CN1155127A - Character sign recognition device and method thereof, and computor controlling device - Google Patents

Character sign recognition device and method thereof, and computor controlling device Download PDF

Info

Publication number
CN1155127A
CN1155127A CN96112731A CN96112731A CN1155127A CN 1155127 A CN1155127 A CN 1155127A CN 96112731 A CN96112731 A CN 96112731A CN 96112731 A CN96112731 A CN 96112731A CN 1155127 A CN1155127 A CN 1155127A
Authority
CN
China
Prior art keywords
stroke
feature
input
character
dictionary
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN96112731A
Other languages
Chinese (zh)
Other versions
CN1094221C (en
Inventor
吉井裕人
新井常一
高须英司
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Canon Inc
Original Assignee
Canon Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Canon Inc filed Critical Canon Inc
Publication of CN1155127A publication Critical patent/CN1155127A/en
Application granted granted Critical
Publication of CN1094221C publication Critical patent/CN1094221C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/048Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0487Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser
    • G06F3/0488Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures
    • G06F3/04883Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures for inputting data by handwriting, e.g. gesture or text

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Character Discrimination (AREA)

Abstract

To provide a character recognition device, its method and a computer controller which recognize character without increasing a storage capacity and a recognizing time. 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

Character recognition device and method and Computer Control Unit
The present invention relates to a kind of character recognition device and method, and a kind of Computer Control Unit of feature identification character of the stroke based on input.
So far, in the character recognition of coordinate entering device of using such as digitizer etc. to input, such character recognition device is arranged, and it can discern the hand-written character such as conitnuous forms character or squeezing characters that is not written (print hand) character (conventional Writing method).In a kind of like this device, except character information based on conventional Writing method, the character information of conitnuous forms character and squeezing characters is stored in the dictionary that is used to discern, this dictionary is stored in storage medium etc., and the hand-written character information of input is compared with the character information in being stored in dictionary, thereby the character of identification input.
Such as, the feature of the pattern 1021 of the conitnuous forms character by keeping character pattern 1020 is in the four-stroke dictionary, wherein character pattern 1020 has been write five strokes when writing with written characters, in the four-stroke dictionary, include the character pattern 1010 that the written characters of not write continuously more than two strokes is arranged as shown in figure 10, even stroke (3) is write as character pattern 1021 continuously with (4), and input character pattern 1010 as the stroke (3) that resembles character pattern 1021 ' four-stroke conitnuous forms character, may export with character pattern 1020 identical characters coding as recognition result.
Also has a kind of method, do not reduce recognition speed in order to discern squeezing characters and conitnuous forms character, prepare to be used for the dictionary and the dictionary that is used for squeezing characters or conitnuous forms character (unconventional Writing method) of written characters (conventional Writing method) respectively, the character that at first is used for the dictionary coupling input of written characters, when recognition result is bad, the character of input and the character match that is stored in the dictionary that is used for squeezing characters or conitnuous forms character.
Yet, in order to discern all squeezing characters or conitnuous forms character, when such formation, promptly at the character of dictionary stored corresponding to all squeezing characters or conitnuous forms character, compare with the dictionary that only is used for written characters, the memory space of this dictionary increases significantly.In addition, need be complementary with the character of a various modes and an input, thereby cause such problem, promptly recognition time is long, and the processing speed that identification is handled descends.Especially, under the situation of the online hand-written character of identification, the possibility of input written characters and the possibility of input squeezing characters or conitnuous forms character are almost equal, then have such problem, when one of preferential selection said two devices, recognition rate reduces significantly.
Consider that the problems referred to above make the present invention, an object of the present invention is to provide a character recognition device and method thereof and a kind of Computer Control Unit, it can be discerned written characters and squeezing characters or conitnuous forms character and not increase memory space and recognition time.
Fig. 1 is a block scheme, shows a kind of structure of the character recognition device of an embodiment;
Fig. 2 is a functional-block diagram, shows the performed functional structure of control of CPU203 in the present embodiment;
Fig. 3 is a process flow diagram, shows a treatment scheme of present embodiment;
