CN1567358A - Identification of character input in improved electronic device - Google Patents
Identification of character input in improved electronic device Download PDFInfo
- Publication number
- CN1567358A CN1567358A CNA031371752A CN03137175A CN1567358A CN 1567358 A CN1567358 A CN 1567358A CN A031371752 A CNA031371752 A CN A031371752A CN 03137175 A CN03137175 A CN 03137175A CN 1567358 A CN1567358 A CN 1567358A
- Authority
- CN
- China
- Prior art keywords
- character
- candidate characters
- electronic equipment
- row
- current
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/142—Image acquisition using hand-held instruments; Constructional details of the instruments
- G06V30/1423—Image acquisition using hand-held instruments; Constructional details of the instruments the instrument generating sequences of position coordinates corresponding to handwriting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Character Discrimination (AREA)
Abstract
When writing on the electronic device (1) to show the character on the current character position (24), the character identification package tries to identify the writing character (22) and generates the first candidate characters list (50). According to the similarity of writing character and candidate character, it arranges the candidate characters of the first list (50). Besides, it searches in the dictionary that whether there is possible characters pair according to the character that next to the current character, and confirms the second candidate characters list (52). Then it compares the two lists, and displays the first list according to its display sequence. Among which, the display sequence maybe different from or same with the initial sequence, that depends on the overlap of the two lists. This invention is suitable for writing the complex character such as Chinese character, or applied in the mini device that with limited memory such as mobile phone, PDA, GPS navigator.
Description
Technical field
The present invention relates to the identification of the character input of improved electronic equipment.The present invention is particularly useful for, but is not limited to, and the character on touch-screen is write in identification, special complex characters such as Chinese character, Japanese or Korea character.
Background technology
A main growth field of screen equipment is a miniature portable equipment, for example: mobile phone, PDA(Personal Digital Assistant), GPS (GPS) navigating instrument etc.These equipment adopt various method incoming symbols or data, for example: adopt true or virtual push button, speech recognition, character recognition etc.Under the situation of character recognition, character to the writing in the district of display, and is used character recognition software to determine what the character of handwriting input is by hand-written.Though dummy keyboard may be quicker than hand-written when carrying out the input of alphabetic character (as: Roman character, Greek alphabet, cyrillic alphabet or the like), yet at the complicated more character of input, then not too suitable during as: Chinese character, Japanese or Korea character, because keyboard seems too small for all characters, perhaps key is too much, so that the user is difficult for finding correct button.
It has been found that simple hand-written character analysis is not enough to provide accurate result, be not enough to often provide accurate result at least.Contextual analysis technology according to the contextual modifications possible outcome of other content that occurs also is necessary.
The structure of touch-screen is well known in the art, and the touch detection can realize that for example: electric capacity or induction detect contact-making switch or the like with many well-known systems.Equally, character recognition software is well-known, although very not perfect usually.
United States Patent (USP) 4,979,227 disclose a kind of method of using dynamic dictionary to proofread and correct the recognition result of character string in the document.Dictionary is to produce by prescan entire document and the obviously correct speech of discerning of storage.During actual scanning, compare between some character string in character string that reads and dictionary.The computation complexity of this method is higher.
United States Patent (USP) 4,654,875 have described a kind of algorithm that is used for text identification/correction.This method reads all speech, and according to base character identification, with in front some letter back orders occur the probability of occurrence of letter and the dictionary content that can accept word analyze.Its computation complexity is higher.
At United States Patent (USP) 6,219, in 449, when reading a speech, form character in this speech each and produce candidate's word table.Expectation fiduciary level according to change information is selected word from possible combination.The method is also remarkable.
United States Patent (USP) 6,137,908 have described segmentation and the context processing method that is used to improve the character recognition precision.The probability of speech in the dictionary is sought to estimate in the dynamic programming path, and derives the recognition result of global optimum (maximum possible).The computation complexity of the method is higher.
United States Patent (USP) 6,442,295 disclose a kind of method of using the dictionary in the storer according to predetermined code.Weight sequence corresponding to the character string that receives is provided for storer.Distance between the speech that calculating will be discerned and the speech of storage, this distance are the weights of each character of the speech that will discern, and the difference summation between the weight of each character of the speech of storage.Stored word with minor increment is a net result.
