KR19980085673A - Character input device and method - Google Patents

Abstract

The present invention relates to a character input apparatus and method for use in a computer, etc. Among these, the character input apparatus comprises: an input plate (1) for recognizing characters input from a user; And an input pen 9 which is a separate external input means for inputting a character into the input panel 1, wherein the input panel 1 is formed to represent both Korean and English vowels and vowels. Character input path 12; A plurality of pressure sensors (11) for detecting the trace of the input pen (9) passing while pressing on the letter input path (12); A character selection button 7 capable of converting input characters into Korean characters, English uppercase letters, and lowercase English letters so that various types of characters can be input by using the single character input path 12; And an input completion button 8 for inputting completion of character input in one independent unit, wherein the character input path 12 comprises: a square path 2; A circular path (3) inscribed in the square path (2); A diagonal path 4 connecting the opposite edges of the square path 2 with each other; A cross path 5 connecting the midpoints of opposite sides of the square path 2 to each other; And a straight path 6 formed at a predetermined distance from the path 2 at an upper portion of the square path 2, and is easy to use by any person without a separate skilled process, and thus can be applied to a baby computer. In addition, it is a very useful invention that can obtain a higher character recognition rate at a lower cost.

Description

Character input device and method

The present invention relates to a character input device and method for use in a computer, and more particularly, having a plurality of pressure sensors in the lower portion along the character input path formed in a form that can represent both Korean and English characters, the character input The present invention relates to a character input device and a method for recognizing an input character by encoding a combination of a pressure sensor sensed by a pressing pressure by a separate input means passing along a path.

Currently, the most widely used character input apparatus is a keyboard, and a character input system using the same includes: a keyboard 10 to which information is input from a user; A parallel input / output device (PIO) 20 for scanning each key on the keyboard 10 and outputting a scan code value when a specific key is pressed; A memory (50) for storing a unique code value corresponding to the scan code value; And a central processing unit 40 for converting a scan code value input from the parallel input / output device 20 into a corresponding character unique code value (ASCII, KSC5601, etc.) in the memory 50. .

In the character input system using the conventional keyboard configured as described above, the parallel input / output device 20 performs one of a plurality of rows in order to confirm the pressing of the character keys matrix-arranged by the electrical signal lines on the keyboard 10. Output a signal with a different state to other rows in the row, and check to see if the state of a particular column changes. This operation is repeated continuously, changing the values of the rows sequentially. If the state of a particular column changes during this process, the character key is pressed, and thus the matrix value (scan code) is transmitted to the CPU 40.

The central processing unit 40 reads the character-specific code data for each scan code stored in the memory 50 with the input scan code value and used it in the subsequent processing as the input character.

However, the character input system using the conventional keyboard configured and operated as described above requires a lot of learning in order for the user to use it freely due to the complicated structure of the keyboard, and thus suitable for use by users of various levels such as infants. There was a problem that can not.

In addition, the handwriting recognition system, which is an improved system compared to the text input system using the keyboard, used the following method.

That is, the handwriting recognition system converts each stroke constituting the handwritten characters input from the user into a direction code, as shown in FIG. 5, in order to recognize a free handwritten character input from the user (S11). ); A matching step (S12) of matching each direction coded data by a character unit consisting of one or more strokes; And a path search step (S13) of finding a path of the character model that is best matched as a whole using the matching result.

In the handwriting recognition system configured as described above, after normalizing the size of the actual character input from the user in the data encoding step (S11), the respective strokes constituting the input character through the distance filtering of the input point are oriented one by one. Convert to code

Subsequently, the information of each direction-coded stroke is matched to form a complete character unit composed of one or more strokes, and then the patterns of the matched character data and the previously stored patterns of character data are compared. Choose the most similar pattern to minimize the difference.

Finally, by sequentially combining the selected characters through the matching and pattern matching process to find the best matching character model path.

However, the conventional handwriting recognition system configured and operated as described above is expensive to implement the system, and when the letters are continuously written, the recognition rate is lowered, the reliability is lowered, and the complexity for the recognition is complicated. There is a problem that increases exponentially.

