JPS61120286A - Recognizer for handwritten character - Google Patents

Abstract

PURPOSE:To decrease the number of characters to be read erroneously, and to improve the accuracy for recognition of characters by detecting the start of a character input operation to a character type designating area through a control processing part, and carrying out with priority the recognition of characters of specific types. CONSTITUTION:An input tablet 2 contains a character input area 8 for handwriting input of characters with a contact of an input pen 7. The area 8 includes character writing frames 9 formed with plural vertical and horizontal lines. A character type disignating area is formed in response to each frame 9 for disignation of input of a character of a specific type. For input of handwritten characters, the pen point of the pen 7 is first brought into contact with the character designating area in the frame 9. Thus a specific character type is collated with a dictionary with priority.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field of the Invention The present invention relates to a handwritten character recognition device that receives handwritten characters input to an input tablet and performs dictionary matching processing to recognize the input characters.

<Summary of the Invention> The handwritten character recognition device of the present invention forms a character type specification area for specifying a specific type of character input (for example, kana characters) corresponding to a plurality of character entry frames set on an input tablet. , the control processing unit detects the start of a character input operation in this character type specification area and prioritizes character recognition for the specified character type. Reduce misreading,
This not only improves character recognition accuracy, but also speeds up corrections to misreadings.

<Background of the Invention> Conventionally, this type of character recognition device recognizes kana characters, English characters,
When handwriting a mixture of numbers, etc., it is easy to confuse characters, especially between different character types, such as the kana character "ri" and the number "7", the kana character "me" and the English character rXJ, etc. , there were problems such as these character inputs being erroneously recognized.

<Object of the Invention> In view of the above-mentioned circumstances, the present invention aims to provide a handwritten character recognition device of a new type that prevents confusion and misrecognition of characters between different character types, thereby improving character recognition accuracy. shall be.

<Structure 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 entry frames are set in an area for inputting characters by hand, and a pen tip that is brought into contact with the input tablet. A handwritten character recognition device includes an input pen for inputting characters by hand, and a control processing section for receiving handwritten characters input from an input tablet and executing character recognition processing. A character type designation area for specifying a specific type of character input is formed corresponding to the character entry frame, and the control processing section is configured to determine the start of a character input operation to the character type designation area and select the character type of the specific type. A method is provided to prioritize character recognition for .

According to the present invention, characters of a specific type that are started by touching the pen tip to the character type designated area are less likely to be confused with other types of characters and misrecognized, and character recognition accuracy is significantly improved. do. Therefore, if you write a specific type of character in the character entry frame and it is misread, if you start writing that character from the character type specification area and input it again, the character will be recognized correctly this time, and the misreading will be avoided. It has a remarkable effect of achieving the purpose of the invention, such as being able to take prompt measures against the problem.

<Description of Embodiments> FIG. 1 shows the overall configuration of a handwritten character recognition device according to the present invention.

The illustrated apparatus includes a personal computer main body 1 (hereinafter simply referred to as "computer main body"), an input tablet notebook 2
, printer 3, floppy disk device fil 4, etc., and these components are electrically connected by cord lines. Computer main unit 1, CRT (Cathode)
Ray Tube) 5 and keyboard 6,
The keyboard 6 is provided with a plurality of function keys for manually entering various commands. Note that the command input section is not limited to the keyboard, but may also be provided on the input tablet 2 and operated using a pen. Inside the computer main body 1, there is a CPU (Central CPU).
PrOcessing Unit), RAM
(Random Access Memory), R
A control processing unit including a memory such as OM (Read Only Memory) is incorporated, and this control processing unit controls input/output operations in each input/output unit, and executes a series of character recognition processes including dictionary matching.

The input tablet 2 is provided with a character input area 8 on its upper surface for inputting characters by hand by bringing an input pen 7 into contact with the character input area 8. Furthermore, this character input area 8 has a character entry frame 9 with a plurality of vertical and horizontal squares. is formed. Each character entry frame 9 is for entering one character (generally including alphabetic letters, numbers, kana characters, kanji, symbols, etc.), and each character entry frame 9 corresponds to a specific type of character. The character type specification area (
(shown by broken lines in FIGS. 2 to 5) is formed.

FIG. 2 shows an example of forming this character type designation area. In (3), a character type designation area 10 is formed on the upper left side of the character entry frame 9, respectively. Normally, one character type designation area 10 is provided for each character entry frame 9, but the invention is not limited to this, and as shown in FIG.
For example, a character type specification area 10A for specifying kana characters and a character type specification area 10B for specifying numbers for each character entry frame 9.
A plurality of character type designation areas may be set as necessary, such as forming a character type designation area.