Summary of the invention
In this instructions and claims, term " comprises " or similar terms is used to represent comprising of non-exclusionism, makes the method or the device that comprise one group of unit not only comprise those unit, and can comprise other unlisted unit.
According to an aspect of the present invention, provide a kind of method that character is input to current character position in the character row in electronic equipment, wherein said character is input in the electronic equipment by writing.This method comprises and receives one first candidate characters table, receives one second candidate characters table, the content of the more described first and second candidate characters tables, and determine the DISPLAY ORDER of candidate characters in the first candidate characters table according to comparative result.The first candidate characters table is possible corresponding to written character, by the candidate characters table that character recognition produces, and this first table has initial order.The second candidate characters table is the candidate characters table that possible produce corresponding to written character, according to the current character location context.The content of more described first and second tables is to determine degree of overlapping therebetween.
According to another aspect of the present invention, provide a kind of circuit that character is input to current character position in the character row in the display of electronic equipment, wherein said character is to be input in the electronic equipment by writing.This circuit comprises an input, a storer, an output and a processor.Described input receives by writing the character that is input in the electronic equipment.Described memory stores first character set, the set that the content of current character row and book character are arranged.Described output is exported the first candidate characters table so that show with DISPLAY ORDER.Described processor produces by character recognition may be corresponding to written character, from the first candidate characters table of described first character set, and wherein first table has initial order.Processor may be corresponding to the second candidate characters table of written character according to the context and the book character arrangement set generation of current character position.The processor further relatively contents of first and second tables and is determined the DISPLAY ORDER of candidate characters in first table at least determining degree of overlapping therebetween according to the degree of overlapping of determining.
According to another aspect of the present invention, a kind of electronic equipment is provided, comprise screen with viewing area, detect the detecting device of the written character in the viewing area, the first memory that contains first character set, the second memory that contains character row, and processor, described processor compares the character in written character and described first character set, produce the first candidate characters table, produce the second candidate characters table, relatively the content of first and second tables, determine the DISPLAY ORDER of candidate characters in first table according to comparative result, and export first according to the described DISPLAY ORDER that will show and show.Described viewing area has a character row district that shows the character row of input character, one be used for written character in case its input is presented at character row the current character position write the district, with a candidate character area, this candidate character area is used for showing according to DISPLAY ORDER first table of output, and allows to select in described a plurality of candidate characters one so that be presented on the current character position of character row.According to producing the first candidate characters table the character of initial order in described first character set.Context according to current character position in the character row produces the second candidate characters table.Relatively the content of first and second tables is to determine degree of overlapping therebetween.
Description of drawings
Understand for convenience and enforcement the present invention, describe exemplary preferred embodiment referring now to accompanying drawing, wherein:
Fig. 1 shows the front end of a PDA of exemplary embodiment;
Fig. 2 is the synoptic diagram of the display circuit of exemplary embodiment;
Fig. 3 is based on the process flow diagram of process of the operation of exemplary embodiment; And
Fig. 4 shows the first exemplary table and second table with initial order.
Embodiment
In institute's drawings attached, identical Reference numeral is represented identical unit.
In brief, in this exemplary embodiment, when to electronic equipment written character so that when being presented at the current character position, the character recognition bag is attempted to discern the character of writing and is produced the first candidate characters table.Place candidate characters in first table according to the similarity between written character and the candidate characters according to initial order.In addition, according to the character that is right after front, current character position, the right dictionary of inquiry possibility character is to determine the second candidate characters table.Independently generation and relatively two tables, and press DISPLAY ORDER demonstration first and show, DISPLAY ORDER may be identical or different with initial order, and this depends on two degrees of overlapping between the table.So, considering the context of correct character position, the present invention is in conjunction with character identification result and the correct possibility of character.Do not exist relevantly between the information of these two tables owing to cause producing, improved the precision that identification is handled.