Fig. 4 A to 4C is a synoptic diagram, is used to illustrate the matching process in the stroke number matching unit 103 of adding at present embodiment;
Fig. 5 A to 5C is a synoptic diagram, is used to illustrate at the matching process in the stroke number matching unit 103 of adding according to another structure of present embodiment;
Fig. 6 is a synoptic diagram, shows the structure that one in the present embodiment set menu;
Fig. 7 is a synoptic diagram, shows the structure of setting menu according to one of another structure of present embodiment;
Fig. 8 is a synoptic diagram, shows the structure of the storage map of a FD, and wherein the procedure stores of the treatment scheme of present embodiment is in this FD;
Fig. 9 shows the example of the adaptable alphabetic character of present embodiment; And
Figure 10 is the explanatory view that a written characters pattern and a squeezing characters or form character pattern continuously.
Below in conjunction with accompanying drawing, describe a most preferred embodiment of the present invention in detail.
Fig. 1 is a block scheme, shows a kind of structure of the character recognition device of present embodiment;
Label 201 expressions are by a digitizer 201a and the coordinate input equipment that 201b constitutes.By a 201b, the coordinate data of character or figure can input to digitizer 201a.By the control of CPU203, display 202 can show the coordinate data of this input.Label 202 expression comprises the display such as CRT etc.By the control of CPU203, display 202 shows a character of input equipments 201 inputs or the coordinate of figure, or the character data that reads from storer 204.
Label 205 expression one FDD (floppy disk driver).FD is loaded into FDD and data can write or read from FD to FD.Also can write a FD (not illustrating) who loads to a program of a treatment scheme (Fig. 3) that will illustrate below, and carry out a processing by the RAM204b that this program is read in this device.In embodiment 1, this procedure stores is in ROM204a, and the treatment scheme (Fig. 3) that will illustrate below carrying out by the control of CPU203.The present invention is not limited to FDD205, but can carry out with above-mentioned identity function, as long as it is a driver, be loaded with such as CD-ROM to it, the outer storage medium of MO etc., and can the read and write data.
Label 206 expression one HDD (hard disk drive), it has the dictionary 206a that is used to discern as following of will illustrate, is used to discern a character from digitizer 201a input.The dictionary 206a that is used to discern in the storage medium stored that can store mass data such as HD (hard disk) etc.
Label 203 expressions one CPU is used to discern the character of input and carries out various controls.The storer that label 204 expressions one are made of ROM204a and RAM204b.Store at ROM204a and to be used for recognizer and the various control program that CPU203 carries out character recognition.The font information of character, character code etc. also are stored in the ROM204a.RAM206b is such zone, be used for storing provisionally the data that the dictionary stored from the HD (not illustrating) that is contained in the HDD206 is read, and the coordinate data and the employed data such as variable of recognizer and various datamations district of storage input.
Also can constitute by this way, promptly CPU203 moves required program in the present embodiment, and dictionary and various parameter are transfused to by a communication line (not illustrating) and are stored in RAM204b, and carry out a process.
Fig. 2 is a functional-block diagram, shows to be controlled the structure of the function of carrying out according to present embodiment by CPU203.
By the control of CPU203, at first activate matching unit 101 (same stroke number).As character data from digitizer 201a input, the stroke number that constitutes input character by on the 201b/following information is counted.Carry out the coupling of the dictionary data of input character data and the stroke number identical with the counting stroke data.After coupling is finished, activate coupling authentication unit 102, and the quality of checking matching result.When matching result is good, activates identification adjustment unit 104, and export one and adjust the result as a recognition result.When matching result is bad, activate one add stroke number matching unit (different strokes number) 103 (below, abbreviate matching unit 103 as), and order carries out the dictionary data and the coupling of input character data until each stroke number of predetermined stroke number, and wherein predetermined stroke number is equal to or greater than the stroke (below be called the input stroke) of input and counts.Finish to the coupling of predetermined stroke number, activate identification adjustment unit 104, and from the stroke number of input stroke, determine best matching result in the matching result of each stroke number of predetermined stroke number, and export the best matching result that this is determined, as recognition result.
The treatment scheme of present embodiment will be described in conjunction with the flow process of Fig. 3.
Fig. 3 is a process flow diagram, shows the treatment scheme of embodiment.
At step S300,, comprise that a character of certain stroke number inputs to digitizer 201a by using a 201B.At step S301, the control of CPU203 activates matching unit 101, thus based on stroke, partial mode, character or character string, coupling is as the stroke information and the stroke information that remains on a dictionary (following title dictionary stroke) on the unit of input stroke feature.
