CN1622121B - Modified handwritten Chinese character input recognition method - Google Patents
Modified handwritten Chinese character input recognition method Download PDFInfo
- Publication number
- CN1622121B CN1622121B CN200310115725A CN200310115725A CN1622121B CN 1622121 B CN1622121 B CN 1622121B CN 200310115725 A CN200310115725 A CN 200310115725A CN 200310115725 A CN200310115725 A CN 200310115725A CN 1622121 B CN1622121 B CN 1622121B
- Authority
- CN
- China
- Prior art keywords
- word
- candidate
- chinese character
- current
- user
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Document Processing Apparatus (AREA)
- Character Discrimination (AREA)
Abstract
The improved hand written Chinese character input and recognition method for electronic device includes the following steps: recognizing Chinese character written in the writing area based on preset regulations to generate candidate Chinese character list, and showing in some display area the candidate Chinese character list. The candidate Chinese character list includes at least the recognized current Chinese character and the combination of the current Chinese character and the recognized and displayed candidate Chinese character.
Description
Technical field
The present invention relates to handwritten Chinese character input identification, particularly a kind of improved hand-written Chinese character input recognition method that is used for electronic installation, it can tolerate overtime overtime influence that even do not receive, and can allow the word of having imported is carried out the part modification.
Background technology
The handwritten Chinese character input is to have the computing machine of touch-screen or touch pad (board) and other electronic equipment input method commonly used.Along with the widespread use of notebook computer, mobile phone and personal digital assistant (PDA) etc., the identification of handwritten Chinese character input fast, efficiently and has accurately become many people's hope.
For writing Chinese characters, different people has different writing styles and writing style.A difficult problem of handwritten Chinese character input identification is exactly how to confirm whether the user has write a word.Therefore, an obstacle that hinders fast and accurately identification is exactly so-called overtime, i.e. stand-by period after the finishing touch because overtime be to detect the most effective the way whether user has write a word.For handwritten Chinese character input identification can actually be used, overtimely be actually very short,, generally can regulate between second at 0.2-2.0 between second at 0.2-0.8 usually by system or user.
Yet, utilize and overtimely confirm whether input is accomplished and have problem sometimes.At first, it is overtime be difficult to set a best, because overtime difference with the user is very different, that is to say that the interruption between writing speed and stroke and the word is depended on and the user to a great extent.Secondly, different overtime different with word.For some word, the user will move a very long segment distance between two strokes.When distance is long, just need be than common stroke to the longer time of required time of stroke.In this case, the user write occur before the word overtime, will occur in overtime before with overtime after the stroke write be identified as the situation of two words.
This owing to the overtime problem that causes in order to solve, common solution is that the form with person's handwriting is that the user provides visual feedback, so that it can be confirmed when identification.In addition, most handwritten Chinese character input identification software all allows the overtime length of the artificial setting of user.Some solution provides two or more handwriting inputs district to the user, only writes a word to force the user an input field.
In fact, the person's handwriting feedback must be arranged on the below of person's handwriting, otherwise the user can not notice ongoing identification.This way can not be used on the device that does not have touch-screen, and for example those have only on the device of handwriting pad.Let user oneself set and overtime can be provided stand-by period (overtime) more flexibly, but can not deal with problems.Because concerning all words, have only one identical overtime.In fact, require to obtain one and be fit to that the really overtime flexibly of different strokes is unpractical in each word and the word.For those portable electric appts that has only a relatively little screen, have only a relatively limited district of writing, therefore, a plurality of districts of writing can't be provided.
In addition and since handwriting recognition be not 100% accurately, the user may rewrite a word sometimes, can be correctly validated to guarantee word.Because Chinese character needs complicated writing process, concerning the user, no matter write a word repeatedly on muscle power or all be a burden psychologically.And, can hinder so fast and effective recognition.In addition, if having only part (for example, a part) correct in the primary candidate, and system provides the candidate word table, and the user also browses all candidate (normally 8-10 word) possibly, and this makes selection course more difficult.And if the user thinks identification error, he must rewrite whole word.