With reference to Fig. 1, wherein show and use a PDA10 of the present invention.This PDA10 has screen 12, and screen 12 has demonstration 14.Show that 14 are divided into following each zone: character row district 16 is used to show the character row 18 of input character; Write district 20, be used for written character 22 its input is presented at the current character position 24 of character row 18; Candidate character area 26 is used for the candidate characters table 28 of display candidate characters 28a-28f, in the hope of one in them corresponding to the character of in writing district 20, writing 22; With instruction area 30, idsplay order menu when needs carry out other operation.In addition, selecting character area 32 is virtual push buttons, is used for selecting the current character of current character position 24 appearance, and allows the input character late.
Though in this embodiment, just corresponding to the screen area of writing district 20 and candidate character area 26 that utilizes this characteristic, screen 12 is touch-screens to screen 12.Especially, it is used to determine writes in the district 20 just in content written, selects which character in candidate characters 28a-28f (if any) table 28 in candidate character area 26, has perhaps selected which other instruction.
In this embodiment, write the content that expectation is write in district 20, user's forward character row 18 input Chinese characters by use.
Fig. 2 is the synoptic diagram of touch screen circuitry and display driving circuit 40, the content that occurs in the demonstration 14 of display driving circuit 40 control screens 12.Here described its function, it is one and is used for according to writing district's written character 22 of 20, and character is input to circuit on the current character position 24 in the character row 18.As people known and in the prior art equipment, this circuit also has other function and the inputs relevant with control demonstration 14, needn't describe at this.
Level and vertical reference 42 and 44 are used to the contact point on the senses touch screen 12.This information offers screen drive circuit 40 as the X and Y coordinates signal Sx of indication contact point, the signal of Sy, so that explain and corresponding responding.
In brief, as follows by the process of writing input character.When detecting contact on touch-screen 12, level and vertical reference 42,44 transmit contact position by signal Sx, Sy to drive circuit 40.Processor 46 is maked decision and make to be shown 14: upgrade by more written character 22 and write district 20, and upgrade candidate character area 26 and current character position 24 to comprise the new table 28 of candidate characters; Or the current character position 24 in the renewal character row 18 is to comprise the character of selection.
Describe in more detail by writing the process of input character referring now to the process flow diagram of Fig. 3.
At step S102, detect touch and transmit position signallings to processor 46 from level and vertical reference 42,44.Processor determines at step S104 whether the position of detected touch is being selected in the character area 32.If then the character that current character position 24 occurs in the character row 18 remains on the original place, and moves the current character position at step S106 along a position.Otherwise, handle advancing to step S108.
At step S108, processor determines that detected touch is whether in the candidate characters 28a-28f of candidate character area 26.If the answer at step S108 is a "Yes", then means and from the table 28 of candidate characters 28a-28f, selected candidate characters.In step S110 update displayed 14, wherein on the current character position 24 of character row 18, import the candidate characters of selecting.Thereafter be step S112, wherein the next position in the character row is considered to current character position 24.Subsequently, processor turns back to step S102 to wait for next detected touch.
At step S114, processor determines that detected touch is whether in writing district 20.If the answer at step S114 is a "No", then meaning does not have character just to be write, and does not have candidate characters 28a-28f just selected in the candidate character area 26, and does not have character just selected in the current character position.Selection can be for example from the instruction of instruction area 30.So at step S116, processor is according to circumstances carried out some other processing.Otherwise,, then handle advancing to step S118 if be "Yes" promptly in the answer of step S114.
If the answer at step S114 is a "Yes",, on the position of detected touch, upgrade the current written character of writing in the district 20 22 then at step S118.Processor is attempted character recognition then according to current written character 22 at step S120.This comprises in the character set of first database of determining storer 48 which character (if there is) is the most similar to current written character 22.Making after this determines, at step S122, processor produces first table of the candidate characters with initial order.The mark that initial order provides for each candidate during character recognition based on processor, mark are the tolerance of confidence level, show that promptly processor thinks the similarity of candidate characters and written character.Yet, too much have the situation of phase equally likely possibility owing to exist, perhaps there is the very few situation that surpasses minimum possibility, can not produce first table sometimes.
