JPH0457034B2 - - Google Patents

Description

[Detailed description of the invention] [Technical field of invention] The present invention relates to a character recognition device.

[Prior art]

In recent years, various character recognition devices have been developed that recognize handwritten input of literature, graphics, etc. on a tablet using a pen-shaped jig, and convert the input information into input information. In this type of character recognition device, when a character pattern is input, the coordinates of the character pattern are detected, and the detected coordinates are sequentially written into a predetermined memory, and then recognized according to the contents of this memory. It's getting old. In this type of device, if a large capacity memory is used, the control circuit becomes complicated and the cost increases, so it is preferable to use a device whose memory capacity is suppressed to a certain extent.

[Problems with conventional technology]

However, when writing coordinates into the memory sequentially, for example, if a character with a long stroke is input even if the character is the same stroke, the memory will overflow in the middle of writing, and no more data will be stored in the coordinate memory. It became impossible to write, and as a result, erroneous recognition occurred.

[Purpose of the invention]

This invention was made against the background of the above-mentioned information, and its purpose is to effectively write coordinates in memory without increasing memory capacity when characters with long strokes are input. An object of the present invention is to provide a character recognition device that prevents recognition from occurring.

[Key points of the invention]

In order to achieve the above-mentioned object, this invention
When writing coordinates from the first coordinate of an input character pattern to a predetermined number into memory sequentially, if a character pattern with a long stroke that exceeds the memory capacity is input, the final coordinates are written. The gist of this is that the data is written to the final storage area of the memory.

〔Example〕

Hereinafter, one embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a circuit diagram of, for example, an electronic wristwatch equipped with a character recognition device. In this case, the input section 1 is formed by further arranging transparent touch electrodes in, for example, a 4x4 matrix on a transparent surface glass fixed to the top surface of the watch case. The 16 touch electrodes are arranged on the XY coordinate system, and for example, 16×16=256
The coordinate position is formed by electrical processing.
Also, if you touch the human body such as a finger to the touch electrode,
At this time, the contact capacitance component generated in the touch electrode is detected and the coordinate position is input, thereby inputting a character pattern. This type of technology has already been proposed in a patent application filed by the applicant (Japanese Patent Application No. 55-35660, title of invention: touch switch device).
You can use . Depending on the input character pattern, a message is displayed on the display section 2, for example, when the alarm time arrives. Note that this display section 2 is composed of a liquid crystal display device, and of course also displays time data.

The ROM (read only memory) 3 stores microprograms that control all operations of this electronic wristwatch, and includes microinstructions AD, DA, OP,
Output NA in parallel. Therefore, micro instructions
AD is applied to ROM 4 and RAM (random access memory) 5 as address data, respectively.
Further, the microinstruction DA is applied to the RAM 5 or the arithmetic unit 6 as data. Furthermore, the microinstruction OP is applied to the operation decoder 7, and the operation decoder 7 accordingly outputs the control signals CS ₁ , CS ₂ , X, Y, Z, R, S,
Output G, etc. The microinstruction NA is applied to the address section 8, and in response, the address section 8
Address data for reading out microinstructions AD, DA, OP, and NA necessary for the next process is output from signals d, c, etc., which will be described later, and is supplied to the ROM 3.

The ROM 4 stores vector strings (described later) of standard character patterns for characters such as numbers and alphanumeric characters as data, and the input unit 1
The character is recognized by comparing it with a vector string of character patterns (described later) input from the character pattern. Note that data is read from the ROM 4 in accordance with the control signal _CS1 .

The RAM 5 has various registers, which will be described later, and is used in various processes such as time measurement and character recognition performed by the arithmetic unit 6, and data is read and written under control signals CS ₂ and R/W.

The calculation unit 6 executes the various calculations described above under the control signal X. The resulting data is then provided to the input section 1, display section 2, and RAM 5. Further, when executing a jump calculation, a signal d indicating the presence or absence of data as a result of the calculation and a signal c indicating the presence or absence of carry occurrence are outputted and supplied to the address unit 8, respectively.

The SR type flip-flop 9 is set or reset by control signals R and S, respectively. The control signal S is output at the start of input of each stroke of a character pattern, and the control signal R is output at the end of input. . The set output signal is then input to AND gate 10 as a control signal. A clock of a predetermined frequency output from an oscillation circuit 11 is input to the other end of this AND gate 10, and the clock output from the AND gate 10 is input to a timer section 12 consisting of a counter and counted. Ru. This timer section 12 measures the input time (ON time) of each stroke and the OFF time from the end of input of each stroke to the time of input of the next stroke. The ON time and OFF time thus obtained are determined by the gate 13, which is opened when the control signal G is output.
The data is supplied to the RAM 5 and the calculation unit 6 via the RAM 5 and the calculation unit 6. Note that the timer section 12 is cleared by the control signal S. Note that the input section 1 and the display section 2 are controlled by a control signal Z or Y, respectively.

