition device of the present invention includes an input tablet having an area for inputting characters by hand, and a pen tip placed on an entry form placed on the input tablet. The input tablet is equipped with an input pen for handwriting inputting characters by touching the input pen, and a recognition processing unit that imports handwritten characters input on the entry form from the input tablet and executes character recognition processing. A printer is provided to print the character recognition results in an area other than the character entry area of the entry form.

According to this invention, even if characters are corrected by overwriting characters when a misreading or writing error occurs, the recognition result is printed in a predetermined area on the entry form, so the input characters can be accurately read. It can increase its value as an archival document. etc., it has achieved the remarkable effect of achieving the purpose of the invention.

<Description of Embodiments> FIG. 1 shows the overall configuration of a handwritten character recognition device according to these three inventions.

The illustrated device includes a personal computer main body 1 (hereinafter simply referred to as "computer main body"), an input tablet 2
, a printer 3, a floppy disk device 4, etc., and these components are electrically connected by cord lines. The computer main body 1 is a CRT (Cathode Ra).
y Tube) 5 and a keyboard 6, and the keyboard 6 is provided with a plurality of function keys for manually inputting various commands. The command input section is not limited to the keyboard, but can also be used with input tablets, which will be described later.
It is also possible to adopt a pen-operated system.

Inside the computer main body 1, there is a CPU (Central CPU).
l Processing Unit), RA
M (Random Access Memory),
A control processing section including a memory such as ROM (Read Only Memory) is incorporated, and this control processing section controls input/output operations in each input/output section and executes a series of character recognition processes including dictionary matching.

The input tablet 2 is provided with a character cantilever 7 on its upper surface for inputting characters by hand, and a plurality of vertical and horizontal squares 8 are defined in the character cannibal 7. As shown in FIG. 2, a predetermined entry form 9 is properly positioned and placed on this human-powered tablet 2, and then characters are entered from above the form 9.
The input pen 10 is brought into contact with the area corresponding to the area corresponding to the area to input characters by hand.

This input pen 10 has a structure in which, for example, when the pen tip touches the character entry canister 7 directly or through the entry form 9 and a predetermined writing pressure is applied, magnetic lines of force are generated at the pen tip. Therefore, by detecting the moving locus of these lines of magnetic force on the input tablet 2, the handwriting from the handwriting to the ryuteki can be obtained as character stroke data. Further, the input pen 1o has, for example, a ballpoint pen structure at its pen tip. Therefore, the handwriting from the handwriting to the ryuteki is written on the entry form 9 at the same time.

Further, the input tablet 2 has, at an upper position on its upper surface,
A pulley 11 is provided (FIGS. 2 and 3). This printer 11 houses a printer hammer 14 in a cover 13 formed on the input tablet 2, and has an insertion slot 12 on the side of the character-filled canister 7 of the cover 13 into which the upper side of the entry form 9 is inserted. It is configured as follows.

When the entry form 9 is inserted with its upper side into the cover 13 through the insertion opening 12 and positioned on the input tablet 2, a plurality of squares 15 are formed corresponding to the squares 8 of the character-filled canister 7. Character entry area 1 with
6 and a reflint area 17 positioned opposite to the printer hammer 14. Therefore, the input pen 1 is provided in the character entry area 16 of the already positioned entry form 16.
When characters are handwritten using 0, the characters are input to the input tablet 2 at the same time, and the printer 11 waits to print the handwritten input characters onto the print area 17 of the entry form 9.

FIG. 4 shows the contents of the memory in the control processing section,
A program area 18 in the figure stores a series of programs for character recognition processing, command processing, and printer control. The work area 19 has a coordinate storage area 20,
Stroke code area 21, square number setting area 22
, a stack area 23, the pen position coordinates of the input pen 10 on the input tablet 2 are stored in the coordinate storage area 23, and the 8-direction code (shown in FIG. 6) for character strokes is stored in the stroke code area 21. In the square number setting area 22, the square numbers of the tablet 2 related to character input are respectively stored. In terms of character storage, the IJA 24 stores the recognition results for all input characters, and the dictionary area 25 stores standard data for dictionary checking regarding alphanumeric characters and kana characters.

5A and 5B show the operational processing flow of the control processing section. First, in step 26, a stack pointer is initialized for the memory of the control processing section, and the contents of the work area 19 are cleared. Next, in step 27, the input to the entry form 9 is initialized, and in step 28, the input to the entry form 9 is initialized.
Initialize each character input. In the next step 29, it is determined whether or not a key requesting tablet input has been pressed on the keyboard 6 of the computer main body 1. When the determination in step 29 is 'YES', the process proceeds to the next step 30. , insert the input pen 1 into any square 15 of the entry form 9 positioned on the input tablet 2.
When a character is entered by touching 0, the number of square 8 of the input tablet 2 corresponding to that square 15 is set in the square number setting area 22 in the work area 19, and the pen position within that square 8 is set. The coordinates of are in the coordinate storage area 20
(step 30).

