JPH09106439A - Character recognizing device, its method and computer controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a character recognition device, its method and a computer controller which recognize a print hand and an abbreviated/running hand without increasing a storage capacity and a recognizing time. SOLUTION: The number of inputted strokes is counted to compare the feature of the inputted strokes and the feature of the same number of strokes with the counted strokes, which are stored in a recognition dictionary 206a storing the feature of strokes composing a standard character. As a result of comparison, in the case a coinciding ratio is lower than a prescribed value, the feature of the inputted strokes and the feature of the more strokes than the same number of strokes stored in the dictionary 206a are compared with each other. Based on the result of comparison, a character composed of the inputted strokes is defined.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character recognition device for recognizing a character based on a characteristic of an input stroke, a method for the character recognition device, and a computer control device.

[0002]

2. Description of the Related Art Conventionally, in recognition of characters input using a coordinate input device such as a digitizer, a character recognition device capable of recognizing handwritten characters such as continuous characters and abbreviated characters, which are not regular writing (regular writing). is there. In this type of device, the recognition dictionary stored in a storage medium or the like stores character information of consecutive characters and abbreviations in addition to the character information of the regular writing,
Character recognition was performed by comparing the character information stored in the dictionary with the input handwritten character information.

For example, by holding a pattern of consecutive characters of a "book" originally written in 5 strokes in a dictionary of 4 strokes including "tree" in a regular drawing, the "book" becomes a continuous stroke of 4 strokes. Even if it is input, the character "book" can be recognized. In order to recognize abbreviations and continuation characters without slowing down the recognition speed, a dictionary for regular writing (regular writing) and a dictionary for abbreviations / continuation (non-regular writing) are separated. There is a method of matching input characters using a dictionary and matching the input characters with the dictionary for abbreviation / continuation characters when the recognition result is not bad.

[0004]

However, in order to be able to recognize all the abbreviations / successive characters, if the configuration is such that the characters corresponding to all the abbreviations / successive characters are also stored in the dictionary, compared to a dictionary containing only regular writing. , The storage capacity of the dictionary will increase significantly. Further, it is necessary to perform matching with many patterns, and as a result, more recognition time is required, which causes a problem that the processing speed of recognition processing is reduced. Also, especially in the case of online handwritten character recognition, the probability of inputting a regular drawing is almost the same as the probability of inputting abbreviations / successive characters, and if either of them is recognized preferentially in advance, the recognition rate will drop significantly. There was a problem that it might happen.

The present invention has been made in view of the above problems, and a character recognition apparatus and method and a computer therefor capable of recognizing a regular character, abbreviations / continuation characters without increasing the storage capacity and the recognition time. The purpose is to provide a control device.

[0006]

A character recognition apparatus according to the present invention for solving the above problems has the following configuration. That is, a character recognition device for recognizing a character based on the characteristics of an input stroke, a dictionary that stores the characteristics of a stroke that forms a standard character, and a counting unit that counts the number of strokes of the input stroke, As a result of the first comparing means for comparing the input stroke characteristics with the stroke characteristics of the same number of strokes stored in the dictionary as the number of strokes counted by the counting means, the matching ratio is predetermined. If it is lower than the value, the second comparison means for comparing the characteristics of the input stroke with the characteristics of the stroke having the stroke count larger than the same number of strokes stored in the dictionary, the first comparison means and the second comparison means. Based on the results of the means
Determining means for determining the character formed by the input stroke.

Preferably, the second comparing means is
The stroke feature stored in the dictionary of the stroke number obtained each time the predetermined number is added to the stroke number is sequentially compared with the input stroke until the stroke number obtained each time the predetermined number is added reaches the predetermined stroke number. This is because it is possible to perform character recognition of abbreviated characters / contiguous characters by comparing with the characteristics of strokes that are equal to or more than the input stroke number.

Preferably, the first comparing means is
The input stroke feature is compared with all stroke features stored in the dictionary and having the same number of strokes as the number of strokes counted by the counting means. Further, preferably, the dictionary is a dictionary storing characteristics of strokes forming the standard character in accordance with a standard stroke order of the standard character.