To rewrite and help input and identification in order to reduce, people have proposed various solutions, for example: according to radical etc. with the word classification or decompose (as: the Five-stroke Method and grey a word used in person's names sign indicating number etc.).United States Patent (USP) 5,923,778 disclose a kind of method of utilizing radicals by which characters are arranged in traditional Chinese dictionaries information to discern, and have proposed a kind of based on radicals by which characters are arranged in traditional Chinese dictionaries information grading structure word database of information.United States Patent (USP) 6,272,243 disclose a kind of method and apparatus of importing to hand-written memo pad of being used for, and it has a predefined sub-input field.Yet it only is suitable for importing Japanese.
Summary of the invention
The object of the present invention is to provide a kind of improved hand-written Chinese character input recognition method, it does not receive overtime influence basically, and can tolerate overtime even do not receive overtime the influence fully.
The present invention also aims to provide a kind of improved hand-written Chinese character input recognition method, it can reduce the rewriting in handwritten Chinese character input identification, increases the speed of input and allows the part of a word is revised.
According to an aspect of the present invention; A kind of improved hand-written Chinese character input recognition method that is used for electronic installation is provided; Comprise the following steps: to write the Chinese character that write in the district of said electronic installation according to predetermined rule identification user; Produce candidate's word table, and show said candidate's word table in a viewing area of said electronic installation in above-mentioned identification step identification; Wherein said candidate's word table comprises the current word and the combination of this current word with the candidate that before this current word, is identified and shows that just has been identified at least.
According to another aspect of the present invention; A kind of improved hand-written Chinese character input recognition method that is used for electronic installation is provided; Comprise the step of writing the Chinese character district write of identification user at said electronic installation; Wherein said Chinese Character Recognition is only just accomplished when satisfying following condition: in a predetermined district that begins, preceding several strokes constitute the word of a completion for the starting point of previous stroke or intermediate point, and the initial point of current stroke or intermediate point begin in the district described.
According to a further aspect of the invention; A kind of improved hand-written Chinese character input recognition method that is used for electronic installation is provided; It comprises that the identification user writes the step of the Chinese character of writing in the district and writes the step that the district is presented at main candidate's word table of said identification step identification said of said electronic installation; Wherein said demonstration is included in the different piece of the font of this primary candidate of background parts prompting, and said notification portion constitutes one and independently write the district, in this writes the district, can discern independently.
Description of drawings
Illustrate embodiments of the invention with reference to the accompanying drawings.
Figure 1A and 1B schematically illustrate the mistake identification that the hand-written Chinese character input recognition method that utilizes the overtime prior art of discerning possibly occur;
Fig. 2 shows an embodiment according to the inventive method;
Fig. 3 is the process flow diagram of the said embodiment of Fig. 2;
Fig. 4 is another embodiment according to the inventive method;
Fig. 5 is the process flow diagram of the said embodiment of Fig. 4;
Fig. 6 A and 6B are another embodiment according to the inventive method.
Fig. 7 is the process flow diagram of the said embodiment of Fig. 6.
Embodiment
Because handwritten Chinese character input identification can utilize variety of way well known in the art to realize, and can use conventional system or software to discern, and for the sake of simplicity, identifying itself is not done detailed description at this.
At first with reference to Fig. 1, wherein Figure 1A schematically illustrates the mistake identification that the hand-written Chinese character input recognition method that utilizes the overtime prior art of discerning possibly occur.Shown in Figure 1A, the user plans to write the bipartite word " bright " of a left right model.He writes " day " on the left side usually earlier, writes " moon " on the right again.Because " bright " is to be made up of the radical " day " that can independently become word and " moon " two parts; He need move long segment distance to the right half part the first stroke of " moon " (top is leaned in the centre of importing/write the district) from the left-half finishing touch of " day " (lower part is leaned in the centre of importing/write the district), and the time that this cost is long also of short duration pause possibly occur between two parts.If the overtime length that the time ratio that pauses is predetermined, system can be identified as the part on the left side " day " or " moon " word, and demonstration " day " and " moon " in candidate's word table.Then, system is identified as " moon " or " day " with the part on the right and in candidate's word table, shows " moon " and " day " discerned.In other words, because overtime, system is identified as two independently words with two parts.
