JPS61169975A - Recognizing device of hand-written character - Google Patents

Abstract

PURPOSE:To simplify character input work and to improve working efficiency by outputting a recognizing character code and information specifying its character frame as a pair of data at the recognizing of the character hand-written in the character frame. CONSTITUTION:A hand-written character recognizing device 1 is constituted of an input tablet 2 and a recognition processing part 3 and connected to a host side device 4 through a code wire. The format of a slip set up on a tablet 2 is stored in a slip format information storing area 61 of a memory in the device 4. At the recognition of an input character,the device 4 is initialized at first, the contents of the area 61 are read out, the format of the slip is displayed on the screen of a CRT5, and then the reception of information can be permitted. In this stage, a recognized character code or an error code and its character frame No. information are sent from the device 1 as a pair of data. These data are received by the device 4 and stored in a memory. Consequently, the character frame corresponding to the character input can be grasped, the character input work can be simplified and the working efficiency can be improved.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a handwritten character recognition device that captures characters input by hand on an input tablet notebook and performs dictionary matching processing to recognize human characters.

<Summary of the Invention> The handwritten character recognition device of the present invention has a plurality of character frames for handwriting characters in the character input area of an input tablet, and when recognizing characters handwritten in the character frames. By outputting the recognized character code and information specifying the character frame as a pair, it is possible to grasp the character frame involved in inputting characters. It is also possible to perform manual input and correction of characters at will, thereby making character input work easier and improving work efficiency.

<Background of the Invention> Conventionally, in this type of character recognition device, a character entry person manually inputs characters one character at a time into an input tablet, and when the text is separated by punctuation marks etc.
It is normal to check the recognition results of input characters by looking at the display on the Ray Tube. In this case, when any of the input characters is misread, the person filling in the characters moves the cursor on the CRT back to the position of the misread character by operating the shift key on the keyboard or input tablet while looking at the CRT. , it is necessary to correct and input new characters. Furthermore, when this correction is completed, the person writing the characters must similarly operate the shift key to advance the cursor on the CRT to its original position and restart the next character input operation.

Therefore, when correcting characters, etc., it is necessary to move the cursor twice, which not only reduces the efficiency of character input work, but also greatly exhausts the nerves of the person writing the characters when positioning the cursor stop position. There were many shortcomings.

<Purpose of the Invention> The present invention provides a device in which a plurality of character frames for handwriting input of characters are set in the character input area of an input tablet, and operations such as character correction can be performed without any cursor movement operation. It is an object of the present invention to provide a novel handwritten character recognition device that can be implemented.

<Configuration and Effects of the Invention> In order to achieve the above object, the present invention provides an input tablet in which a plurality of character frames for inputting characters by hand in a character input area are set, and a pen tip is brought into contact with the input tablet. an input pen for handwriting characters, and a recognition processing section that takes in handwritten human characters from the character input area of the input tablet, executes character recognition, and outputs a recognized character code, and the control processing section: The present invention is configured to specify and obtain a character frame related to character input, and output the character frame specifying information as a pair with the recognized character code.

According to the device of the present invention, the character frame required for character correction or character input can be grasped using the character frame identification information, so the conventional cursor movement operation is no longer required when character correction etc., and this makes character input work easier. Efficiency can be greatly improved. Furthermore, since determining the cursor stop position does not involve staring at the CRT screen and concentrating one's nerves, the character writer does not have to exhaust his or her nerves, achieving the remarkable effect of achieving the object of the invention.

<Description of Embodiments> FIG. 2 shows an overall schematic configuration of a character processing system incorporating the handwritten character recognition device 1 of the present invention.

The illustrated handwritten character recognition device 1 includes an input tablet 2
and a recognition processing section-3, and each of these components is electrically connected to the host side device 4 by a cord wire.

Here, when the handwritten character recognition device 1 is connected to a dedicated terminal for a bank, for example, the host-side device 4 refers to the dedicated terminal, that is, the "immediately superior" terminal, and the host-side device 4 in the illustrated example is In this case, CRT (Cathode Ray
It consists of a host computer main body 7 equipped with a tube 5 and a keyboard 6, and a printer 8 and a floppy disk device 9 connected thereto.

The recognition processing unit 3 includes a CPU (CentralPr
In addition to RAM (Ra
ndam AccessMemory), ROM (
The recognition processing unit 3 controls input processing of handwritten characters, a series of character recognition processing including dictionary matching, and output processing of recognition results. Ru.