Preferably, the first comparison means is
Up to a predetermined order of the input order of the input strokes, the characteristics of the strokes that are the same as the number of strokes counted by the counting means stored in the dictionary and that are in the same order as the input order After the predetermined order, the feature of the input stroke is compared with all the features of the stroke stored in the dictionary and having the same number of strokes as the number of strokes counted by the counting means.

Further, preferably, the predetermined value is a value obtained by a function having at least one of the stroke number of the input stroke, the input time, and the input speed as a variable. Also, preferably, the function is an increasing function. Further, preferably, the determining means is a character corresponding to the comparison object having a highest proportion of matching with the feature of the stroke of the dictionary compared by the first comparing means and the second comparing means. Is determined as the recognition character.

Further, preferably, the determining means further comprises an adding means for adding a predetermined bias value to each of the comparison results obtained by the first comparing means and the second comparing means, and the determining means obtains by the adding means. Based on the new results
Determine the character that the input stroke comprises.

Further, preferably, the predetermined bias value is a bias value determined based on at least one of the stroke number of the input stroke, the input time, and the input speed. A character recognition method according to the present invention for achieving the above object has the following configuration. That is, a character recognition method for recognizing a character based on a characteristic of an input stroke, a counting step of counting the number of strokes of the input stroke, and a stroke forming a standard character with the characteristic of the input stroke. As a result of the first comparing step of comparing the number of strokes counted in the counting step and the features of the same number of strokes stored in the dictionary storing the features of the first comparing step, the matching ratio is lower than a predetermined value. In this case, the second comparison step of comparing the characteristics of the input stroke with the characteristics of the stroke having the stroke count larger than the same number of strokes stored in the dictionary, and the results of the first comparison step and the second comparison step And a determining step of determining a character formed by the input stroke.

A computer controller according to the present invention for achieving the above object has the following configuration. That is, a computer control device for controlling a computer by reading a predetermined program from a memory medium, wherein the memory medium counts the number of strokes of the input dictionary and a dictionary that stores the features of strokes that form a standard character. The procedure code of the counting step, the procedure code of the first comparing step of comparing the characteristics of the input stroke with the characteristics of the strokes stored in the dictionary, the stroke number being the same as the number of strokes counted in the counting step, and As a result of the first comparison step, if the matching ratio is lower than a predetermined value, the second comparison for comparing the feature of the input stroke with the feature of the stroke having the number of strokes larger than the same number of strokes stored in the dictionary. The character formed by the input stroke is determined based on the procedure code of the process and the results of the first comparison process and the second comparison process. And a procedure code of constant step.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the character recognition device according to the present embodiment. 201
Is a coordinate input unit, and includes a digitizer 201a and a pen 201.
b. Digitizer 20 with pen 201b
It is possible to input coordinate data of characters and figures on 1a. The input coordinate data can be displayed on the display unit 202 under the control of the CPU 203. Reference numeral 202 denotes a display unit, which is a display such as a CRT, and which displays coordinate data of characters and figures input by the coordinate input unit 201 and C.
The character data read from the memory 204 or the like under the control of the PU 203 is displayed.

Reference numeral 205 denotes an FDD (floppy disk drive), which is equipped with the FD and can read / write data. Further, it is also possible to execute a process by writing a program of a processing flow (FIG. 3) described later in the attached FD (not shown) and reading the program into the RAM 204b of the present apparatus. In the first embodiment, the program is stored in the ROM 204a and the CPU 20
The process flow (FIG. 3) described later is executed under the control of No. 3.

Reference numeral 206 denotes an HDD (hard disk drive), which has a built-in recognition dictionary 206a, which will be described later, for recognizing characters input from the digitizer 201a. The recognition dictionary 206a is stored in a storage medium such as a hard disk (HD) capable of storing a large amount of data. Reference numeral 203 denotes a CPU that performs recognition of input characters and performs various controls. Reference numeral 204 denotes a memory, which includes a ROM 204a and a RAM 204b. R
The OM 204a stores a recognition program and various control programs used by the CPU 203 for character recognition. Also, it stores character font information and character codes. The RAM 204b is an HD built in the HDD 206.
An area for temporarily storing data read from a dictionary stored in (not shown), input coordinate data, data such as variables used by the recognition program, or a work area for various data.