In this way, normal handwritten character input is performed by first bringing the nib of the input pen 7 into contact with the character entry frame 9 (in the figure, P is the first In this case, the recognition process first performs dictionary matching for alphabetic characters and numbers before dictionary matching for kana characters.

On the other hand, the character type specification area 10 is used when the character input based on the above procedure is misread.
As shown in the figure, the pen tip of the input pen 7 is first brought into contact with the character type designation area lo (in the figure, P indicates the first writing point). According to this re-input operation procedure, dictionary collation of kana characters is first performed with priority, thereby making it possible to avoid misreading of kana character input.

The above character entry frame 9 may be formed continuously as shown in Fig. 5+11 (21), or may be formed continuously as shown in Fig. 5 (
They may be made independent one by one as in 3).

Furthermore, the character type specification area 1o may be added outside the character entry frame 9 as shown in FIG.
It may also be assigned inside the character entry frame 9, such as +31.

The writing surface 7 has a structure in which, for example, when the pen tip touches the input tablet 2 and a predetermined writing pressure is applied, magnetic lines of force are generated at the pen tip, and therefore, the movement of these magnetic force lines on the input tablet 2 By detecting the locus, each stroke of an input character from writing to writing is obtained as data. In this embodiment, a special key is provided on the keyboard 6 of the computer main body 1 for inputting an input end command, and each time this key is operated, all strokes in each square are included, and character recognition processing is completed. It is about to start.

In the above embodiment, when inputting characters into the character type designation area 10, dictionary matching of kana characters is given priority;
The dictionary check for symbols, etc.) may be configured to be performed subsequently, or a method may be adopted in which it is not performed at all.

Furthermore, the character type designation area 10 is not limited to being used for re-inputting in the case of misreading as described above, but there are no particular restrictions on how this device can be used, such as using this area spontaneously when inputting kana characters.

FIG. 6 shows the contents of the memory in the control processing section,
A program area 11 in the figure stores a series of programs for character recognition processing.

The work area 12 has a coordinate storage area 13, a stroke code area 14, a square number setting area 15, a mode storage area 16, etc. In the coordinate storage area 13, the pen position coordinates of the input pen 7 on the input tablet 2 are stored. The code area 14 contains the 8-way code (shown in Figure 9) for character strokes, the square number setting area 15 contains the number of the square (character entry frame) for character input, and the mode storage area 16 contains the character type specification. Mode designation data indicating whether or not characters are to be input into the area 10 is stored respectively. Further, the character storage area 17 stores the recognition results for all input characters, and the dictionary areas 18 and 19 store standard data for dictionary checking of alphanumeric characters and kana characters.

FIG. 7 shows the operation processing flow of the control processing section. First, in step 20, a stack pointer is initialized for the memory of the control processing section, and the contents of the work area 12 are cleared. Next, step 21 determines whether there is any key input from the keyboard 6 of computer body 1, and when a key requesting tablet input is pressed, it is determined in step 21 whether there is a command input. and,
The judgment of “Start input” in step 22 is both “YES”
Then, proceed to step 23. and input tablet 2
When the input pen 7 comes in contact with the character input area 8 and a predetermined writing pressure is applied, the determination of ``Drawing'' in step 23 becomes ``YES'', and the mode designation data is stored in the mode memory based on the area where the pen tip is located. At the same time, the coordinates of the pen position within the character entry frame 9 or the character type designation area 10 are written into the coordinate storage area 13 (step 2).
4, 25).

FIG. 8 shows the detailed flow of step 24. First, in step 50, it is determined whether the above-mentioned "writing" is the first writing on the character entry frame 9 or the character designation area 10. .

When the determination in step 50 is "YES", in the next step 51, the pen tip position is in the character type specification area 10.
It is determined whether or not it is a character type designation input based on whether or not it is within the range. If the first writing position is within the character type specification area 10, the determination of “Character type specification input?” in step 51 is “
YES'', the process advances to step 52, and the kana mode designation data is stored in the mode storage area 16.

On the other hand, when the determination in step 51 is "No", step 5
3, the general mode data is stored in the mode storage area 16, and then, in step 54, the square number corresponding to the pen tip is set in the square number setting area 15.

In this way, when the pen tip moves and a character stroke is drawn within the same square, the determination of ``stroke separation'' in step 26 is NO.
, the determination of “Is it within the same square” in the following step 27 is “YES”.
S'', and the constituent coordinates of the character stroke are sequentially stored in the coordinate storage area 13. In this case, if the character stroke protrudes outside the frame, step 27 is changed to "N".
"O" and an error sound is generated in step 28. Then, when the pen tip is separated from the input tablet 2, the determination of "Is it a good handwriting" in step 26 or step 29 is "Y"?
ES'', and further, in step 30, it is checked whether or not the input pen touches the input tablet 2 again, that is, whether or not the next character stroke exists.