Same problem also can appear at the identification of the word that two parts of mo(u)ld bottom half constitute.Shown in Figure 1B, Chinese character " fortunately " is to be made up of radicals by which characters are arranged in traditional Chinese dictionaries " soil " and the latter half of not constituting complete word.Owing to overtime, system may " fortunately " word recognition become two words, i.e. " soil " word and " sheep " word that looks like the latter half.
When practical application, adjustment stand-by period (overtime) does not have anything to help, because overtime different and different with the user's with Chinese character, stroke, writing style.Even same user, concerning him, fixing overtime also also inapplicable all word.
Fig. 2 shows an embodiment according to improved method of the present invention.This embodiment utilizes commonly used and vital candidate's word table concerning handwritten Chinese character input identification, makes it when comprising current word, also comprises the combination of a word that is identified before the current therewith word of this current word and shows.This candidate's word table may be displayed on (under the situation of board, needing independently usually, display screen is presented at the word of writing on the board) on touch-screen or the independent display screen.This embodiment has just increased the most probable combination of a word that is identified before the current therewith word of current word and shows in candidate's word table, do not require that increasing extra display unit takies display screen.
As shown in Figure 2, suppose that same user will write same bipartite word " bright ".Rules for writing according to stroke; He will at first write left-half " day " with the speed that is fit to its writing style; If owing to its writing style or other reason, beginning that a pause is arranged before writing right half part, and it is longer than predetermined time-out time to pause; System will think that input accomplishes, and left-half is identified as " day " word by known mode." day " is used as certain position that primary candidate is presented at a display screen, in the present embodiment, candidate be presented at write the district above.
The user continues to write remainder, i.e. right half part " moon ".System produces new candidate's word table after overtime, listed the combination of " moon " and " moon " that he has just write and " day " of before this identification and demonstration, i.e. his " bright " word that will write.Then, " bright " word can be used as one of primary candidate or candidate word be presented at write the district above.Can pass through system's setting, make " bright " word as primary candidate get into text area automatically or: click or otherwise, and choose this word by user by selecting, for example.After the selection, this word is moved to text area.
In this embodiment, the word that can be identified needs not to be complete Chinese character, and can be the ingredient of words such as radical.For example: if user writing " companion " word; After accomplishing left-half, occur overtime; And left-half is a radical but not complete Chinese character continues to write right half part " partly " before at him, and system will discern this radical and in candidate's word table, show this radical.After the user has write right half part, system will produce new candidate's word table, comprise current word of just having discerned " partly " and the word " companion " that is combined with the radical of discerning before this and show by " partly " in this table.The user can select this word, and system moves on to text area with this word.
In addition, described combination must constitute a complete word.According to the size and the layout of display screen, candidate's word table can comprise maximum 10 words, so that select.
Like this, through only in candidate's word table, increasing the most probable combination of a word that is identified before the current therewith word of current word and shows, word just can improve identification efficiency greatly usually, and need not increase display unit and occupy display screen.
Fig. 3 shows the process flow diagram of Fig. 2 embodiment.When the user begins on touch-screen or board, to write, detect beginning.Whether system's detection at any time has the interruption (overtime) (step 102) of being longer than the schedule time between stroke.If system detects overtime, then will discern by the word (step 201) that stroke constituted before overtime, then, in candidate's word table, show the word (step 202) of identification.Simultaneity factor continues to detect (step 103).If detect another overtime (step 104), then discern (step 105) and show (step 106).In step 106, system is presented at the word of step 105 identification and at the word of step 105 identification with in the combination of step 201 and 202 identifications and the word that shows.If do not detect overtimely, then continue to detect.After the demonstration of step 106, system is according to user's selection, and the word that the user is selected places the text editing district.
Fig. 4 is another embodiment according to the inventive method.Except utilizing above-mentioned candidate's word table,, can almost completely eliminate overtime influence to identification through the position of stroke on touch-screen or board of detected words.