FIG. 2 is a functional block diagram showing the configuration of the functions executed under the control of the CPU 203 of this embodiment. Under the control of the CPU 203, first, the same stroke number matching unit 101 is activated, and in the character data input from the digitizer 201a, the number of strokes forming the input character is counted. Then, the character data is matched with the dictionary having the same number of strokes as the counted number of strokes.
After the matching is completed, the same stroke number result examination unit 102 is activated to examine whether or not the matching result is good. If the matching result is good, the recognition result adjusting unit 10
4 is activated and the result is output as a recognition result. On the other hand, when the matching result is bad, the positive stroke number matching unit 103 is activated and the input stroke (hereinafter,
It is sequentially matched with the dictionary in each stroke number up to a predetermined stroke number which is greater than or equal to the stroke number of (input stroke). When the matching up to the predetermined number of strokes is completed, the recognition result adjusting unit 104 is activated, and the optimum matching result is determined from the matching results of the strokes of the input stroke to the predetermined number of strokes, and the determined matching result is determined. Is output as a recognition result.

Next, regarding the processing flow of this embodiment,
This will be described with reference to the flowchart of FIG. FIG. 3 is a flowchart showing the processing flow of this embodiment. In step S300, a stroke is input to the digitizer 201a using the pen 201b. In step S301, C
The same stroke number matching unit 101 is activated under the control of the PU 203, and the stroke information of the input stroke and the stroke information held in the dictionary (hereinafter referred to as dictionary stroke) are set in stroke units, partial pattern units, character units, or character strings. Match by unit.

As a matching method, the order of input strokes and the order of dictionary strokes stored in the recognition dictionary 206a may be matched one-to-one, or each order may be ignored. You may match by brute force. The recognition dictionary 206a used by the same stroke count matching unit 101 holds dictionary strokes having the same stroke count as the input stroke count. As the stroke information having the same stroke number, for example, information on strokes forming a regular writing (regular writing) and information on strokes forming frequently used abbreviations / contiguous characters (writing other than regular writing) may be stored. . In this case, since the abbreviation / successive stroke information is included in the recognition dictionary 206a, a certain recognition rate for abbreviated / successive characters can be secured at this point.

When the matching by the same stroke number matching unit 101 is completed in step S302, the same stroke number result examining unit 102 is activated to determine the matching result of the same stroke number matching unit 101. When it is determined that the matching result is good (YES in step S302), the matching result is stored in the RAM 204b, and step S30 is performed.
Go to 7. When it is determined that the matching result is bad (NO in step S302), the matching result is stored in the RAM.
It stores in 204b, and progresses to step S303.

As a method of determining the matching result, for example, the matching result is determined based on whether or not the matching probability of the candidate character of the character recognized as the result of the same stroke number matching unit 101 is larger than a predetermined threshold value. So, for example,
The threshold of 80% is set, and the matching probability of candidate characters is 8
When it is 0% or more, this result is judged to be good. Further, abbreviated characters / successive characters are often written faster than regular writing. Therefore, the threshold may be set higher as the number of input strokes is larger, and may be set lower as the number of input strokes is smaller.

Further, for example, the threshold value may be defined by the following expression expressed by a linear function. Threshold = 60 + 2 × (number of strokes of input stroke) In this case, if 20 or more characters are input, it is unconditionally determined that the matching result of the same stroke number matching unit 101 is bad.

The threshold is not limited to the setting of the threshold value using the linear function for the stroke number of the input stroke as described above, and any increasing function for the stroke number (need not be monotonic increase) may be used. Good. In addition to the number of strokes of the input stroke, the writing time of the input stroke, the writing speed of the input stroke, and the like can be used to easily set the threshold value.