The input tablet 2 has a character input area 10, a touch key area 11, and an error lamp 12 on the tablet surface.
Each of the areas 10 and 11 is formed so as to be able to sense the touch of the pen tip of a dedicated input pen 13. The touch key area 11 is arranged with a plurality of function keys that are selectively operated with the tip of the input pen 13 and used to input various commands manually. Character input area 10
is for inputting desired characters by hand by bringing the pen tip of the input pen 13 into contact with the tablet surface and moving the pen tip, and a matrix circuit is provided inside the tablet surface corresponding to the character input area 10. A coordinate detection circuit is incorporated to detect the X and Y coordinates of the pen tip in contact with the character input area 10.

In the character input area 10, square-shaped character frames 14 are formed in series over multiple rows and multiple stages. Within each of these character frames 14, a single character can be input by hand, and after positioning a form on which the same character frame is printed on the tablet surface, the input pen 13 is inserted into the character frame. When characters are handwritten one after another using the , each character is recognized in sequence, and the recognized character code and information specifying the character frame 14 for character input are output as a pair to the host device 4. In the illustrated example, this character frame identification information is a series of character frame numbers set corresponding to each character frame 14 (indicated by numbers 1, 2, 3, etc. in FIG. 5 (1)). However, the information is not limited to this, and may be information specifying the column and column of the character frame 14, for example, as shown in FIG. 5(2).

Furthermore, as shown in FIG. 5(3), when the character frame 14 is set in the character input area 10 in block units, the character frame specifying information includes the block number and the number of characters in each block. Character frame numbers (numbers 1, 2, 3, etc. in the diagram)
..., 10).

The input pen 13 has a structure in which, for example, magnetic lines of force are generated at the pen tip when the pen tip touches the input tablet 2 and a predetermined writing pressure is applied. By detecting each stroke of the input character, from writing to writing, each stroke of the input character is obtained as data.

FIG. 3 shows the contents of the memory in the recognition processing section 3, and a program area 15 in the figure stores a series of programs for character recognition processing.

The work area 16 includes a coordinate counter 17, a coordinate storage area 18, a substroke counter 19, a direction code storage area 20, a stroke counter 21, a stroke code storage area 22, a character frame number storage area 23, a flag setting area 24, etc. The coordinate counter 17 stores the number of coordinate points that the input pen 13 has passed over the person's stroke/throat 2, and the coordinate storage area 18 stores the coordinates of each pen position of the input pen 13 as it moves on the input tablet 2. The number of 8-direction codes (shown in FIG. 4) included in the character stroke is stored in the direction code storage area 20, and the 8-direction code of each sub-stroke (described later) that constitutes the character stroke is stored in the stroke counter 21. indicates the number of character strokes that make up a character, the stroke code storage area 22 indicates a code specifying the type of stroke input, and the flag setting area 24 indicates whether or not the character recognition result has been output to the host device 4. The flags (hereinafter referred to as "character output flags") are selected respectively. Furthermore, the stroke recognition dictionary area 25 contains standard data for collation of all the strokes that make up a character, and 1
Standard collation data for characters composed of strokes (for example, the number "2") is stored in the character recognition dictionary area 26, and characters composed of two or more strokes (for example, the number "5") are stored in the character recognition dictionary area 26. ) are stored respectively.

FIG. 1 shows the control operation of the recognition processing section 3.
First, in step 31, the recognition processing section 3 is initialized. In the next step 32, it is determined whether a command has been input from the touch key area 11 of the human-powered tablet 2. If the determination is "YES", the command code is transmitted to the host device 4 (step 33). ). If the command input is an input start command that requests tablet input, the next step 34 is "YES".
In step 35, the contents of the work area 16 of the memory are cleared.

In this way, when the pen tip of the human-powered pen 13 comes into contact with the character input area 10 of the input tablet 20 and a predetermined writing pressure is applied, the determination in step 36 "Is there command input?" is NO, and the determination in step 37 "Writing?" ” is determined as “YES” and the process proceeds to the next step 38.

In this step 38, after the coordinates of the writing point are detected by the coordinate detection circuit, a character frame number calculation operation is executed using the coordinate data, and the calculated character frame number data is transferred to the work area 16. It is stored in the character frame number storage area 23 of . Subsequently, in step 39, the contents of the coordinate counter 17 and stroke counter 21 in the work area 16 are cleared, and then in the next step 40, the subroutine for stroke input processing shown in FIG. 6 is called and executed. Ru.