Because Chinese-character writing follows from left to right and rule from top to bottom, for example: on touch-screen/board, write normally and begin from the upper left corner of writing the district; Finish in the lower right corner; Can promptly write the upper left quarter in district to the position of a word first stroke of a Chinese character, be defined as one " beginning the district ".The size that begins to distinguish cans be compared to most writes the 1/4th big slightly of district or writing frame, so that most of user can begin to write beginning the district.If a word comprises two or more parts, for example: about or two parts up and down, can begin to write another part from the place beyond beginning to distinguish.In the present embodiment, whether these characteristics of utilizing Chinese character to import detect a word and have write, and whether have begun writing of another word.
According to present embodiment, under scenario, a stroke will be considered to the beginning of a new word: 1) starting point of previous stroke or preceding several strokes or intermediate point are not beginning the district; 2) preceding several stroke constitutes a complete word; 3) intermediate point of the starting point of current stroke or current stroke is beginning the district.That is to say that only when above-mentioned three conditions satisfy simultaneously, think that just the word that is made up of preceding several strokes write, current stroke is the beginning of a new word.Therefore, needn't wait timeout, system will rather than overtimely confirm whether a word has been write according to above-mentioned condition.
As shown in Figure 4, begin to distinguish the upper left corner, and can mark with line or figure writing district or writing frame, begin to write in this zone with the prompting user.
Be example with " bright " word equally, this word is made up of left and right sides two parts and should begins to write from " day " word on the left side is other.The starting point of first three stroke that " day " word is other and intermediate point are all in beginning the district.Though the 4th starting point may begin beyond the district, first three does not constitute a complete word.Therefore, the 4th beginning is not considered to the beginning of a new word.Right half part " moon " generally begins to write from beginning to distinguish in addition.Though left-half " day " also can constitute a complete word as radical, the first stroke of right half part " moon " can not be considered to the beginning of a new word, because the first stroke of " moon " is to begin not satisfy above-mentioned the 3rd condition beyond the district.System can detect the starting point of each stroke, to determine whether to have write a word according to this rule.The intermediate point that makes system detect each stroke also can be set, and promptly whether the part of the centre of stroke drops on and begins the district.The word of having write can be identified and be presented at the candidate block.
Owing to the user may begin to distinguish " moon " that begins to write right half part owing to reasons such as ignorance; And this may cause system that " moon " mistake of right half part is identified as complete " moon " word; Therefore; Candidate's word table also should comprise current word and the combination that is close to this current word word before, as described with reference to Fig. 2 except comprising current word.In this example, " moon " and " bright " all is displayed in candidate's word table, therefore, and owing to the user fails beginning to begin to write beyond the district the correct identification that mistake that right half part " moon " causes can not influence system.
Fig. 5 shows the flow process of the said embodiment of Fig. 4.When the user begins on touch-screen or board, to write, detect beginning.System at first detects the starting point/intermediate point of previous stroke and whether distinguishes (step 301) in predetermined beginning.If not, whether several strokes constitute a complete Chinese character (step 302) before then detecting.If, then continue to detect, do not beginning district's (step 401) up to the starting point/intermediate point that detects a stroke.If preceding several stroke constitutes a complete Chinese character, then detect current stroke starting point/intermediate point and whether beginning district's (step 303).If the result who detects is sure, the word that then preceding several strokes is constituted is discerned (step 304), and shows that (step 305) is at the word of step 304 identification and the word that detects neutralization identification and show at this previous round.If the result of step 303 negates then to get back to step 401.At last, system produces selected word is placed the text editing district according to user's selection.
Fig. 6 A and 6B are another embodiment according to the inventive method, and it makes the user can partly revise a word.
According to Hanzi font directory system (SKIP), Chinese character can be divided into four kinds of fonts, left right model promptly as follows (for example: " bright " word), last mo(u)ld bottom half (for example: " Wu " word), the encirclement type (for example: " state " word) and solid type (for example: " greatly " word).Thereby, except preset one of the district of writing at display screen begins the district, can be by some way, for example: color or gray scale (under the situation of monochromatic display screen) prompting is with the font of the word of identification.