As matching process, mode is mated the order of importing stroke and the order that remains on the dictionary stroke of dictionary 206a one to one, perhaps can be when not considering those orders, and corresponding whole strokes are carried out coupling.
The dictionary stroke of the stroke number identical with the input stroke number remains on by in the matching unit 101 employed dictionary 206a, also can store the stroke information of same stroke number and the stroke information of formation written characters (conventional Writing method), such as, constitute the very high squeezing characters of frequency of utilization or the stroke information of conitnuous forms character (non-regular Writing method).In this case, owing in dictionary 206a, comprise the stroke information of squeezing characters or conitnuous forms character, at this moment, can keep certain speed to the discrimination of squeezing characters or conitnuous forms character.
At step S302, when the coupling of matching unit 101 is finished, activate coupling authentication unit 102, thereby judge the matching result of matching unit 101.When definite matching result good (step S302 for " be ") time, matching result is stored in RAM204b, enters step S307 then.When judging matching result bad (be at step S302 " or not), matching result is stored in RAM204b, enters step S303 then.
As a kind of determination methods of matching result, such as, whether judge matching result by inspection as the relative possibility of the candidate characters of matching unit 101 results' the character that will discern greater than a predetermined threshold value.Such as, in RAM204b, set storage one such as 80% threshold value as a parameter, when the relative possibility of candidate characters is equal to or greater than 80%, judge that matching result is for well.Because usually squeezing characters or conitnuous forms character are write soon than written characters, can constitute by this way, promptly when the input stroke number is big, threshold setting is a value greatly, when importing stroke a few hours, threshold setting is a little value.In this case, also can constitute by this way, promptly in the parameter storage unit of RAM204b, store the threshold value of each stroke number, and when judging, use with forms mode.Value in the form can be upgraded by the user.
Also have, such as, can determine this threshold value by following linear function.
Threshold value=60+2 * (stroke number of input)
In this case, when character more than or equal to 20 strokes of one of input, can determine unconditionally that the matching result of matching unit 101 is bad.
Pass through above-mentioned equation shown in the present invention is not limited to, be used to import the method that the property line function of stroke number comes setting threshold,, can use any increasing function (being not restricted to monotonically increasing function) for stroke number.
Except the input stroke number, by using the writing time of input stroke, the writing speed of input stroke etc. obtain aforesaid threshold values as this function parameters, because of the writing time of input stroke or the threshold value that writing speed changes can easily be set.
When the matching result of step S302 is bad, a kind of like this possibility is arranged, promptly stroke is by squeezing characters or the input of conitnuous forms character.Therefore, in the treatment step of following, stroke number increases in the step of coupling target, and carries out identification and handle.
At step S303, a stroke is added to the number (i) of input stroke, thereby upgrades stroke number (i) (i is a positive integer).At step S304, (n is an integer: n>i) to the stroke number (i) that check to upgrade greater than a predetermined stroke number (n).As a value of (n), use the parameter that is stored in the RAM204b.When stroke number (i) greater than predetermined stroke number (n) (step S304 for " be ") time, enter step S306.When stroke number (i) be less than or equal to predetermined stroke number (n) (step S304 for " or not) time, enter step S305.At step S305, activate matching unit 103, based on stroke, partial mode, a character or an alphabetic character string, coupling has the dictionary stroke that remains on dictionary 206a of identical stroke number with renewal stroke number (i).Repeated execution of steps S303 to S305 reaches to predetermined stroke number (n) until stroke number (i).Each matching result that upgrades each stroke number (i) gained sequentially is stored in RAM204b.
Such as, with to coupling authentication unit 102 setting thresholds similar mode, the upper limit of predetermined stroke number (n) is confirmed as importing the stroke number of stroke, a function of writing time or writing speed.
Such as, matching process as the matching unit in step S305 103, can with mode one to one mate input stroke order and remain on the order of dictionary stroke in the dictionary 206a, perhaps they also can be corresponding other and do not consider these orders and mate.Particularly,, and not have significantly when compressing the trend of which stroke, wish to carry out the following a kind of matching process that will will describe 4C in conjunction with 4A when input one squeezing characters or a conitnuous forms character.
Fig. 4 A to 4C is a synoptic diagram, is used to illustrate the matching process in the matching unit 103 of present embodiment.