If the matching result in step S302 is poor, there is a possibility that strokes have been input in abbreviated / continuous characters. Therefore, in the subsequent steps, the number of strokes is increased and the recognition processing is performed in the matching target step. In step S303, +1 stroke is added to the stroke count i of the input stroke to update the stroke count i (i = a positive integer). Step S304
And the updated stroke number i is a predetermined stroke number n (n = a positive integer,
However, it is determined whether or not n> i). When the stroke number i is larger than the predetermined stroke number n (YE in step S304)
S), the process proceeds to step S306, and the stroke number i is the predetermined stroke number n.
If it is smaller (NO in step S304), the process proceeds to step S305. In step S305, the plus stroke number matching unit 103 is activated, and the dictionary strokes held in the recognition dictionary 206a having the same stroke number as the updated stroke number i are matched in stroke units, partial pattern units, character units, or character string units. To do. Then, steps S303 to S3 are performed until the number of strokes i reaches a predetermined number of strokes n.
The matching result obtained each time the stroke number i is updated by sequentially repeating the processing of 05 is sequentially stored in the RAM 204b.

The upper limit of the predetermined stroke number n is determined as a function of the stroke number of the input stroke, the writing time, and the writing speed, like the threshold value set in the stroke number examining unit 102.
In addition, the plus stroke number matching unit 10 in step S305.
In the matching method in 3, the input order of the input strokes and the order of the dictionary strokes stored in the recognition dictionary 206a may be matched in a one-to-one correspondence. You may match by winning. In particular, when abbreviated characters or continuous characters are input and there is no clear tendency as to which stroke is abbreviated, it is desirable to perform the matching method described below with reference to FIG.

FIG. 4 is a diagram for explaining a matching method in the plus stroke number matching unit 103 of this embodiment. For example, as shown in (a) of FIG. 4, the character formed by the input stroke is a continuous character of “book”, and the input pattern and the stroke order, which are the position composition of the character,
The matching is performed with the stroke order and the dictionary pattern, which is the position configuration of the dictionary strokes forming the "book" of the open-face type, stored in the recognition dictionary 206a as shown in (b). In the matching method in this case, as shown in FIG. 4C, each input order of the input strokes is matched with the dictionary stroke shown in FIG. 4B. The numbers shown in (c) of FIG. 4 are the input order of the input strokes and the input order of the dictionary strokes, and the combinations for which matching is performed are indicated by line segments connecting the orders. In this case, a matching method is shown in which each order of input strokes is matched with each order of dictionary strokes.

If the input order of a character or a character string composed of input strokes is generally abbreviated in the subsequent input strokes, for example, as shown in FIG. For the input stroke having the smallest input order, that is, the first stroke and the second stroke of the input stroke are matched with the same stroke number of the first stroke and the second stroke of the dictionary stroke, respectively. The third stroke and the fourth stroke of the input stroke may be matched with each stroke of the dictionary stroke.

In step S306, the matching result in the same stroke number matching unit 101 stored in the RAM 204b and each stroke number i in the positive stroke number matching unit 103 are stored.
The optimum matching result is determined from the matching result in. As the determination method in step S306, the character corresponding to the best matching result out of the matching result in the same stroke number matching unit 101 and the matching result in each stroke number i in the positive stroke number matching unit 103 is set as the recognized character. . In addition, a predetermined bias value is added to the matching result of the same stroke number matching unit 101 and the matching result of the positive stroke number matching unit 103, respectively, and the character corresponding to the best matching result is added from the added matching results. You may select as a recognition character.

In general, when a character or a character string is written early, the character is likely to be an abbreviation / continuation character, and the result of the plus stroke number matching unit 103 is likely to be correct. On the other hand, when a character or a character string is written late, it is highly likely that the character was written in a square pattern, and the result of the same stroke count matching unit 101 is likely to be correct. Therefore, in particular, from the matching result obtained by adding a predetermined bias value, in the method of determining the optimal matching result, the number of strokes, the writing time, the writing speed and the like, further using the index indicating the characteristics of the character or the character string, The predetermined bias value may be changed appropriately.

In step S307, the font information and the character code of the character corresponding to the determined matching result are set to R.
It is read from the OM 204a and is output to the display unit 202 as a recognition result, for example. As for the character recognition described in the above-mentioned flow chart, some users often write it in a regular drawing, and some people write a continuous character well. For example, when the result of the plus stroke number matching unit 103 in the embodiment is output as a recognition result to a person who often writes a regular script,
It is often wrong. On the other hand, even if the matching result of the same stroke number matching unit 101 in the embodiment is output as a recognition result to a person who often writes continuous characters, it is often incorrect.