Mo(u)ld bottom half encirclement type solid type on the left right model
Each is with the part of color tips; For example: the left half of left right model and right half constitute an independently cog region respectively; At this, system can discern a complete word or a radical etc., is a complete word but should make the whole primary candidate of discerning in the district of writing.
Shown in Fig. 6 A, when the writing district input Chinese character and " mix " of touch-screen, beginning that the user (begins the district) from the upper left corner.System discerns user's input.Identification can be carried out with the described method of Fig. 2, also can carry out with the method for Fig. 4, even can discern in a known manner.Because for example: careless handwriting is that " companion " rather than he will write " mixing " writing primary candidate that the district shows, it is left right model that the background of writing the district is pointed out this word with different colors or gray scale.At this moment, the user needn't rewrite whole word, because primary candidate " almost being correct " looks promptly to look like the word that the user will write that just left-half is not right.At this moment, the user can only rewrite left-half.When the user rewrote left-half, the background color of left field (gray scale) changed, and only revise left-half with the affirmation user, and right half part remained unchanged.System produces a new primary candidate according to user's modification and " mixes ", and this is the word that will write of user just.
Candidate also can be simultaneously displayed in candidate's word table, for example: in the candidate's word table on write the district shown in Figure 4.
Another kind of mode is to make candidate's word table comprise that all (quantity according to the word that can show is decided) has the radical of user's selection or the word of radicals by which characters are arranged in traditional Chinese dictionaries.In this case, the user can write the correct part of district's selection, rather than revises wrong part.Shown in Fig. 6 B, when the user notice the primary candidate that is shown be not he in fact to write word the time, he can select correct part, for example: correct part is selected in the zone (right side area) that has correct part through click.Then, all words that have correct part " partly " all are displayed in candidate's word table, in this example, and " companion ", " mixing " stumbles " etc.The user can pass through from candidate's word table, for example: click the word (" mixing ") of selecting him to write and select " mixing " word.
Present embodiment is not suitable for the board that has independent display screen.
Fig. 7 is the process flow diagram of the said embodiment of Fig. 6.The user has write after the word; System is with the method for Fig. 2 or Fig. 4, or known mode is discerned the primary candidate (step 502) that demonstration subsequently is identified; And with the font (step 503) of this word of color/gray scale prompting, various colors/gray scale is represented the not same district of font.If the word that shows is exactly the word (step 504) that the user writes, then system produces selected word and is placed on text editing district (step 505) according to user's selection.If the character segment in that step 503 showed is correct, the user can select to confirm the part of correct part or direct modification mistake.If the user selects to confirm correct part (step 601), system will produce candidate's word table (step 602), and this table comprises all words that contain this correct part.If the user selects the part (step 603) of direct modification mistake, the user will point out, the part behind identification and the display update through changing color/gray scale of revising part in system.
Though with color/gray scale is the mode that example has been explained the prompting user, obviously also can adopt alternate manner well known to those skilled in the art.
More than invention has been described with the mode of embodiment, but the present invention is not limited to said embodiment, those skilled in the art can make various modifications to the present invention within the scope of the appended claims.
Claims (12)
1. improved hand-written Chinese character input recognition method that is used for electronic installation; Comprise the following steps: to write the Chinese character that write in the district of said electronic installation according to predetermined overtime rule identification user; Produce candidate's word table, and show said candidate's word table in a viewing area of said electronic installation in above-mentioned identification step identification; Wherein said candidate's word table comprises the current word and the combination of this current word with the candidate that before this current word, is identified and shows that just has been identified at least.
2. according to the process of claim 1 wherein that the word in said candidate's word table comprises complete word and structure character segment.
3. according to the method for claim 2, wherein said structure character segment comprises radical.
4. according to the process of claim 1 wherein that the described current word that just has been identified is a complete word or incomplete word.
5. according to the method for claim 1, also comprise the step of from said candidate's word table that said candidate viewing area shows, selecting a word.