As the example that is used to illustrate embodiment that will describe below one, use just like the dictionary mode shown in the character alternate figures 4B who is write continuously shown in Fig. 4 A with two strokes.Yet,, also can carry out similar processing for the pattern 91 of pattern in the dictionary as shown in Figure 9 92 and input.
Such as, the character that the input stroke of being write by a 201b constitutes is just like the conitnuous forms character shown in Fig. 4 A, and input pattern for a positional structure with a kind of like this character and sequential write (1 to 4), corresponding to a dictionary mode, carry out coupling, wherein this dictionary mode has a positional structure and the sequential write of the dictionary stroke (1 to 5) of written characters shown in the pie graph 4b.As a kind of matching process in this case, as shown,, carry out coupling respectively according to Fig. 4 B dictionary stroke ([1] is to [5]) by a target of each the straight line coupling in Fig. 4 C for each input sequence ([1] is to [4]) of input stroke.Label shown in Fig. 4 C represents that input pen draws the input sequence of input sequence and dictionary stroke.And show a combination mating by a line segment that connects the numeral of expression one stroke.In this case, showed a kind of matching process of between each input order of strokes and all dictionary order of strokes, carrying out coupling.
Usually, hypothesis has a kind of like this trend now, promptly Shu Ru stroke is compressed, wherein input in the stroke the character that constitutes by the input stroke or the input sequence of character string be anti-order, can carry out coupling in such a way, such as, input stroke to a minimum input sequence shown in Fig. 5 C, first and second strokes of promptly importing stroke respectively with the dictionary stroke in the same stroke number of first and second strokes be complementary, and third and fourth stroke of the back of input stroke and the corresponding stroke number of dictionary stroke are complementary.
At step S306, the matching result of matching unit 101 and the matching result of each stroke number (I) of matching unit 103, determine best matching result in being stored in RAM204b.
As a kind of definite method in step S306, be set to an identification character corresponding to the character of the best matching result of the matching result of each stroke number (i) of the matching result of matching unit 101 and matching unit 103.Also can constitute in such a way, promptly a predetermined inclined to one side value adds to the matching result of matching unit 101 and the matching result of matching unit 103 respectively, and based on the addition matching result of gained, can also select character, as an identification character corresponding to best matching result.
Usually, when writing a character or a character string fast, very likely be a squeezing characters or conitnuous forms character, that is, as matching result, the correctness height that the result of matching unit 103 and squeezing characters or conitnuous forms character are complementary.On the contrary, when writing a character or a character string at a slow speed, character is the possibility height of writing with written characters, and promptly the result of matching unit 101 is correct possibility height.Therefore, from matching result, determining in the method for best matching result, by using an indication such as stroke number, writing time, a traction of the character of writing speed etc. or character string feature, can suitably change predetermined value partially, wherein matching result obtains by adding predetermined value partially.
At step S307, from ROM204a, read font information and character code, and export to as a recognition result corresponding to the character of determined best matching result, such as, display 202.
In the illustrated character recognition of above-mentioned flow process, the user of a written characters written character commonly used or the user of a conitnuous forms character written character commonly used are arranged.Such as, if the result of matching unit in the present embodiment 103 exports the user who often writes written characters to as a recognition result, mistake will repeatedly appear.On the other hand, if the matching result of matching unit in the present embodiment 101 exports the user who often writes the conitnuous forms character to as a recognition result, also mistake will appear repeatedly.
Therefore, for threshold value illustrated in above-mentioned flow process and inclined to one side value, a use 201b etc. imports an instruction by the coordinate input equipment user, thereby allows, such as, on display 202, show the setting menu shown in Fig. 6 or 7.It is to constitute like this, and promptly according to the setting menu that shows, the user at random parameter storage unit in RAM204b resets threshold value or inclined to one side value.Make in this way, improved the recognition speed of character recognition, further meet user's requirement.Can easily realize setting menu by following formation, i.e. the procedure stores of Ji Huoing, such as, be presented on the display 202 at ROM204a and the operation by the user.
Among Fig. 6, setting item as threshold value, by providing a " default value " project (in Fig. 6, default value is 60) with a default value of setting threshold, one " increases progressively traction " project (in Fig. 6, increase progressively traction and be stroke number) draw to set one of a pointing character feature (such as, stroke number, writing time, writing speed), this character feature is as the reference of the threshold value of a definite matching unit 103, and a " increases progressively unit " project (among Fig. 6, increasing progressively unit is 2), be used for setting the number that upgrades the traction value by the predetermined value in a time.Threshold value can be set according to user's taste.