Now, suppose that the first stroke of a character (for example, the number "4") consisting of the first and second character strokes is written in the character frame 14. First, step 7 in FIG.
1, it is determined whether the writing point in step 37 is within the character frame 14, and if the determination is "YES", the contents of the coordinate counter 17 are incremented by 1, and the coordinate detection circuit in the input tablet 2 The detected pen position coordinates are stored at a predetermined location in the coordinate storage area 18 corresponding to the contents of the coordinate counter 17 (steps 72 and 73).

When the pen tip moves and a character stroke is drawn within the same character frame 14, the determination in step 74 is NO', and the process returns to step 71, where the character stroke is drawn within the same character frame 14. ”
The determination becomes "YES", and as above, the coordinate counter 1
7 is added and the pen position coordinates are stored (steps 72, 73).

This processing operation is repeatedly executed every time the pen tip position passes a coordinate point, and when the pen tip leaves the input tablet 2 upon completion of the first stroke input, step 74 is performed.
The determination becomes "YES" and the process advances to step 75.

In step 75, a point where the coordinates of the stroke input abruptly change (for example, a point where it changes by 90 degrees) is detected as a change point, and the character stroke is divided into a plurality of sub-strokes at each change point.

FIGS. 7(1) and (2) show examples of substroke division for characters r2J r5J. In the figures, black dots 27 are writing points, black dots 28 are writing separation points, white points 29 are change points,
Straight lines 30 each indicate a substroke.

Each of the above-mentioned sub-strokes is converted into a direction code shown by numbers 1 to 8 in FIG. Then, in the following step 77, the direction code of each substroke is stored in the direction code storage area 20.

FIG. 8 shows an example of direction code conversion for the first stroke of the number "5", and in the figure, ■■■■■■■■ indicates the direction code of each sub-stroke.

In this code conversion, when adjacent direction codes are continuous (in the illustrated example, the portions ``■'' to ``■''), a so-called rounding process is performed to make the direction code ``■■■'' associated with a long stroke representative.

Figure 9 shows the direction code string of the number "5" after rounding, where the first stroke has a direction code of ■■■■, and the second stroke has a direction code of ■. .

Next, returning to the example of number "4", the first stroke input is checked against the contents of the stroke recognition dictionary area 25 in step 78, and a direction code string matching the direction code string of the first stroke input is determined. It is checked whether there is standard data with When there is matching data, the determination of "Is it a match?" in step 79 becomes "YES", and after incrementing the stroke counter 21 by 1 in step 80, in the next step 81, the character stroke related to the standard data is calculated. A code for specifying a stroke (hereinafter referred to as a “stroke code”) is stored in the stroke code storage area 22 according to the contents of the stroke counter 21
Set to the specified location and return to the main routine. On the other hand, if the standard data related to matching does not exist in the dictionary area 25, the result in step 79 is NO'', and in this case, it is considered a human error and the process returns to the main routine.

If the pen tip protrudes outside the character frame 14 during the process of drawing a character stroke, the determination of "Are they in the same character frame?" in step 71 will be "NO" and an error sound will be generated in step 82. , in this case, the next step 83 is ``''YES'' due to the writing of the pen tip from the input tablet 2.
” and returns to the main routine as an input error.

In this way, if the above step 79.83 returns as an input error, the determination in step 41 in FIG. 1 is "Y".
ES", an error sound is generated in step 7, step 42.43, and the error lamp 12 lights up.

On the other hand, if the input is completed and the main routine returns, the determination in step 41 is "NO", and then in step 44, the stroke input so far (in the example with the number "4", only the first stroke It is checked whether standard data matching the characters consisting of (input completed) exists in the dictionary area 25,26. Number "
4", the first stroke alone does not constitute a character, so the determination in step 45 "Does it match?" is NO."
Then, in the next step 46, character frame number storage area 2 is
After the character frame number, which has already been set to 3, and an error code indicating mismatch are sent as a pair to the host device 4, the contents of the flag setting area 24 are cleared in step 47.

If the input character consists of only the first stroke (for example, the number "2"), the determination in step 45 becomes "YES", and the character frame number and its recognized character code are determined in step 48. After the pair is sent to the host device 4, a character output flag "1" is set in the flag setting area 24 in step 49.