6. according to the process of claim 1 wherein the constitute complete word of said current word with the candidate that before this current word, is identified and shows.
7. an improved handwritten Chinese character that is used for electronic installation is imported identification equipment; Comprise with lower device: be used for a device of writing the Chinese character of writing in the district at said electronic installation according to predetermined overtime rule identification user; Be used to produce a device at candidate's word table of above-mentioned recognition device identification, and the device that is used for showing in a viewing area of said electronic installation said candidate's word table; Wherein said candidate's word table comprises the current word and the combination of this current word with the candidate that before this current word, is identified and shows that just has been identified at least.
8. according to the equipment of claim 7, the word in wherein said candidate's word table comprises complete word and structure character segment.
9. according to Claim 8 equipment, wherein said structure character segment comprises radical.
10. according to the equipment of claim 7, the wherein said current word that just has been identified is a complete word or incomplete word.
11., also comprise the device of from said candidate's word table that said candidate viewing area shows, selecting a word according to the equipment of claim 7.
12. according to the equipment of claim 7, the complete word that constitutes of wherein said current word and the candidate that is identified before this current word and shows.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200310115725A CN1622121B (en) | 2003-11-28 | 2003-11-28 | Modified handwritten Chinese character input recognition method |
| CN201210249413.4A CN102981753B (en) | 2003-11-28 | 2003-11-28 | The hand-written Chinese character input recognition method improved |
| HK05109897.0A HK1077900A1 (en) | 2003-11-28 | 2005-11-07 | Improved chinese handwriting recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200310115725A CN1622121B (en) | 2003-11-28 | 2003-11-28 | Modified handwritten Chinese character input recognition method |
Related Child Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006101538074A Division CN100410966C (en) | 2003-11-28 | 2003-11-28 | Modified hand-written Chinese character input recognition method |
| CN201210249413.4A Division CN102981753B (en) | 2003-11-28 | 2003-11-28 | The hand-written Chinese character input recognition method improved |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1622121A CN1622121A (en) | 2005-06-01 |
| CN1622121B true CN1622121B (en) | 2012-08-29 |
Family
ID=34760531
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200310115725A Expired - Fee Related CN1622121B (en) | 2003-11-28 | 2003-11-28 | Modified handwritten Chinese character input recognition method |
| CN201210249413.4A Expired - Fee Related CN102981753B (en) | 2003-11-28 | 2003-11-28 | The hand-written Chinese character input recognition method improved |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210249413.4A Expired - Fee Related CN102981753B (en) | 2003-11-28 | 2003-11-28 | The hand-written Chinese character input recognition method improved |
Country Status (2)
| Country | Link |
|---|---|
| CN (2) | CN1622121B (en) |
| HK (1) | HK1077900A1 (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104656938B (en) * | 2013-11-19 | 2018-07-06 | 阿尔派株式会社 | Input device and character input method |
| US10095673B2 (en) | 2014-11-17 | 2018-10-09 | Lenovo (Singapore) Pte. Ltd. | Generating candidate logograms |
| CN104598148B (en) * | 2014-12-16 | 2017-09-19 | 任文 | A kind of computer contact screen handwritten Chinese character method |
| CN107870678A (en) * | 2016-09-26 | 2018-04-03 | 中兴通讯股份有限公司 | A kind of hand-written inputting method and device |
| TWI664604B (en) * | 2018-03-20 | 2019-07-01 | 王佑鈞 | System for construction and application of handwriting chinese character glyph sets to be used in a computer system |
| US20220122477A1 (en) * | 2020-10-20 | 2022-04-21 | Holistic Language Solutions LLC | Computerized method and apparatus for determining accuracy of written characters and stroke order and compliance with rules and providing visual and audio feedback |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1197333A (en) * | 1998-03-09 | 1998-10-28 | 柏凯 | Computer hand writing input machine codes |