On the other hand, setting item for inclined to one side value, by providing a " default value " project (in Fig. 6, add to the matching result of the matching result of matching unit 103 and matching unit 101 with 2: 1 ratios and to increase progressively unit, as inclined to one side value) to set a default value of inclined to one side value, the default value of this inclined to one side value is added to the matching result of matching unit 101 and the matching result of matching unit 103; One " increases progressively traction " project (in Fig. 6, increase progressively traction and be writing speed) draw to set one of a pointing character feature (such as, stroke number, writing time, writing speed), this character feature is as the reference of the threshold value of a definite matching unit 103, and a " increases progressively unit " project (among Fig. 6, increasing progressively unit is 0.1), be used for when upgrading stroke number, setting value partially at every turn.Threshold value can be set according to user's taste.
Can also use setting menu setting threshold shown in Figure 7.
The setting menu of Fig. 7 is the setting menu that constitutes in such a way, such as, set the situation of the character of writing by written characters and the situation of the character write by the conitnuous forms character as the traction of character recognition.As shown in Figure 7, by operating a button 701, the user can at random set the ratio between written characters and the conitnuous forms character.For the user who often writes squeezing characters or conitnuous forms character, the menu that provides a " never to write written characters ", thus the user can set a kind of pattern, with the activation sequence of matching unit 101 in the change present embodiment and 103.By checking check key 702, the user can change the activation sequence of matching unit 101 and 103.
Fig. 6 and 7 shows the example of setting menu, but the present invention is not limited to this.Such as, can provide various setting menus according to user's application, and be easy to correctly set them.
According to above-mentioned present embodiment, in order to discern written characters and squeezing characters or conitnuous forms character, matching unit (same stroke number) 101 and matching unit (different strokes number) 103 is provided, judge the matching result of stroke in input stroke and the dictionary, and the matching unit 101 and 103 that selection will be used based on judged result, thereby identification character.Like this, owing to do not need to be provided for the recognition dictionary of squeezing characters or conitnuous forms character, the recognition dictionary that only is provided for written characters just can be discerned written characters and squeezing characters or conitnuous forms character, and does not need to be provided for the recognition dictionary of squeezing characters or conitnuous forms character.Because the characteristics according to user's ways of writing can be set the condition for identification of written characters and squeezing characters or conitnuous forms character suitably, thereby can improve the recognition rate and the processing speed of character recognition.
The purpose of the present invention that function or method realized by FDD can realize by the storage medium such as a FD etc., store the program that realizes above-mentioned present embodiment in this storage medium.Just, read from storage medium and realized a kind of new function of the present invention because storage medium is connected to FDD and program itself.Of the present invention a kind of architectural feature according to this purpose is shown in Fig. 8.
In order to realize the control of present embodiment, FD is made of five modules, promptly as shown in the figure, and dictionary 1011, counting module 1012, the first comparison modules, 1013, the second comparison modules 1014 and a determination module 1015.According to being stored in the storage medium inner module, carry out processing according to the order of step S1002 to S1005 shown in Figure 8, " counts ", and " first is " relatively, and " second compares " and " determines ".Be stored in each module in the storage medium, in counting module 1012, carry out " counting " (step S1002) corresponding to the step S301 in the flow process of Fig. 3.
The " first that carries out in first comparison module 1013 relatively " (step S1003) handles corresponding to the step S302 in the flow process of Fig. 3.The " second that carries out in second comparison module 1014 relatively " (step S1004) handles corresponding to the step S303 to S305 in the flow process of Fig. 3.The " that carries out in determination module 1015 determines that " (step S1005) handles corresponding to step S306 in the flow process of Fig. 3 and S307.
The present invention can be applied to the system that is made of a plurality of equipment, also can be applied to comprise the device of an equipment.Clearly the present invention can embody situation of the present invention by program to a system or a device are provided.In this case, the storage medium that stores according to program of the present invention constitutes the present invention.By from the storage medium read routine to system or device, by predetermined method operating system or device.