Whether step S124 determines character recognition success fully, promptly whether has produced first table, if do not produce first table, then handles and turns back to step S102 to wait for next detected touch.If produced first table, then next procedure S126 determines whether the confidence level tolerance of first candidate in first table surpasses certain threshold value, and whether at least also surpasses the predetermined threshold for the confidence level tolerance of second candidate in first table.If the answer at step S126 is a "Yes", think that then first candidate in first table will be correct candidate, and upgrade current character position 24 showing that this first candidate is used as default candidate at step S128, and upgrade candidate character area 26 to be presented at the first complete candidate characters table that step S122 produces according to initial order.
If the answer at step S126 is a "No", then step S130 produces second table according to the context of current character position 24.This comprises the identification current context and it is compared with the set of context in second database of storer 48.Each context in the set of context in second database of storer 48 has a character on the relative current character position 24.If a context in the set of context is identical with current context, then with storer in character on this context-sensitive current character position 24 be placed in second table of candidate characters.In this preferred embodiment, employed context is restricted to alphabetic string.Effectively, this means that second table is the table that maximum possible is in the character of current character position 24 under the situation of considering the front character.In most preferred embodiment, only based on being right after in front character (be a left side or be that input sequence is then depended on the right side) in the character row.So, produce second content of showing in the mode that is totally independent of first content of showing.
In following step S132, relatively first and second show.If there be not overlapping (promptly not having common character) at all, then produce the DISPLAY ORDER identical with initial order at step S134.If there be certain overlapping (promptly having a common character at least), then at step S136, produce DISPLAY ORDER, wherein overlapping character is placed in the section start of table, with relative to each other identical order, just as they appear in the initial order.Non-overlapped character moves down in table, and with relative to each other identical order, just as they appear in the initial order.So if exist 100% overlappingly, then DISPLAY ORDER is identical with initial order.According to the DISPLAY ORDER of determining in step S134 or S136, step S138 update displayed 12 is upgraded to show first candidate (as default candidate) in first table current character position 24.Candidate character area 26 also is updated, thereby according in step S134 or the definite DISPLAY ORDER of S136, is presented at the first complete candidate characters table that step S122 produces.Handle and then turn back to step S102 to wait for the next detection that touches.
Fig. 4 shows two 2 tables 50 and 52.A table 50 is exemplary first tables with initial order, and another table 52 is exemplary second tables.These tables are for example generations of writing by the character shown in Fig. 1 22.Can find by these two tables relatively, only between first character of second character of first table 50 and second table 52, exist overlapping.Under the processing of above-mentioned steps S136, this means that public character is moved to the head of table for the DISPLAY ORDER of first table, first character of initial order is moved to the second place, and all the other characters are kept intact.So the DISPLAY ORDER of first table is identical with initial order, but preceding 2 characters are exchanged.This is at shown in the table 28 of the candidate characters 28a-28f in Fig. 1 candidate character area 26 in proper order.
As shown in Figure 4, first and second tables do not need the character of equal number.Sometimes first table can be longer than second table, sometimes can be conversely, and in that they can be identical At All Other Times.Yet, if desired, can produce dictionary, make second table that produces have the member of fixed qty.Can produce first table with equal number member equally, at every turn.The quantity of candidate characters 28a-28f in the candidate character area 26 is not limited to 6.In addition, it does not need to keep constant, can change according to the size of first table.If the character of first table that produces then can be clipped excessive character more than the free space of candidate character area 26 inside,, can appear on one second screen if perhaps the user wishes to see them.
In the above-described embodiments, first character of first table of DISPLAY ORDER appears in the current character position 24, and appears in the candidate character area 26 as the first candidate characters 28a.Alternatively, first character can only appear in the current character position 24, and the table 28 of candidate characters 28a-28f contains the remainder of first table.
The processing of above-mentioned generation candidate characters table is quite simple, yet has improved availability and precision.A special advantage of this scheme is relatively low calculated level and memory-size requirement.This makes them be applicable to portable electric appts, for example: mobile phone, PDA(Personal Digital Assistant), GPS (GPS) navigating instrument etc.