The oscillation circuit 11 constantly oscillates a reference frequency signal of, for example, 32,768 Hz, and supplies it to the frequency dividing circuit 14.
A 16 Hz signal is output from the frequency dividing circuit 14 and applied to the address section 8. In response to this, timing processing is executed once every 1/16 seconds.

Next, the standard vector string stored in the ROM 4 and the vector string of the character pattern input from the input section 1 will be explained. Figure 2 shows the situation when the number "2" with a stroke number of 1 is input from the input section 1. As shown in Figure 2A, when the character pattern "2" is input, its coordinate position data is are sequentially written to RAM5. In this case, a maximum of 32 pieces of coordinate position data are written in the RAM 5 in one stroke. Then, as shown in Figure 2B, the character pattern "2"
When the finger leaves the touch electrode of the input unit 1 after inputting , the stroke length is calculated from the coordinate position data written in the RAM 5, and then the stroke length is divided into six equal parts. Then, as shown in Figure 2c, each equally divided point is approximated by a straight line from the starting point to the ending point, and the vector of each part is determined according to the vectors (8 types from 0 to 7) in Figure 3. , a vector sequence is calculated.
The calculated vector string is then compared with the standard vector string in the ROM 4, and the most similar character pattern is output.

FIGS. 4, 5, 6, and 7 show standard vector sequences of character patterns having stroke counts of 1, 2, 3, and 4, respectively, and are stored in the ROM 4.

Next, the configuration of the RAM 5 will be explained with reference to FIG. The T register is used to store current time, the TM register is used to store timer time, and the AL register is used to store alarm time. In this case, a fixed time is set in the TM register to detect character breaks, and if the OFF time of the timer section 12 is smaller than the fixed time of the TM register, it is considered as the end of writing one stroke; For example, it is the end of one character, that is, the separation between characters. Also,
The F _A , F ₁ , and F ₂ registers are for flag storage.
Z is a stroke number counter that counts the number of strokes for one character. Also, Y is
The address counter counts the addresses for sequentially addressing the 32 memory areas M ₁ to M ₃₂ each time the coordinates from the input section 1 change. Furthermore, in the memory areas M ₁ to M ₃₂ , 1
Coordinate position data for a maximum of 32 strokes are stored sequentially. Furthermore, the memory areas m ₁ , m ₂ , m ₃ , and m ₄ are for storing vector sequences of the first stroke, second stroke, third stroke, and fourth stroke of one character. In this example, it is possible to input up to four strokes per character.

Next, the operation of this embodiment will be explained with reference to FIGS. 9 to 11.

First, an overview of the overall operation will be explained with reference to the general flow shown in FIG. This general flow is such that every time the signal 16Hz is output from the frequency dividing circuit 14,
That is, execution starts every 1/16 seconds. First, the time measurement process of step _S1 is executed, and the calculation unit 6
calculates the current time data by performing a predetermined operation on the previous data in the T register of the RAM 5. This current time data is then sent to the display section 2 and displayed.

Next, timer processing in step _S2 is executed.
This timer processing requires some processing to be performed for a certain period of time in the flow described later, and if a certain period of time is preset in the TM register, the specified period of time is subtracted each time this process is executed.

Next, a process for determining whether or not the message setting mode is in effect is executed in step _S3 . In this judgment process, it is determined whether the message setting mode is set depending on whether a mode switch (not shown) for message setting is turned on or not, and if "YES", the step is executed. Proceed to S ₈ and proceed to the character recognition processing routine, and on the other hand,
If "NO", the process proceeds to step _S4 . In this judgment process, it is judged whether or not the alarm time preset in the AL register has been reached.
If ``YES'', the process advances to step _S5 , where the message data is read out from the RAM 5 and displayed, and when it is displayed for a certain period of time, this is determined in step _S6 , and the message is displayed in step _S7 . The display will be erased. _On the other hand, if the determination at steps _S4S6 is "NO", this general flow process ends and other processes (not shown) are started.