Claims (45)

1. a character recognition device is used for feature identification one character based on the stroke of input, comprising:
One dictionary is used to store the stroke feature that constitutes standard character;
Counting assembly is used to count the described stroke number of input;
First comparison means is used for the feature of the stroke of the identical standard character of the feature of the described stroke of input the stroke number with described counting assembly counting interior with being stored in described dictionary is compared;
Second comparison means, when reason owing to the described first comparison means result, when coincidence rate is lower than a predetermined value, be used for the feature of the described stroke of input is compared greater than the feature of the described stroke with identical stroke number with the stroke number in being stored in described dictionary; And
Determine device,, be used for a character of the described stroke formation of definite input based on the result of described first and second comparison means.
2. according to a kind of device of claim 1, it is characterized in that, described second comparison means is being stored in the described dictionary, the feature of resultant stroke was compared with the feature of the described stroke of input when each predetermined number was added to described stroke number, and repeated described comparison until reaching a predetermined stroke number by increasing the resulting stroke number of described predetermined number.
3. according to a kind of device of claim 1, it is characterized in that described first comparison means is compared the feature of the stroke of all identical standard characters of the feature of the described stroke of input the stroke number with described counting assembly counting interior with being stored in described dictionary;
4. according to a kind of device of claim 1, it is characterized in that according to the standard sequential write of standard character, described dictionary storage constitutes the feature of the stroke of described standard character.
5. according to a kind of device of claim 1, it is characterized in that, first comparison means is compared the feature of the feature of the described stroke of input the stroke with following properties interior with being stored in described dictionary, the stroke that is stored in the described dictionary has the identical stroke number of counting with described counting assembly of stroke number, and have described the enter order identical with the predefined procedure that enters order of the described stroke of extremely importing, and behind described predefined procedure, described first comparison means the feature of the described stroke of input be stored in interior all features of described dictionary and compare with stroke identical stroke number that described counting assembly is counted.
6. according to a kind of device of claim 1, it is characterized in that described predetermined value is such value, it obtains by a function, and the described stroke number of input is used as a variable at least in this function.
7. according to a kind of device of claim 1, it is characterized in that described predetermined value is such value, it obtains by a function, is used as a variable in this function at least the input time of the described stroke of input.
8. according to a kind of device of claim 1, it is characterized in that described predetermined value is such value, it obtains by a function, and the input speed of the described stroke of input is used as a variable at least in this function.
9. according to a kind of device of claim 1, it is characterized in that described predetermined value is such value, it obtains by an increasing function, and the information of the described stroke of importing in this increasing function is used as a variable.
10. according to a kind of device of claim 1, it is characterized in that, as a recognition result, described definite device is determined corresponding to a character that has the comparison object of ceiling rate in all ratios, in these ratios, the feature of the described stroke of input is consistent with the feature of the stroke of the comparison object that is used as described first and second comparison means in described dictionary.
11. a kind of device according to claim 1 is characterized in that,
Described definite device has adding device, is used for a predetermined inclined to one side value is added to respectively by resulting each comparative result of described first and second comparison means; And
Based on the new result that described adding device obtains, described definite device is determined the character that the described stroke of input constitutes.
12. according to a kind of device of claim 11, also have display device, be used to show that one sets menu, to set described inclined to one side value.
13. according to a kind of device of claim 1, also have display device, when judging whether described second comparison means carries out comparison, show that one sets menu, to set a predetermined value.
14., also have a graphic tablet, to import described stroke according to a kind of device of claim 1.
15. a kind of device according to claim 1 is characterized in that, described stroke is an information of lifting pen from starting to write.
16. a character identifying method is used for feature identification one character based on the stroke of input, comprising:
A counting step is used to count the number of the described stroke of input;
First comparison step, the feature of the stroke of the identical standard character of the feature of the described stroke of input the stroke number with described counting step counting interior with being stored in described dictionary is compared, in described counting step, the feature that constitutes the stroke of standard character is stored;
Second comparison step, when reason owing to described first comparison step, when coincidence rate is lower than a predetermined value, be used for the feature of the described stroke of input is compared greater than the feature of the stroke of the described stroke with identical stroke number with the stroke number in being stored in described dictionary; And
One determining step based on the result of described first and second comparison step, is used for a character of determining that described stroke constitutes.
17. a kind of method according to claim 16, it is characterized in that, in described second comparison step, in fact being stored in the described dictionary, the feature of resultant stroke was compared with the feature of the described stroke of input when each predetermined number was added to described stroke number, and repeated described comparison step until reaching a predetermined stroke number by increasing the resulting stroke number of described predetermined number.
18. a kind of method according to claim 16, it is characterized in that, in described first comparison step, the feature of the described stroke of input be stored in the described dictionary with described counting step in the feature of stroke of identical all standard characters of the stroke number counted compare;
19. a kind of method according to claim 16 is characterized in that, according to the standard sequential write of described standard character, described dictionary storage constitutes the feature of the stroke of described standard character.
20. a kind of method according to claim 16, it is characterized in that, in described first comparison step, the feature of the stroke with following properties that the feature of the described stroke of input is interior with being stored in described dictionary is compared, the stroke that is stored in the described dictionary has the identical stroke number of counting with described counting step, and have described the enter order identical with the predefined procedure that enters order of the described stroke of extremely importing, and behind described predefined procedure, the feature of the described stroke of input be stored in all features with stroke identical stroke number that described counting step is counted in the described dictionary and compare.