Accordingly, an object of the present invention is to provide a character input device and a method which can be easily used without a skilled process by a user and have an improved character recognition rate.

Another object of the present invention is to provide a character input method capable of automatically distinguishing and recognizing continuous phoneme inputs without a separate input process for distinguishing between phoneme inputs.

1 is a block diagram showing a preferred embodiment of a character input apparatus according to the present invention,

2 is a flowchart illustrating a method of recognizing a character using a character input apparatus according to the present invention;

(A) is a flowchart of a method of confirming whether or not a character is input by the input completion button,

(B) is a flowchart of a method of confirming whether or not the character input is completed by comparing the position of the input point.

3 illustrates an example of a method of inputting a Korean or English consonant inputted to a character input apparatus according to the present invention;

FIG. 4 is a table showing a wiring relationship between input sensors of the character input device of FIG. 1 and input output signals of the parallel input / output device. FIG.

(A) of Figure 5 shows the initialization state of the input code consisting of the unit bits assigned to each pressure sensor,

(B) indicates the input code value corresponding to the letter d,

6 shows a general configuration of a character input system using a conventional keyboard,

7 shows a flow of a conventional handwriting recognition system.

* Description of the symbols for the main parts of the drawings *

1: Input panel 2: Square path

3: circular path 4: diagonal path

5: cross path 6: straight path

7: Character selection button 8: Input completed button

9: input pen 11: pressure sensor

12: Character input path

10: keyboard 20, 200: parallel input / output port

40, 400: central processing unit 50, 500: memory

b0, b1, b2 ... b36: Position indicator on the data bit to specify the pressure sensor

Character input device according to the present invention for achieving the above object, the character input path formed to represent both the Korean and English vowels; And a plurality of sensing means which are discretely installed along the character input path, for detecting a moving section on the character input path of an external input means, so that the selected combination can distinguish all phonemes of a desired language. Sensing means for detecting a movement path of the external input means; And a fixed path for fixing a movement between the sensing means of the external input means.

In addition, the character input method according to the present invention comprises: a first step of identifying and storing a position value by identifying a sensing means which changes state upon moving a path on the character input apparatus of claim 1 of any input means; A second step of repeating the first step until the phoneme input of the independent unit by the input means is completed; And a third step of correspondingly recognizing the phoneme code corresponding to the combination of the identified and stored position values.

According to another aspect of the present invention, there is provided a text input method, comprising the steps of: checking whether a new state change detection means is adjacent to a state change detection means immediately before; And stopping the repetition of the first step when the two sensing means are not adjacent to each other and the position of the newly changed state sensing means is the position of the sensing means corresponding to the discontinuous phoneme. Is there.

In the character input system using the character input apparatus according to the present invention configured as described above, a path of various geometric shapes formed on the character input path while the user moves on the character input path by using any input means is appropriate. When combined to make a desired phoneme, each of the sensing means located on the movement path of the input means of the sensing means provided along the character input path, the output state value is changed, the sensing means change state The input phoneme is recognized as follows using the position of.

While storing the positions of the sensing means for which the output state value is changed and confirming that the phoneme input of the unit independent from the user is completed, the random position is generated by combining the positions of the sensing means stored up to that point. That is, when the character is input using the external input means, an arbitrary code is generated using only the sensing means located on the input character path.

Subsequently, when the unique character code corresponding to the generated arbitrary code is read out from a memory or the like which stores the corresponding relationship, the standard code value corresponding to the phoneme input from the user is correctly recognized.

Hereinafter, a preferred embodiment of a character input apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing a preferred embodiment of a character input system using a character input apparatus according to the present invention, the character input apparatus comprising: an input plate 1 for recognizing characters input from a user; And an input pen 9 which is a separate external input means for inputting a character into the input panel 1;

The input panel 1 includes a character input path 12 formed to represent both Korean and English vowels and vowels; A plurality of pressure sensors (11) for detecting the trace of the input pen (9) passing while pressing on the letter input path (12); A character selection button 7 capable of converting input characters into Korean characters, English uppercase letters, and lowercase English letters so that various types of characters can be input by using the single character input path 12; And an input completion button 8 for inputting completion of character input in one independent unit, wherein the character input path 12 comprises: a square path 2; A circular path (3) inscribed in the square path (2); A diagonal path 4 connecting the opposite edges of the square path 2 with each other; A cross path 5 connecting the midpoints of opposite sides of the square path 2 to each other; And a straight path 6 formed on the square path 2 at a predetermined distance from the path 2.