In this way, when the pen tip moves to another square or when the predetermined time (for example, about 2 seconds) set for inputting one character has elapsed, step 31 or step 32 returns "YE5".
It is then determined that the input of one character has been completed, and the process moves to the next step 34, where character recognition processing is executed.

FIG. 5B shows the subroutine of step 34.
First, in step 35, the conversion process of the 8-way code is executed for each stroke of the input character, followed by step 36.
The direction code converted in is stored in the stroke code area 21 of the work area 19.

FIG. 7 shows an example of code conversion for the kana character "a", and the symbols ■■■ in the figure indicate codes for each direction of the character stroke. In this code conversion, for example, as shown in FIG. 8, when adjacent direction codes (■■ in the illustrated example) are consecutive,
A rounding process is performed to make the direction code (■) associated with a long stroke a representative stroke code.

Next, in step 37, dictionary matching processing is performed with reference to the dictionary area 25 for the input character pattern consisting of the direction code. If the input character pattern matches one of the standard patterns, the judgment in step 38 becomes ``YES'' and the process proceeds to step 39, where the recognition result is output to the CRT 5 and displayed on the CRT screen. The characters to be recognized are displayed, and in step 40, the recognized characters are stored in the area of the corresponding square number in the character storage area 24.Meanwhile, in step 3
If the determination in step 8 is "No", proceed to step 41, and
An error message indicating unrecognizability is displayed on RT5.

When the character recognition process in step 34 is completed, the process returns to step 28, where the work area 19 is initialized for inputting one character, and waits for the next character input. Therefore,
If characters are written in each square 15 on the entry form 9 one after another with the input pen 10, the character recognition process for each input character is sequentially performed through the routes of steps 29, 30, 31 (or 32), and 34 and 28. executed.

If a command other than the tablet input start command is input by key operation on the computer main unit 1,
Either step 29 is "NO", step 43 is "YES" and the process advances to step 44, or step 33 is "YES".
YES”, and after the process of step 42, the process goes to step 4.
Proceed to step 4.

This step 33 assumes that another command is input before the timeout of step 32, and in this case, character recognition processing is performed for the last character in step 42 of the same subroutine as step 34. After
Proceed to step 44.

Steps 44 and 45 are for determining the type of command. If the input command is, for example, a character correction command, step 44 determines "Is this a tablet input end command?" and step 45 determines "Is this a tablet input end command?""Is this a cancel command?" are both determined as "NO", and in step 46, the process corresponding to the command (in this case, character correction process) is executed. This character correction process is performed through the same flow as steps 30 to 33 and step 34 described above, and by executing this character correction, the corresponding square number area stored in the character storage area 24 is The character data will be replaced with the corrected character data.

In this way, when the tablet input end command is input by key operation, the determination in step 44 becomes "YES" and the process proceeds to step 47, where the character content stored in the character storage area 24 is transferred to the entry form 9 by the printer 11. The character data is printed in the print area 17, and furthermore, in step 48, the character data in the character storage area 24 is output to an external information processing device.

In step 45, if the command input is a cancel command, the process returns to step 27 and the input data is erased; on the other hand, if the command input is a command other than the cancel command, the process proceeds to step 46, where the command input is canceled. Corresponding command processing is executed.

[Brief explanation of drawings]

Fig. 1 is a perspective view showing the overall configuration of a handwritten character recognition device according to the present invention, Fig. 2 is a perspective view of a human-powered tablet with a printer on which entry forms are set, Fig. 3 is a cross-sectional view thereof, and Fig. 4 is the device. 5 is a flow chart showing the operation of the device, FIG. 6 is a diagram for explaining the 8-direction code, FIG. 7 is a diagram showing the direction code for the kana character "A", and FIG. is a diagram for explaining rounding processing in code conversion. 1...Computer body 2...Input tablet 7...Character entry pad 9...Entry form 10...Input pen 11... ...
Printer 16...Character entry area 17...
...Print area patent Applicant Nippon Tsushin Construction Co., Ltd. Patent applicant Tateishi Electric Co., Ltd. ＼ To n 〜〜〜 → 6 Figure 1 Calorie code Me 1 Sarachizuki Ribo → 7 Figure Power direction coat・i ;ta moon sear 7F5 figure A fork (! phantom θ Kawaguchi

Claims (1)

[Scope of Claims] An input tablet having an area for inputting characters by hand; an input pen for inputting characters by hand by bringing a pen tip into contact with an entry form placed on the input tablet; and a recognition processing section that takes in handwritten character input on a paper from an input tablet and executes character recognition processing, and the input tablet receives the character recognition results from the recognition processing section in areas other than the character entry area of the entry form. A handwritten character recognition device equipped with a printer that prints into the area.