The 21 a kind of methods according to claim 16 is characterized in that, described predetermined value is such value, and it obtains by a function, and the described stroke number of input is used as a variable at least in this function.
22. a kind of method according to claim 16 is characterized in that, described predetermined value is such value, and it obtains by a function, is used as a variable in this function at least the input time of the described stroke of input.
23. a kind of method according to claim 16 is characterized in that, described predetermined value is such value, and it obtains by a function, and the input speed of the described stroke of input is used as a variable at least in this function.
24. a kind of method according to claim 16 is characterized in that, described predetermined value is such value, and it obtains by an increasing function, and the information of the described stroke of importing in this increasing function is used as a variable.
25. according to claim 16 method, it is characterized in that, in described definite device, determine that character corresponding to a comparison object that has ceiling rate in all ratios is as a recognition result, in these ratios, the feature of the described stroke of input is consistent with the feature of the stroke of the comparison object that is used as described first and second comparison means in described dictionary.
26. a kind of method according to claim 16 is characterized in that,
Described determining step has an addition step, is used for a predetermined inclined to one side value is added to respectively by resulting each comparative result of described first and second comparison step; And
Based on the new result that described addition step obtains, determine a character of the described stroke formation of input.
27. according to a kind of method of claim 26, also have a step display, be used to show that one sets menu, to set described inclined to one side value.
28. according to claim 16 method, also have step display, when judging whether described second comparison step carries out comparison, show that one sets menu, to set a predetermined value.
29., also have a step of importing described stroke by graphic tablet according to claim 16 method.
30., it is characterized in that described stroke is an information of lifting pen from starting to write according to claim 16 method.
31. a storage medium is used to be stored in the program of carrying out in the Computer Control Unit, this Computer Control Unit is from described storage medium fetch program and control computer, and described storage medium comprises:
One is used to count the process coding of the described stroke number of input;
One is used for the process that the feature of the stroke of the identical standard character of the feature of the described stroke of input the stroke number with described counting step counting interior with being stored in described dictionary is compared is encoded, in described counting step, the feature that constitutes the stroke of standard character is stored;
One is used for when the reason owing to described first comparison step, when coincidence rate is lower than a predetermined value, the process that the feature of the described stroke of input and the stroke number that is stored in the described dictionary are compared greater than the feature of the stroke of the described stroke with identical stroke number is encoded; And
One is used for the result according to described first and second comparison step, the process coding of the character that the described stroke of definite input constitutes.
32. a kind of storage medium according to claim 31, it is characterized in that, in described second comparison step, in fact being stored in the described dictionary, the feature of resultant stroke was compared with the feature of the described stroke of input when each predetermined number was added to described stroke number, and repeated described comparison step until reaching a predetermined stroke number by increasing the resulting stroke number of described predetermined number.
33. a kind of storage medium according to claim 31, it is characterized in that, in described first comparison step, the feature of the described stroke of input be stored in the described dictionary with described counting step in the feature of stroke of identical all standard characters of the stroke number counted compare;
34. a kind of storage medium according to claim 31 is characterized in that, according to the standard sequential write of described standard character, described dictionary storage constitutes the feature of the stroke of described standard character.
35. a kind of storage medium according to claim 31, it is characterized in that, in described first comparison step, the feature of the stroke with following properties that the feature of the described stroke of input is interior with being stored in described dictionary is compared, the stroke that is stored in the described dictionary has the identical stroke number of counting with described counting step, and have described the enter order identical with the predefined procedure that enters order of the described stroke of extremely importing, and behind described predefined procedure, the feature of the described stroke of input be stored in all features with stroke identical stroke number that described counting step is counted in the described dictionary and compare.
36. a kind of storage medium according to claim 31 is characterized in that, described predetermined value is such value, and it obtains by a function, and the described stroke number of input is used as a variable at least in this function.
37. a kind of storage medium according to claim 31 is characterized in that, described predetermined value is such value, and it obtains by a function, is used as a variable in this function at least the input time of the described stroke of input.
38. a kind of storage medium according to claim 31 is characterized in that, described predetermined value is such value, and it obtains by a function, and the input speed of the described stroke of input is used as a variable at least in this function.
39. a kind of storage medium according to claim 31 is characterized in that, described predetermined value is such value, and it obtains by an increasing function, and the information of the described stroke of importing in this increasing function is used as a variable.
40. according to claim 31 storage medium, it is characterized in that, in described definite device, determine that character corresponding to a comparison object that has ceiling rate in all ratios is as a recognition result, in these ratios, the feature of the described stroke of input is consistent with the feature of the stroke of the comparison object that is used as described first and second comparison means in described dictionary.
41. a kind of storage medium according to claim 31 is characterized in that,
Described determining step has an addition step, is used for a predetermined inclined to one side value is added to respectively by resulting each comparative result of described first and second comparison step; And
Based on the new result that described addition step obtains, determine a character of the described stroke formation of input.
42. according to a kind of storage medium of claim 41, also have the process coding of a step display, show that one sets menu, to set described inclined to one side value.
43. according to claim 31 storage medium, also have a process coding of a step display,, show a setting menu, to set a predetermined value when the whether execution of judging described second comparison step relatively the time.
44., also have a process coding of importing the step of described stroke by graphic tablet according to claim 31 storage medium.
45., it is characterized in that described stroke is an information of lifting pen from starting to write according to claim 31 storage medium.
CN961127317A 1995-10-12 1996-10-11 Character sign recognition device and method thereof, and computor controlling device Expired - Fee Related CN1094221C (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP7263919A JPH09106439A (en) 1995-10-12 1995-10-12 Character recognizing device, its method and computer controller
JP263919/95 1995-10-12