In addition, the parallel input and output device 200, which is the same as the conventional configuration of FIG. 6, as a configuration circuit for sensing an electric change of a specific pressure sensor 11 on the character input path 12 and identifying an input element; A central processing unit 400; And a memory 500.

2 (a) is a flow chart showing the flow of the character input method by the character input system using the character input apparatus according to the present invention, hereinafter the operation of the preferred embodiment of the character input apparatus of FIG. This will be described in detail in parallel with the flowchart.

First, the pressure sensors 11 on the character input path 12 and the input / output signals of the parallel input / output device 200 are connected as shown in the table of FIG. 4, and the central processing unit 400 is Each of the pressure sensors 11 located on the character input path 12 internally assigns a unique position number b0, b1, b2 ... b36, and the pressure sensor 11 assigned a unique position number. Initializing the input code data as shown in Figure 5 (a) consisting of more than the total number of bits (bit), the parallel input and output device 200 is a signal state of a specific row in sequence as in the conventional Change the value and then read the thermal state repeatedly. In such a state, the user selects and presses one of the character selection buttons 7 according to the type of the character to be input_national character, English uppercase letter or lowercase letter _, and then uses the input pen 9 to input the input plate. A desired character is input along the character input path 12 formed in (1) (S31).

As shown in FIG. 1, the character input path 12 is formed of a plurality of paths having a geometric shape such as a square path 2 and a circular path 3, and when the paths are properly combined, As shown in FIG. 3, all the letters and vowels of Korean and English characters can be input accurately.

When a user starts to input a desired character on the character input path 12, for example, a consonant d, using the character pattern shown in FIG. 3, the character input path 12 Among the plurality of pressure sensors 11 formed on the pressure path, the pressure sensors 11 and b0 positioned on the path_ of the input pen 9 passing through the text input path 12 of the input text_12. , b1, b2, b3, b4, b11, b20, b19, b18, b17, b16, b25, b32, b33, b34, b35, b36 detect the pressure by pressing the input pen 9 ( S32) The output potential is changed.

The pressure sensor 11 at which the output potential changes is specified by the parallel input / output device 200, which means that when the state value of the column (c0, c1, ...) to be read is changed, This is done by sending the output value and the reading of the heat to the CPU 400. That is, while the consonants of the bars and starts to _{enter, FEFE (16), FEFD (} 16), FEFB (16), FEF7 (16), ···, scan code value in turn, the central processing unit of BFEF (16) 400 is inputted, and the central processing unit 400 determines that a series of pressure sensors b0, b1, b2, ... b35, b36 of the pressure sensor 11 have been pressed. The bits in the input code are sequentially set to 1 in the input order (S33).

Finally, when the input completion button 8 is selected by the user to complete character input in one independent unit, that is, when the value of the BFBF ₁₆ is input from the parallel input / output device 200 (S34). In this case, one code having a value as shown in FIG. 5B is generated by the scan code value transmitted from the parallel input / output device 200 until then (S35).

The CPU 400 compares the generated input code with the character-specific code (ACSII, KSC5601, etc.) data for each input code stored on the memory 500 (S36), thereby inputting the input character path. The corresponding standard code (A4A9 ₍₁₆₎ ) can be obtained, the character corresponding to the standard code can be displayed on the screen (S37).

In this case, a method of checking whether the input completion button 8 is pressed to confirm that input of one independent character is completed is used. However, the pressing of the separate input completion button 8 is not confirmed. Without checking, input completion can be confirmed using a method as shown in FIG.