The next step 50.51 is to wait for a certain period of time (e.g. CPU
It is checked whether or not there is writing with the input pen 13 in the character input area 10 within a period of time (about 2 to 5 seconds is set in the timer). In the case of the example of the number "4", since the second stroke is continuously written in the same character frame 14, the determination of "Is the time over?" in step 50 is "NO",
The determination in step 51 as to "Is it drawn on?" is YES', and the determination as to "in the same character frame?" in step 52 as "YES", and the process proceeds to step 40, where the second stroke is shown in FIG. The subroutine is executed.

If the second stroke is not input within the set time in the standby state of step 50.51, the determination of "Is the time over?"
” and proceeds to step 53, where the host side device 4
It is checked whether the recognized character code has been output to the flag setting area 24, that is, whether the character output flag rlJ has been set in the flag setting area 24. In this case, the determination in step 53 is “
"No", so in steps 54 and 55 an error sound is generated and the error lamp 12 is lit to inform the character entry person that the input of the first stroke is invalid, and then the process waits for the next character entry. (Step 37).

Note that if input of all strokes constituting a character has been completed, steps 54 and 55 are not executed and the input standby state of step 37 is entered.

Thus, in the input process for the second stroke of the number "4", the determination of "Verification match?" in step 45 is "Y".
ES'', and at this stage the recognized character code is sent to the host device 4 together with the character frame number (step 48).

If you want to correct the input characters that have already been recognized during the character input process, place the tip of the human pen 13 in the character frame 14 where the character to be corrected is written, and write a new character over it. It can be handwritten.

In this case, after a sufficient amount of time has passed after character input is completed, both step 50 and step 53 are
YES", and furthermore, the determination of "Is it drawn?" in step 37 becomes "YES", and in the next step 38, the number of the character frame 14 is calculated and stored, and in step 39, the coordinate counter 17 and the stroke counter are After the clear processing of 21 is sequentially executed, a correction process based on the subroutine of FIG. 6 is executed at step 40.

On the other hand, if the process moves to character correction immediately after completing one character, step 50 is "NO" and step 51 is YES.
", Step 52 becomes "NO" and the process proceeds to Step 56. In this step 56, it is checked whether or not the character output flag "1" is set in the flag setting area 24. In this case, the determination is is "YES", and the process advances to step 38, where character correction processing is executed.

Further, in the case where the character correction processing is started immediately after the first stroke of the number "4" in the above embodiment is input, step 50 is ``NO'', step 51 is ``YES'',
Step 56 is 'NO' and steps 57.58
After generating an error sound and lighting up the error lamp 12 to notify the character entry person that the human power of the first stroke is invalid, the process advances to step 38 and character correction processing is executed.

Note that the correction input processing based on the subroutine of FIG. 6 is similar to the normal character manual processing described above, and the explanation thereof will be omitted here.

FIG. 10 shows the contents of the memory in the host device 4. As shown in FIG. In the figure, a program area 60 stores a program for receiving and processing commands and character recognition data sent from the handwritten character recognition bag fl, as well as a series of programs related to system processing. The area 61 stores the format of the form placed on the input tablet 2 (the number of character frames, position coordinates of the character frames, etc.).

The work area 62 has a received character storage area 63, a character frame number storage area 64, etc.
The recognized character code sent from the handwritten character recognition device 1 is set in the character frame number storage area 64, and the character frame number associated with the received character is set in the character frame number storage area 64. The character storage area 65 is for storing received characters one after another at addresses corresponding to the value of the character frame number.

FIG. 11 shows the control operation of the host side device 4. First, initialization is performed in step 91, and the contents of the work area 62 of the memory are cleared.

Next, the contents of the form form information storage area 61 are read out and the form format is displayed on the screen of the C-RT 5, after which information reception from the terminal (including the handwritten character recognition device l) is permitted. (steps 92 and 93).

At this stage, if the handwritten character recognition device 1 transmits a recognized character code or error code and its character frame number information as a pair, and this is received by the host device 4, the determination in step 94 "Received?" “YES”, step 9
5, "Is the received content a command?" is determined as "NO", and the received information is stored in the received character storage area 63 and the character frame number storage area 64, respectively.

If a command is received under the reception acceptance state, step 95 becomes "YES'" and step 96
After inhibiting reception from the terminal in step 97, the process corresponding to that command is executed in step 97.