Particularly, relevant owing to not existing between the generation of these two tables, error recovery is unlikely propagated.Even contextual analysis is wrong, the user still can recover the result easily, because it only causes the rearrangement of written character recognizer output.
In a preferred embodiment, dictionary is the database of character double word (bi-gram).
In dictionary, the two-word record of identical bebinning character is less than number of thresholds.A reason that increases this restriction is limited memory-size.Another reason is to avoid wrong from normal moveout correction.If second table is oversize, then can will has the candidate of extremely hanging down confidence level tolerance mistakenly and be arranged to first candidate according to dependency rule.So dictionary only is made up of the two-word record with high frequency of occurrences.Also can use a plurality of dictionaries.System or user are according to the type selecting dictionary of input content.
If do not produce first table at step S118, then be different from and turn back to the processing mode that does not have table in step S102 and the display candidate characters district 26, processor can continue to produce second table at step S126, and step S130 can cause showing that second table is with as candidate characters table 28a-28f in candidate character area 26.So, seldom may exist under the situation of correct character, the user can select this processing immediately, rather than further written character.One optional " button " arranged probably, allow the user directly to produce second table of character late with possibility according to the character of previous input.
Consider two degrees of overlapping between the table, be used for determining that the rule of DISPLAY ORDER can be different from above-mentioned given rule.For example, the character in second table also can have relative confidence level tolerance mark, and DISPLAY ORDER can be considered the relative mark of two characters in the table.
For purpose of description in this instructions, the term " char " comprises the mark of any identification, and is not only letter or speech.
Preferably use the input pen (not shown) to come written character and the position of selecting on the touch-screen 12, because input pen has higher touch or contact precision than finger.
In addition, alphabet is Roman alphabet not necessarily, also can be the alphabet that Greece, Cyrillic, Arab or any other can be replaced by character (for example Chinese character, Japanese or other).Equally, numeric character can be arabic numeral, Chinese-character digital or other numeral.
Preceding detailed description only provides preferred embodiment, scope of the present invention, the scope of application or structure is not produced any restriction.Otherwise the front detailed description of the preferred embodiment can realize the preferred embodiments of the present invention in order to make those skilled in the art.Should be appreciated that under the prerequisite that does not depart from aim that appended claims of the present invention limits and scope, can on the function of unit and layout, carry out various changes.
Claims (15)
1. method that in electronic equipment, character is input to current character position in the character row, wherein said character is imported in the electronic equipment by writing, and this method comprises:
Reception may be corresponding to written character, by the first candidate characters table that character recognition produces, and this first table has initial order;
The second candidate characters table that reception may produce corresponding to written character, according to the context of current character position;
With the content of more described first and second tables of electronics mode to determine degree of overlapping therebetween; With
At least determine the DISPLAY ORDER of candidate characters in described first table according to determined degree of overlapping.
2. according to the method for being used for of claim 1, wherein produce first and second tables independently of one another at electronic equipment current character position input character in character row.
3. according to the method for being used for of claim 1,, then do not change between DISPLAY ORDER and the initial order if wherein between described first and second tables, do not exist overlappingly at electronic equipment current character position input character in character row.
4. according to the method for being used for of claim 1,, then do not change between DISPLAY ORDER and the initial order if wherein between described first and second tables, exist overlappingly fully at electronic equipment current character position input character in character row.
5. according to the method for being used for of claim 1 at electronic equipment current character position input character in character row, overlap if wherein between described first and second tables, exist, then DISPLAY ORDER begins with the character that all occurs in two tables, order is identical with order in initial order, be the character that only appears in first table subsequently, order is identical with order in initial order.
6. according to the method for being used for of claim 1, comprise that also detection is by writing the character of input electronic equipment at electronic equipment current character position input character in character row.
7. according to the method for being used for of claim 1, also comprise by character recognition producing the described first candidate characters table at electronic equipment current character position input character in character row.
8. according to the method for being used for of claim 7, wherein produce the described first candidate characters table and comprise at electronic equipment current character position input character in character row:
Reception is by writing the character that is input in the electronic equipment;
The character that receives is compared with the character set in one first database; With
According to a plurality of candidate characters of selecting to be used for described first table the character set of comparative result in described first database.