Publications (2)

Publication Number Publication Date
CN1155127A true CN1155127A (en) 1997-07-23
CN1094221C CN1094221C (en) 2002-11-13

Family

ID=17396098

Family Applications (1)

Application Number Title Priority Date Filing Date
CN961127317A Expired - Fee Related CN1094221C (en) 1995-10-12 1996-10-11 Character sign recognition device and method thereof, and computor controlling device

Country Status (3)

Country Link
JP (1) JPH09106439A (en)
KR (1) KR100232975B1 (en)
CN (1) CN1094221C (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110796917A (en) * 2018-08-02 2020-02-14 义守大学 Computerized writing evaluation and training method

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7298341B2 (en) * 2019-06-28 2023-06-27 株式会社リコー ELECTRONIC INFORMATION BOARD, ELECTRONIC INFORMATION BOARD SYSTEM AND THEIR CONTROL METHOD AND PROGRAM

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110796917A (en) * 2018-08-02 2020-02-14 义守大学 Computerized writing evaluation and training method

Also Published As

Publication number Publication date
KR970022832A (en) 1997-05-30
CN1094221C (en) 2002-11-13
KR100232975B1 (en) 1999-12-01
JPH09106439A (en) 1997-04-22

Similar Documents

Publication Publication Date Title
CN1107283C (en) Method and apparatus for character recognition of handwriting input
US5657396A (en) Method and apparatus for pattern recognition and validation, especially for hand-written signatures
JP4698289B2 (en) Low resolution OCR for documents acquired with a camera
US6212299B1 (en) Method and apparatus for recognizing a character
CN1163840C (en) Reducing handwriting recognizer errors using decision trees
CN1550069A (en) Component-based, adaptive stroke-order system
US5125039A (en) Object recognition system
CN1607539A (en) System and method for detecting a list in ink input
EP0378762A2 (en) Graphics images data compression method
CN1110759C (en) Information processing method and apparatus
CN1124537C (en) Online character input device
CN110008997A (en) Image texture similarity recognition method, device and computer readable storage medium
CN1094221C (en) Character sign recognition device and method thereof, and computor controlling device
CN1469621A (en) Portable information apparatus for processing inputted data from external apparatus and method thereof
CN1122928A (en) Information processing method and apparatus
CN1029534C (en) Handwriting Chinese character online identifying method and system
US7016535B2 (en) Pattern identification apparatus, pattern identification method, and pattern identification program
CN1262974C (en) System and method for limiting cursor shifting range in any images
CN114187388A (en) Animation production method, device, equipment and storage medium
CN1641681A (en) Method for rapid inputting character information for mobile terminal with pickup device
JP2001084336A (en) Device and method for character recognition, and program recording medium
CN1277229C (en) Result detection for voice file loading and reloading method
CN117275013B (en) Chinese character stroke writing identification method on mobile terminal
CN1222860C (en) Character input method and system suitable for multiple input methods
CN114296894A (en) Call chain intelligent analysis method and device based on log data and electronic equipment

Legal Events

Date Code Title Description
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C06 Publication
PB01 Publication
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20021113

Termination date: 20151011

EXPY Termination of patent right or utility model