In this way, when the input of characters is completed and the key instructing the end of input is pressed on the keyboard 6, the determination of "command input" is made in step 31, and the "command input" is determined in step 32.
"Is the input completed?" are both "YES", and in the next step 33, the conversion process of the 8-direction code is executed for each stroke of the input character, and then in the following step 3.
The direction code converted in step 4 is stored in the stroke code area 14 of the work area 12.

FIG. 10 shows an example of code conversion for the kana character "A", and in the figure, ■■■ indicates each direction code of the character string ``o''. In addition to this code conversion, for example, as shown in FIG. 11, 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 35, it is determined whether the mode designation data already posted in the mode storage area 16 is kana mode designation data, and the determination in step 35 is "YES".
S'', in the next step 36, the input character pattern consisting of the direction code is first subjected to dictionary matching processing that refers to the kana character dictionary area 19.Then, if the input character pattern is one of the standard kana characters, If it matches the pattern, the determination in step 37 as to whether it matches is YES and the process proceeds to step 40, where the recognition result is output to the CRT 5 and the characters to be recognized are displayed on the CRT screen.

On the other hand, when the determination in step 37 is "No", in the next step 38, a dictionary matching process is performed with reference to the alphanumeric dictionary area 18 for the input character pattern. If the input character pattern matches any alphanumeric standard pattern, the determination in step 39 as to whether it matches is ``YES'' and the process proceeds to step 40, where the corresponding character is displayed on the CRT 5. If the judgment in step 39 is
If "No", the process advances to step 41, and an error message indicating unrecognizability is displayed on the CRT 5.

Next, in step 35, when it is determined that the mode designation data is not kana mode designation data, the determination in step 35 becomes "NO", in which case alphanumeric dictionary collation processing is first performed in step 42, and then step At step 44, dictionary matching processing for kana characters is performed. As a result, if there is a pattern match in any of the dictionary matches, the determination of "match match" in step 43 or step 45 becomes "YES" and the process proceeds to step 40, where the corresponding character is displayed on CλT5. In this case, when the characters displayed on the CRT 5 are erroneously recognized as input Kana characters, in order to correct this, the character type specification area 10
First, the pen tip of the input pen 7 is brought into contact with the input pen 7 to rewrite the misidentified kana characters. Meanwhile step 43
When the determinations in step 45 are both "No", the process proceeds to step 41, where an error is displayed on the CRT 5 indicating that the recognition is not possible.

In addition, in steps 22 and 32, if the command input is other than instructing to start or end tablet input, the respective determinations are "No" and the process proceeds to step 46, where the command processing corresponding to the command input is executed. be done.

[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, FIGS. 2 and 3 are diagrams for explaining an example of forming a character type designation area, and FIG. 4 is an illustration of an example of character entry. 5 is a diagram for explaining an example of forming a character entry frame and a character type designation area. FIG. 6 is a diagram showing a memory map of the device. FIGS. 7 and 8 are flowcharts showing the operation of the device. , FIG. 9 is a diagram for explaining the 8-way code, FIG. 10 is a diagram showing the direction code for the kana character "a", and FIG. 11 is a diagram for explaining the rounding process in code conversion. 1... Computer body 2... Input tablet, 7... Input pen 8... Character input area 9...
・Character entry frame 10...Character type designated area Patent applicant Tateishi Electric Co., Ltd. ++27 Old (2) (3) Mark J2 (2J Take 9 fig.

Claims (5)

[Claims] (1) An input tablet in which multiple character entry frames are set within an area for inputting characters by hand; an input pen for inputting characters by hand by touching the pen tip with the input tablet; and handwritten character input. and a control processing unit that executes character recognition processing by fitting into the input tablet, and the input tablet has a character type designation area for designating a specific type of character input corresponding to each character entry frame. The handwritten character recognition device further comprises: the control processing unit having means for detecting the start of a character input operation in the character type designation area and giving priority to character recognition for the specific character type. (2) The handwritten character recognition device according to claim 1, wherein the character type designation area is added outside the character entry frame. (3) The handwritten character recognition device according to claim 1, wherein the character type designation area is partitioned and provided inside a character entry frame. (4) The handwritten character recognition device according to any one of claims 1 to 3, wherein the character type specification area is an area for specifying kana character input. (5) The handwritten character recognition device according to claim 1, wherein a plurality of types of character type designation areas are formed on the input tablet in correspondence with each character entry frame.