According to Claim 8 be used for method at electronic equipment current character position input character in character row, wherein produce the described first candidate characters table and also comprise:
Each candidate characters at first table produces the possibility mark, represents that this candidate characters is detected as the possibility of the supposition that writes character; With
Make its corresponding candidate characters of each possibility mark relevant.
10. according to the method at electronic equipment current character position input character in character row of being used for of claim 9, wherein said initial order is based on the possibility mark of the candidate characters of first table.
11. the method at electronic equipment current character position input character in character row of being used for according to claim 1 also comprises the described second candidate characters table that produces input character according to the context of current character position.
12., wherein produce described second table and also comprise according to the method for being used for of claim 11 at electronic equipment current character position input character in character row:
Inquiry contains one second database that set arranged in character; With
Under the situation of given at least a portion current character row, select candidate characters,, then can arrange set and produce character arrangement from described character if wherein said selection candidate characters is input to the current character position.
13. according to the method at electronic equipment current character position input character in character row of being used for of claim 12, wherein said character is arranged set and is comprised the character combination set.
14. according to the method for being used for of claim 13, wherein at electronic equipment current character position input character in character row
The set that set comprises the combination of two alphabetic characters arranged in described character; And
Select the candidate characters of second table to comprise:
Determine the tight front of which character current character position in character row, if any;
Determine in the described two alphabetic character composite sets which is that character with the tight front of current character position in the character row is beginning; With
Select second character from described two alphabetic characters combination, as the candidate characters of described second table, described two alphabetic characters combination is confirmed as beginning with the character of tight front in the character row.
15. the method at electronic equipment current character position input character in character row of being used for according to claim 1 also comprises:
On the display of electronic equipment, show described first table according to DISPLAY ORDER; With
Detect user's selection to a candidate characters from first table that shows; With
With the candidate characters input character selected capable in the current character position.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031371752A CN1303564C (en) | 2003-06-12 | 2003-06-12 | Identification of character input in improved electronic device |
| PCT/EP2004/051003 WO2004111921A1 (en) | 2003-06-12 | 2004-06-02 | Improved recognition for character input in an electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031371752A CN1303564C (en) | 2003-06-12 | 2003-06-12 | Identification of character input in improved electronic device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1567358A true CN1567358A (en) | 2005-01-19 |
| CN1303564C CN1303564C (en) | 2007-03-07 |
Family
ID=33546186
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB031371752A Expired - Lifetime CN1303564C (en) | 2003-06-12 | 2003-06-12 | Identification of character input in improved electronic device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN1303564C (en) |
| WO (1) | WO2004111921A1 (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008151466A1 (en) * | 2007-06-14 | 2008-12-18 | Google Inc. | Dictionary word and phrase determination |
| CN101221481B (en) * | 2007-01-12 | 2011-08-24 | 华硕电脑股份有限公司 | Handheld communication device and predictive input method thereof |
| US8010344B2 (en) | 2007-06-14 | 2011-08-30 | Google Inc. | Dictionary word and phrase determination |
| CN102298696A (en) * | 2010-06-28 | 2011-12-28 | 方正国际软件(北京)有限公司 | Character recognition method and system |