Furthermore, if the setting of the message setting mode is determined in step _S3 and the process proceeds to step _S8 , in this step _S8 , the flag F _A in RAM5 is
It is determined whether or not the value is "0". Therefore, it is normally set to ``0'' when the character recognition device is not running, and therefore, the process proceeds to step _S9 .
Data "1" is set in the flag F _A , and it is stored that the character recognition process is being executed. Then, in accordance with the flow described later, recognition processing of the character pattern whose coordinates are input from the touch electrode of the input unit 1 is executed (step S ₁₀ ), and the input message data is stored in the RAM 5 (step S ₁₁ ).
Furthermore, the display is confirmed on the display section 2 (step _S12 ).
Thereafter, the flag F _A is cleared (step S ₁₃ ) and the character recognition process execution state is released. Incidentally, in step _S8 , if the flag F _A is not "0", the process returns to the process being executed previously.

FIG. 10 is a flowchart showing the specific content of the character recognition process in step _S10 . That is, when entering the character recognition processing step, first, the initialization processing of step S _A is executed. That is, the _F1 register in RAM5 and
Flag “1” is set in the _F2 register, and
The contents of the stroke number counter Z are cleared and, moreover, the address counter Y is set to "1". In this case, the _F1 register is set to "0" at the beginning of one stroke and "1" at the end, and the _F2 register is set to "0" at the beginning of one stroke and "1" at the end. is set. Then, in the next step _S2 , after executing the process of reading the coordinate position data from the input section 1,
Proceeding to step _S3 , it is determined whether or not a character pattern has been input from the input unit 1, in other words, whether or not the touch electrode has been touched.If no touch input has been made yet, the process proceeds to step _S15 , and _F1 is entered. It is determined whether the contents of the register are "1" or not. Since it is initially "1", the process returns to step _S2 and the same process is repeated, and the state is in a standby state until there is a touch input.

When there is a touch input, the process advances to step _S4 and the same process as the above step _S15 is executed, but since it is initially "1", the contents of the timer section 12 are cleared and then , starts the timing operation for the above-mentioned OFF time (step _S5 ). After clearing the contents of the _F1 register in the next step _S6 , the process proceeds to the next step _S7 , where it is determined whether the contents of the _F2 register are "1" or not. At first, “1”
Therefore, the process advances to step _S8 , and an increment process is executed to increment the contents of the stroke number counter Z by 1. As a result, when you start writing one stroke of a character, the content of the stroke number counter Z is "1".
This is the value for the first stroke. Then F ₂
After the contents of the register are cleared (step _S9 ), the process proceeds to step _S10 , where it is determined whether or not there is a coordinate change. In this case, when the character pattern is written by moving the fingers, the coordinates change accordingly, but if the coordinates do not change, the process returns to step _S2 . In other words, when you are just touching the touch electrode and not moving your finger, steps S ₂ , S ₃ ,
It enters a standby state that goes through a loop of S ₄ , S ₇ , and S ₁₀ . Then, if there is a coordinate change, proceed to step _S11 ,
Whether the contents of address counter Y is less than "32" or not,
That is, in this embodiment, since it is possible to read a maximum of 32 coordinates per stroke, this judgment process detects whether or not the final coordinates of one stroke have been read. Most importantly, since the value of the address counter Y is "1", the process proceeds to step _S13 , and the memory area M _Y to be addressed is selected from among the memory areas M ₁ to _{M 32} according to the contents of the address counter Y. That is, the first input coordinate position data is written into the memory area _M1 . Then, in the next step _S14 , the contents of the address counter Y are incremented by 1, and then the process returns to step _S2 . Therefore, steps _S13 and _S14 are repeatedly executed until the value of address counter Y reaches "32", and as a result, 31 coordinate position data are stored in address counter Y during input of one stroke. The data is sequentially written to memory areas _M1 to _M31 according to the content.
When the value of the address counter Y reaches "32", the process proceeds from step _S11 to step _S12 , and after the 32nd coordinate position data is written to the final memory area _M32 , the process proceeds to step _S2 . return. In this case, 1
During the input of a stroke, once the value of the address counter Y reaches "32", step _S12 is repeatedly executed, and 32 or more coordinate position data are written into the memory area ₃₂ .

Therefore, characters with long strokes, e.g.
As shown in FIG. 11, when the number "8" is input, the coordinate position data from point to is sequentially written into the corresponding memory areas _M1 to _M31 , but the coordinates from point to The position data is written to the final memory area _M32 . Therefore, since the contents of the memory area _M32 are rewritten sequentially starting from the point, the data of the coordinate position of the point, that is, 1
The stroke becomes the final coordinate position data. As a result, the points are approximated by a straight line in terms of coordinates.