| CN1232204A (en) * | 1998-04-16 | 1999-10-20 | 国际商业机器公司 | Chinese handwriting identifying method and device |
| CN1324068A (en) * | 2000-03-29 | 2001-11-28 | 松下电器产业株式会社 | Explanatory and search for handwriting sloppy Chinese characters based on shape of radicals |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1040340A (en) * | 1996-05-01 | 1998-02-13 | Nippon Steel Corp | Handwritten character recognizing device and recording medium |
| JPH10162101A (en) * | 1996-11-28 | 1998-06-19 | Hitachi Ltd | Device and method for recognizing handwritten character |
| JPH11191138A (en) * | 1997-12-25 | 1999-07-13 | Toshiba Corp | Kanji input method utilizing handwritten character recognizing function and character input device |
| JP3986708B2 (en) * | 1999-06-28 | 2007-10-03 | シャープ株式会社 | Character recognition error correction device and computer-readable recording medium recording character recognition error correction program |
-
2003
- 2003-11-28 CN CN200310115725A patent/CN1622121B/en not_active Expired - Fee Related
- 2003-11-28 CN CN201210249413.4A patent/CN102981753B/en not_active Expired - Fee Related
-
2005
- 2005-11-07 HK HK05109897.0A patent/HK1077900A1/en not_active IP Right Cessation
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1197333A (en) * | 1998-03-09 | 1998-10-28 | 柏凯 | Computer hand writing input machine codes |
| CN1232204A (en) * | 1998-04-16 | 1999-10-20 | 国际商业机器公司 | Chinese handwriting identifying method and device |
| CN1324068A (en) * | 2000-03-29 | 2001-11-28 | 松下电器产业株式会社 | Explanatory and search for handwriting sloppy Chinese characters based on shape of radicals |
Non-Patent Citations (1)
| Title |
|---|
| 图1. |
Also Published As
| Publication number | Publication date |
|---|---|
| HK1077900A1 (en) | 2006-02-24 |
| CN1622121A (en) | 2005-06-01 |
| CN102981753A (en) | 2013-03-20 |
| CN102981753B (en) | 2015-10-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5303312A (en) | Handwriting recognition by character template | |
| CA2477637C (en) | Component-based, adaptive stroke-order system | |
| JP5831411B2 (en) | Kanji stroke order learning device and kanji stroke order learning program | |
| US20030099398A1 (en) | Character recognition apparatus and character recognition method | |
| US20040243389A1 (en) | Method and apparatus for selecting symbols in ideographic languages | |
| EP1513053A2 (en) | Apparatus and method for character recognition | |
| CN101281449B (en) | Hand-written character recognizing method and system | |
| JP3292752B2 (en) | Gesture processing device and gesture processing method | |
| CN1622121B (en) | Modified handwritten Chinese character input recognition method | |
| US20130082985A1 (en) | Content display apparatus, and content display method | |
| CN100410966C (en) | Modified hand-written Chinese character input recognition method | |
| JP2004523034A (en) | Character generation system | |
| KR100298547B1 (en) | Character input apparatus | |
| US20020085772A1 (en) | Intelligent correction key | |
| US6973214B1 (en) | Ink display for multi-stroke hand entered characters | |
| Akita et al. | SliT: character input system using slide-in and tap for smartwatches | |
| KR101880868B1 (en) | Apparatus for inputting chinese character and method for inputting chinese character using the same | |
| CN102156616B (en) | Touch screen key-writing input method | |
| Dell et al. | Improving form-based data entry with image snippets | |
| CN111143541A (en) | Character generation system for deep learning based on handwriting | |
| JP5381185B2 (en) | Electronic device and information display program | |
| JPH07200723A (en) | Method and device for recognizing character | |
| CN101311883B (en) | Method for inputting Chinese characters using national language phonetic symbol | |
| JP3210778B2 (en) | Handwritten character recognition device | |
| JP3153704B2 (en) | Character recognition device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1077900 Country of ref document: HK |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: GR Ref document number: 1077900 Country of ref document: HK |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20151223 Address after: Espoo, Finland Patentee after: Technology Co., Ltd. of Nokia Address before: Espoo, Finland Patentee before: Nokia Oyj |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120829 Termination date: 20191128 |