For this reason, the threshold value and the bias value described in the above-mentioned flowchart are displayed on the display unit 202, for example, by displaying the setting menu as shown in FIG. 6 or 7, and the user can freely set the setting menu according to the setting menu. By enabling this, it is possible to further improve the recognition rate of character recognition. In addition, the setting menu,
For example, it can be easily realized by storing it in the ROM 204a and displaying it on the display unit 202 by a user's operation.

In FIG. 6, as the threshold setting item, for example, a “default value” item (default value = 60 in FIG. 6) for setting a default value of the threshold value, and for determining the threshold value in the plus stroke number matching unit 103. An “increment index” item (FIG. 6) for setting an index (for example, the number of strokes, writing time, writing speed) indicating the characteristics of the reference character.
Then, the increment index = the number of strokes), and the “increment frequency” item (increment frequency = 2 in FIG. 6) for setting the number for updating the index value by a predetermined value is provided. A threshold can be set.

On the other hand, as a bias value setting item, a "default value" item (FIG. 6) for setting the default value of the bias value added to the matching result of the same stroke number matching unit 101 and the matching result of the positive stroke number matching unit 103.
Then, the increment frequency is added as a bias value to the matching result of the positive stroke number matching unit 103 and the matching result of the same stroke number matching unit 101 at a ratio of 2: 1), and becomes a reference for determining the threshold value in the positive stroke number matching unit 103. “Increment index” item (increment index = writing speed in FIG. 6) for setting an index (for example, stroke number, writing time, writing speed) indicating the character characteristics, and “increment” set every time the stroke number updates the bias value. By providing the “frequency” item (increment frequency = 0.1 in FIG. 6), the threshold value can be set according to the user's preference.

Alternatively, the setting may be performed using a setting menu as shown in FIG. The setting menu of FIG. 7 is, for example, a setting menu configured as an index of character recognition when a regular writing and a continuous writing are written. FIG.
As shown in, the user can freely set the recognition ratio of the regular characters and the continuous characters by moving the knob 701. In addition, for a person who writes extremely abbreviated characters / successive characters, the user can set the order of activating the same stroke number matching unit 101 and the positive stroke number matching 103 in the first embodiment so that the user can set “Never use a regular character The menu "Don't write" is provided so that the user can check button 702
By checking the
The order of activating 1 and the positive stroke number matching 103 can be exchanged.

6 and 7 are examples of setting menus, and the present invention is not limited to these. For example, in addition to this, various setting menus can be provided according to the use of the user and appropriate setting can be easily performed. As described above, according to the present embodiment, in order to recognize the regular characters and the abbreviations / continuation characters,
Same stroke number matching unit 101 and positive stroke number matching unit 1
03 is provided to determine the matching result between the input stroke and the dictionary stroke, and character recognition is performed by selectively using the same stroke number matching unit 101 and the positive stroke number matching unit 103 based on the determination result. As a result, since it is not necessary to provide a recognition dictionary for abbreviations / continuation characters, it is not necessary to secure a recognition dictionary for abbreviations / continuation characters in the storage medium. Character recognition can be enabled. In addition, since the conditions for recognizing regular characters and abbreviations / successive characters can be set in a timely manner according to the writing characteristics of the user, the recognition rate of character recognition and the processing speed can be improved.

The object of the present invention achieved by the function of the FDD or the function of the method can also be achieved by a storage medium such as an FD in which the program of the above-described embodiment is stored. That is, the storage medium is mounted on the FDD, and the program itself read from the storage medium achieves the novel function of the present invention. The structural features of the program according to the present invention for this purpose are as shown in FIG.

In order to realize the control of this embodiment, F
D is composed of five modules. As shown in the figure, there are five: a dictionary 1011, a counting module 1012, a first comparing module 1013, a second comparing module 1014, and a determining module 10151. According to the modules stored in this storage medium, the processes are executed in the order of steps S1002 to S1005, "counting", "first comparison", "second comparison", and "decision" as shown in FIG. “Count” executed by the counting module 1012 in the modules stored in each storage medium
(Step S1002) The process corresponds to step S301 in the flowchart of FIG.