That is, while the input pen 9 is moved to input a desired character (S51), when the output state value of the pressure sensor 11 which detects the pressure of the input pen 9 (S32) is changed, the center is changed. The processing apparatus 400 determines the corresponding position according to the above-described method (S53), and compares the position of the pressure sensor 11 immediately before the state is changed (S54), and compares the two pressure sensors 11 In step S55, it is determined whether the character input path 12 is adjacent to each other.

If the two pressure sensors 11 to be compared are adjacent to each other, that is, located near each other on the letter input path 12, it is determined that the input of the letter is not yet completed by the user and the position of the pressure sensor immediately inputted. After updating, the value of the unit bit allocated to the pressure sensor newly changed in the input code is set to 1, and the output value of the parallel input / output device 200 is continuously detected.

However, when the two pressure sensors 11 to be compared are not located in the mutual rectus, that is, when the newly changed state of the pressure sensor 11 is located in a non-adjacent position from the pressure sensor 11 identified immediately before, In order to form a phoneme, the position of the newly changed state of the pressure sensor 11 should be in a specific positional relationship with the pressure sensor 11 of which the state is changed immediately before, as shown in the letter input path shown in FIG. 3. Bar, in view of the combination of the pressure sensors 11 confirmed to date, it is checked whether or not the pressure sensor 11 is located at a specific position that can be discontinuously positioned (S56).

If the pressure sensor 11 in the position where discontinuous input is possible, e.g. b34 and b32 (in this case, the scan code is FBBF ₁₆ and FEBF ₁₆₎ is continuously changed. In this case, since the character is still determined as being input (Korean 영문, English I or J, etc.), after setting the value of the unit bit allocated to the pressure sensor newly changed in the input code to 1 (S60), the pressure sensor Continue to detect the pressing of (11).

However, if it is confirmed that the input sensor is located at a position where it is not possible to input non-adjacently, that is, if b34 is confirmed after b34, this means that the input of the existing character is completed and the newly changed pressure sensor 11 ) Means to input a new character, so that the input code generated until immediately before the state change detection of the pressure sensor is completed (S57), and the input completion button 8 corresponds to the input path through the same process. You will get a unique character code value.

In addition, instead of the above method of separately checking whether the input is completed, whenever the pressure sensor 11 that detects the pressure of the input pen 9 changes its output state value, the pressure sensor changed state until then ( It is also possible to recognize the corresponding character code as the input character when the same code is present while comparing the unique character code data for each input code stored in the memory 500 by using the input code generated by 11). .

Character input device and method according to the present invention configured and acting as described above, is easy to use without any skilled process is applicable to a baby computer, and also very useful inventors to obtain a higher character recognition rate at a lower cost will be.

Claims (6)

A character input path formed to represent both Korean and English vowels; And And a plurality of sensing means arranged discretely along the character input path, for detecting a moving section on the character input path of an external input means. The method of claim 1, The character input path, Square path; Diagonal paths connecting opposite edges of the rectangular paths to each other; A cross-shaped path connecting the midpoints of opposite sides of the rectangular path to each other; An inward path connecting the midpoints of neighboring sides of the rectangular path to each other in a curved shape; And And a straight path formed on the rectangular path at a predetermined distance from the path. The method according to claim 1 or 2, The sensing means is a character input device, characterized in that the pressure sensor is installed in the lower portion of each interval between the intersection of the character input path and the neighboring intersection. Sensing means arranged to distinguish all phonemes of a desired language from the selected combination, and detecting a movement path of an external input means; And And a fixed path for fixing the movement between the sensing means of the external input means. A first step of identifying and storing a position value by identifying a sensing means which changes state when the path is moved on the character input apparatus of claim 1 of any input means; A second step of repeating the first step until the phoneme input of the independent unit by the input means is completed; And And a third step of correspondingly identifying the phoneme code corresponding to the combination of the identified and stored position values. The method of claim 4, wherein The second step, Checking whether the newly changed state detection means and immediately before the state change detection means are adjacent; And Stopping the repetition of the first step if the two sensing means are not adjacent to each other and the position of the newly changed state sensing means is the position of the sensing means corresponding to the discontinuous phoneme. Character input method.