On the other hand, when you finish writing the first stroke of a character and start writing the next stroke, or for characters with a stroke count of 1, the first stroke of the next character to be input. When you release your finger from the touch electrode of the input section 1 to start writing, this is detected in step _S3 and the process proceeds to step _S15 , but in this case,
Since the contents of the _F1 register are "0", the process further advances to step _S16 , where it is determined whether the contents of the _F2 register are "0" or not. In this case as well, since it is "0", the process proceeds to step _S17 . Here, RAM ₅
The length (l) of the stroke is calculated from the coordinate position data written in each of the memory areas M ₁ to _{M 32} . The stroke length calculated in this way is then divided into six equal parts in the next step _S18 .
After the equally divided points are extracted, the vectors of each part are determined according to the vectors in FIG. 3, and a vector sequence is calculated (step _S19 ). Then, this vector sequence is stored in memory areas m ₁ to _{m 4} of RAM 5.
The contents of the stroke number counter Z are written to the addressed memory area mz (step
_S20 ). Therefore, the vector sequence of the first stroke is written to the memory area _m1 . In the next step _S21 , the _F2 register is set to "1", and then the process proceeds to step _S22 , where the OFF time of the timer section 12 is set for a certain period of time set in the TM register in the RAM5. It is determined whether or not it has been reached. Here, if the above-mentioned OFF time is longer than the above-mentioned fixed time, it is determined that one character has been inputted, but if it is smaller, the process returns to step _S2 and waits for the next stroke to be input.

Then, when the second stroke is input, the coordinate reading process is executed in the same manner as described above, and then,
Steps _S17 to _S22 are executed. Therefore, for each character with a stroke count of 22, 3, and 4, the step
Since the process of S ₂₀ is repeated 2, 3, and 4 times,
Vector sequences of the second stroke, third stroke, and fourth stroke are written to memory areas m ₂ , m ₃ , and m ₄ in the RAM 5.

In this way, when one character has been input, this is detected in step _S22 and the process proceeds to step _S23 , where _F1
"1" is set in the register. Then, as described above, the vector strings written in the memory areas m ₁ to _{m 4} are classified by number of strokes, and the vector strings for each number of strokes are compared with the standard vector string in ROM 4 (step S ₂₄ ). As a result, the most similar character pattern is extracted (step
_S25 ). In this case, the direction difference between each vector in the vector string obtained by inputting the character pattern and the standard vector string is detected, and the smallest sum of these is extracted as the most similar character pattern. . The character pattern extracted as described above is then displayed on the display section 2 (step _S20 ).

In addition, in the above example, the XY coordinate system is 16×16
256 points arranged in a matrix, but the scale of the coordinates is not limited to the above example. Further, in the above embodiment, the number of strokes of characters to be inputted is up to four strokes, but the number can be arbitrary, and the types of characters can also be katakana, hiragana, kanji, symbols, etc. Further, in the above embodiment, a maximum of 32 pieces of coordinate position data are read in one stroke, but the number may be arbitrary.

〔Effect of the invention〕

As explained in detail above, this invention is capable of writing input character patterns that have long strokes that exceed the memory capacity when sequentially writing coordinates from the first coordinate of an input character pattern to a predetermined number into memory. When a character pattern is detected, the final coordinates are written into the final storage area of the memory, so even long stroke character patterns can be recognized accurately without increasing the memory capacity.

[Brief explanation of drawings]

The drawings show an embodiment of the present invention; FIG. 1 is a circuit diagram of an electronic watch equipped with a character recognition device, and FIGS. , a diagram showing the process of obtaining the vector sequence, Figure 3 is a diagram explaining the relationship between the numerical values and vectors that make up the standard vector sequence,
Figures 4, 5, 6, and 7 are diagrams showing standard vector sequences of character patterns with stroke numbers of 1, 2, 3, and 4, respectively, and Figure 8 is a diagram of the RAM configuration. , FIG. 9, and FIG. 10 are flowcharts for explaining the operation, and FIG. 11 is a diagram showing a state in which the coordinates are imported when the number "8" is input. 1...Input section, 3, 4...ROM, 5...
RAM, 6...Arithmetic unit, _M1 to _M52 ...Memory area.

Claims (1)

[Scope of Claims] 1. Character pattern input means; detection means for detecting coordinate data of the input character pattern while the character pattern is being input from the character pattern input means; and a predetermined number of memory areas. a first storage means for storing coordinate data detected by the detection means up to the predetermined number while a character pattern of one character is input from the character pattern input means; a second storage means for storing final coordinate data of the character pattern of one character when the detection means detects coordinate data of the predetermined number or more while the character pattern of the character is being input; A character recognition device comprising: means for recognizing the character pattern according to coordinate data stored in first and second storage means.