The "first comparison" (step S1003) processing executed by the first comparison module 1013 corresponds to step S302 in the flowchart of FIG. The “second comparison” (step S1004) process executed by the second comparison module 1014 corresponds to steps S303 to S305 in the flowchart of FIG. “Determination” executed by the determination module 1015 (step S100
5) The process is step S30 in the flowchart of FIG.
6, corresponding to step S307.

The present invention may be applied to a system composed of a plurality of devices or an apparatus composed of a single device. Further, it goes without saying that the present invention can be applied to the case where it is implemented by supplying a program to a system or an apparatus. In this case, the storage medium storing the program according to the present invention constitutes the present invention. Then, by reading the program from the storage medium to the system or device, the system or device operates in a predetermined manner.

[0040]

As is apparent from the above description, according to the present invention, the storage capacity and the recognition time are not increased.
It is possible to provide a character recognizing device, a method thereof, and a computer control device capable of recognizing regular characters and abbreviations / successive characters.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a character recognition device of the present embodiment.

FIG. 2 is a functional block diagram showing a configuration of functions executed by control of a CPU 203 according to the present embodiment.

FIG. 3 is a flowchart showing a processing flow of the present embodiment.

FIG. 4 is a plus stroke number matching unit 103 according to the present embodiment.
4 is a diagram for explaining a matching method in FIG.

FIG. 5 is a diagram for explaining a matching method in a plus stroke number matching unit 103 according to another configuration of the present embodiment.

FIG. 6 is a diagram showing a configuration of a setting menu according to the present embodiment.

FIG. 7 is a diagram showing a configuration of a setting menu according to another configuration of the present embodiment.

FIG. 8 is a diagram showing a structure of a memory map of an FD in which a program of the processing flow of this embodiment is stored.

[Explanation of symbols]

 201 coordinate input unit 201a digitizer 201b pen 202 display unit 203 CPU 204 memory 204a ROM 204b RAM 205 FDD 206 HDD 206a recognition dictionary

Claims (21)