The next step 98 is to determine whether or not the received information is an error code. In the embodiment with the number "4", the error code is detected on the host side at the end of the first stroke recognition. It has already been mentioned that it is sent to device 4 (step 46 in FIG. 1). Therefore, in this case, the determination at step 98 is "YES", and the display position of the CRT 5 corresponding to the received character frame number is cleared and cleared, and a display indicating the human error is displayed at the appropriate place on the CRT 5 ( Steps 101, 102). Furthermore, an error sound is emitted in step 103, and the next step 10
4, the address of the character storage area 65 corresponding to the character frame number is cleared.

On the other hand, at the end of the second stroke recognition of the number "4", the recognized character code is sent to the host device 4 together with the character frame number (step 48 in FIG. 1), so in this case, step 98 The determination is "NO", and in step 99, the CRT 5 corresponding to the received character frame number is
The character corresponding to the recognized character code (in this case, the number "4") is displayed at the display position of step 100.
, the received character code is stored in the address of the character storage area 65 corresponding to the character frame number.

The above-mentioned series of processes is repeatedly executed every time there is a reception, and as a result, the received characters are displayed one after another on the CRT 5, and the received information is stored in the character storage area 65 (step 99, 100).

Also, the above applies not only when inputting new characters, but also when inputting new characters.
The same applies to character correction; in this case as well, the corrected character is sent to the host device 4 together with the character frame number involved in the correction, so that the corrected character is correctly displayed at the character correction position on the CRT 5.

[Brief explanation of drawings]

FIG. 1 is a flowchart showing the control operation of the handwritten character recognition device according to the present invention, FIG. 2 is an external view showing the overall configuration of the handwritten character recognition device, FIG. 3 is a diagram showing the memory map of the handwritten character recognition device, and FIG. 4 is a diagram for explaining the 8-direction code, FIG. 1) (21 is a diagram showing an example of substroke division for characters r2J r5J, Figure 8 is a diagram showing an example of direction code conversion for the first stroke of character "5", and Figure 9 is a diagram showing an example of the direction code conversion of the character "5" after rounding processing. ”
10 is a diagram showing a memory map of the host side device, and FIG. 11 is a flowchart showing the control operation of the host side device. 1...Handwritten character recognition device 2...Input tablet 3...Recognition processing unit 10...Character input area 13...Input pen 14 ...Character frame patent applicant: Nippon Tsushin Construction Co., Ltd. Patent applicant: Tateishi Electric Co., Ltd.
；Kashiichi・] Bustling T-i diagram "y+3E Ntyi
tt, IM mosquito engineering sail mesorimatsu 2゜1-17 41()”J y, 2) a-1itI71/!u?
v7”21. Enter Kata 7shi・Vto3 Book 1kichi%, J Konoma Hirop, 0.*Primary input area

Claims (5)

[Claims] (1) An input tablet with a plurality of character frames set in the character input area for inputting characters by hand; an input pen for handwriting characters by touching the pen tip to the input tablet; and a recognition processing unit that takes in handwritten input characters from a character input area, executes character recognition, and outputs a recognized character code, and the recognition processing unit specifies and obtains a character frame related to character input, and converts the character frame into the character frame. A handwritten character recognition device configured to output specific information as a pair with the recognized character code. (2) On the input tablet, a plurality of character frames are set in multiple rows, and the recognition processing unit is configured to generate a series of characters set corresponding to each character frame for each character frame for character input. 2. A handwritten character recognition device according to claim 1, wherein a frame number is determined and the character frame number information is output as a pair with a recognized character code. (3) On the input tablet, a plurality of columns of character frames are set over multiple rows, and the recognition processing unit determines the rows and rows of the character frames related to character input, and identifies the rows and rows. A handwritten character recognition device according to claim 1, wherein the information is output in pairs with a recognized character code. (4) On the input tablet, a character frame spanning multiple columns and multiple stages is set over multiple blocks, and
The recognition processing unit determines, for each character frame involved in character input, the number of the block to which the character frame belongs and a series of character frame numbers set corresponding to each character frame for each block, and calculates these numbers. A handwritten character recognition device according to claim 1, wherein the information is output in pairs with a recognized character code. (5) The handwritten character recognition device according to any one of claims 1 to 4, wherein the recognition processing section is configured with a computer including a CPU and a memory.