[Claims] 1. A character recognition device for recognizing a character on the basis of the characteristic of an input stroke, a dictionary for storing the characteristic of the stroke constituting a standard character, and a counting for counting the number of strokes of the input stroke. Means for comparing the input stroke feature with the stroke feature stored in the dictionary and having the same number of strokes as the number of strokes counted by the counting means, and the result of the first comparing means matches. When the ratio is lower than a predetermined value, a second comparison unit that compares the input stroke feature with a stroke feature that has a greater number of strokes than the same number of strokes stored in the dictionary, and a first comparison unit. A character recognizing device comprising: a determining unit that determines a character formed by an input stroke based on a result of the second comparing unit. 2. The second comparing means is obtained every time when the predetermined number is added to the stroke feature stored in the dictionary and having the stroke number obtained every time the predetermined number is added to the stroke number. The character recognition device according to claim 1, wherein the comparison is performed sequentially until the number of strokes reaches a predetermined number of strokes. 3. The first comparison means compares the input stroke feature with all stroke features stored in the dictionary and having the same number of strokes as the number of strokes counted by the counting means. The character recognition device according to claim 1. 4. The character recognition device according to claim 1, wherein the dictionary is a dictionary that stores characteristics of strokes that form a standard character according to a standard stroke order of the standard character. 5. The first comparing means, up to a predetermined order of the input order of the input strokes, has the same number as the feature of the input strokes and the number of strokes counted by the counting means stored in the dictionary. And the stroke features in the same order as the input sequence are compared, and after the predetermined sequence, the stroke features having the same number as the stroke feature counted in the dictionary and the feature of the input stroke are stored in the dictionary. The character recognition device according to any one of claims 1 to 4, wherein the character recognition device is compared with all the features described in (1) to (4). 6. The character according to claim 1, wherein the predetermined value is a value obtained by a function having at least one of the stroke number of the input stroke, an input time, and an input speed as a variable. Recognition device. 7. The character recognition device according to claim 6, wherein the function is an increasing function. 8. The character corresponding to the comparison object having a highest ratio of matching with the feature of the stroke of the dictionary compared by the first comparing device and the second comparing device. The character recognition device according to claim 1, wherein is determined as a recognition character. 9. The deciding means further comprises an adding means for adding a predetermined bias value to each of the comparison results obtained by the first comparing means and the second comparing means, and a new one obtained by the adding means. The character recognition device according to claim 1, wherein the character formed by the input stroke is determined based on the result. 10. The bias value according to claim 9, wherein the predetermined bias value is a bias value determined based on at least one of the stroke number of the input stroke, the input time, and the input speed. Character recognizer. 11. A character recognition method for recognizing a character based on a characteristic of an input stroke, the counting step of counting the number of strokes of the input stroke, the characteristic of the input stroke and a standard character. A first comparison step of comparing the stroke count of the same number as the stroke count stored in the dictionary storing the features of the constituent strokes, and as a result of the first comparison step, the matching rate is a predetermined value. If it is lower than the above, the second comparison step of comparing the characteristics of the input stroke with the characteristics of the stroke having the number of strokes larger than the same number of strokes stored in the dictionary, the first comparison step and the second comparison step. And a determination step of determining a character formed by the input stroke, based on the result of 1. 12. The second comparing step is obtained every time when the predetermined number is added to the stroke feature stored in the dictionary of the stroke number obtained every time when the predetermined number is added to the stroke number. The character recognition method according to claim 11, wherein the comparison is sequentially performed until the number of strokes reaches a predetermined number of strokes. 13. The first comparison step is characterized by comparing the input stroke feature with all stroke features stored in the dictionary and having the same number of strokes as the number of strokes counted in the counting step. The character recognition method according to claim 11. 14. The character recognition method according to claim 11, wherein the dictionary is a dictionary that stores the characteristics of strokes forming the standard character in accordance with the standard stroke order of the standard character. 15. In the first comparing step, up to a predetermined order of the input order of the input strokes, the same number of stroke features as the input strokes and the number of strokes counted in the counting step are stored in the dictionary. And the stroke features in the same order as the input sequence are compared, and after the predetermined sequence, the stroke features having the same number as the stroke feature counted in the counting step and the feature of the input stroke are stored. 15. The character recognition method according to claim 11, wherein the character recognition method is compared with all of the characteristics of. 16. The character according to claim 11, wherein the predetermined value is a value obtained by a function having at least one of the stroke number of the input stroke, an input time, and an input speed as a variable. Recognition method. 17. The character recognition method according to claim 16, wherein the function is an increasing function. 18. The character corresponding to the comparison object having the highest ratio of matching with the feature of the stroke of the dictionary compared in the first comparing step and the second comparing step in the determining step. 12. The character recognition method according to claim 11, wherein is determined as a recognition character. 19. The determination step further comprises an addition step of adding a predetermined bias value to each of the comparison results obtained in the first comparison step and the second comparison step, and a new value obtained in the addition step is added. The character recognition method according to claim 11, wherein the character formed by the input stroke is determined based on the result. 20. The bias value according to claim 19, wherein the predetermined bias value is a bias value determined based on at least one of the number of strokes of the input stroke, the input time, and the input speed. Character recognition method. 21. A computer control device for controlling a computer by reading a predetermined program from a memory medium, wherein the memory medium stores a dictionary that stores characteristics of strokes that form a standard character, and a dictionary of the input strokes. A procedure code of a counting step for counting the number of strokes, and a procedure code of a first comparing step for comparing the feature of the input stroke with the feature of the same number of strokes stored in the dictionary as the stroke number counted in the counting step. And, as a result of the first comparing step, when the matching ratio is lower than a predetermined value, the feature of the input stroke is compared with the feature of the stroke having the stroke number larger than the same number of strokes stored in the dictionary. A sentence formed by the input stroke based on the procedure code of the second comparison process and the results of the first comparison process and the second comparison process. Computer controller, characterized in that it comprises a procedure code determining step of determining.