| CN102364425A (en) * | 2011-10-17 | 2012-02-29 | 宇龙计算机通信科技(深圳)有限公司 | Input method and device |
| CN102646201A (en) * | 2011-02-21 | 2012-08-22 | 富士施乐株式会社 | Character recognition apparatus and character recognition method |
| CN101149806B (en) * | 2006-09-19 | 2012-09-05 | 北京三星通信技术研究有限公司 | Method and device for hand writing identification post treatment using context information |
| CN101097487B (en) * | 2006-06-30 | 2013-08-07 | 2012244安大略公司 | Method for learning character fragments and related hand-hold electronic equipments |
| US9171234B2 (en) | 2006-06-30 | 2015-10-27 | Blackberry Limited | Method of learning a context of a segment of text, and associated handheld electronic device |
| US9286288B2 (en) | 2006-06-30 | 2016-03-15 | Blackberry Limited | Method of learning character segments during text input, and associated handheld electronic device |
| CN111079758A (en) * | 2019-07-11 | 2020-04-28 | 广东小天才科技有限公司 | A method and terminal device for determining the writing order in a dictation process |
| CN114880063A (en) * | 2022-06-09 | 2022-08-09 | 北京火山引擎科技有限公司 | A text display method, device, equipment, medium and product |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1804860A (en) * | 2005-01-13 | 2006-07-19 | 摩托罗拉公司 | Identification of script signs on user interface |
| US7805004B2 (en) | 2007-02-28 | 2010-09-28 | Microsoft Corporation | Radical set determination for HMM based east asian character recognition |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR1543777A (en) * | 1966-12-23 | 1900-01-01 | Ibm | Identifying characters by using context |
| US4058795A (en) * | 1972-10-03 | 1977-11-15 | International Business Machines Corporation | Method and apparatus for context-aided recognition |
| JPS5839377A (en) * | 1981-09-02 | 1983-03-08 | Toshiba Corp | Character recognizing device |
| US4654875A (en) * | 1983-05-23 | 1987-03-31 | The Research Foundation Of State University Of New York | System to achieve automatic recognition of linguistic strings |
| EP0325233B1 (en) * | 1988-01-18 | 1994-07-27 | Kabushiki Kaisha Toshiba | Character string recognition system |
| TW490643B (en) * | 1996-05-21 | 2002-06-11 | Hitachi Ltd | Estimated recognition device for input character string |
| US6219453B1 (en) * | 1997-08-11 | 2001-04-17 | At&T Corp. | Method and apparatus for performing an automatic correction of misrecognized words produced by an optical character recognition technique by using a Hidden Markov Model based algorithm |
-
2003
- 2003-06-12 CN CNB031371752A patent/CN1303564C/en not_active Expired - Lifetime
-
2004
- 2004-06-02 WO PCT/EP2004/051003 patent/WO2004111921A1/en not_active Ceased
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9286288B2 (en) | 2006-06-30 | 2016-03-15 | Blackberry Limited | Method of learning character segments during text input, and associated handheld electronic device |
| US9171234B2 (en) | 2006-06-30 | 2015-10-27 | Blackberry Limited | Method of learning a context of a segment of text, and associated handheld electronic device |
| CN101097487B (en) * | 2006-06-30 | 2013-08-07 | 2012244安大略公司 | Method for learning character fragments and related hand-hold electronic equipments |
| CN101149806B (en) * | 2006-09-19 | 2012-09-05 | 北京三星通信技术研究有限公司 | Method and device for hand writing identification post treatment using context information |
| CN101221481B (en) * | 2007-01-12 | 2011-08-24 | 华硕电脑股份有限公司 | Handheld communication device and predictive input method thereof |
| US8412517B2 (en) | 2007-06-14 | 2013-04-02 | Google Inc. | Dictionary word and phrase determination |
| CN101779200B (en) * | 2007-06-14 | 2013-03-20 | 谷歌股份有限公司 | Dictionary word and phrase determination method and device |
| WO2008151466A1 (en) * | 2007-06-14 | 2008-12-18 | Google Inc. | Dictionary word and phrase determination |
| US8010344B2 (en) | 2007-06-14 | 2011-08-30 | Google Inc. | Dictionary word and phrase determination |
| CN102298696B (en) * | 2010-06-28 | 2013-07-24 | 方正国际软件(北京)有限公司 | Character recognition method and system |
| CN102298696A (en) * | 2010-06-28 | 2011-12-28 | 方正国际软件(北京)有限公司 | Character recognition method and system |
| CN102646201A (en) * | 2011-02-21 | 2012-08-22 | 富士施乐株式会社 | Character recognition apparatus and character recognition method |
| CN102364425A (en) * | 2011-10-17 | 2012-02-29 | 宇龙计算机通信科技(深圳)有限公司 | Input method and device |
| CN111079758A (en) * | 2019-07-11 | 2020-04-28 | 广东小天才科技有限公司 | A method and terminal device for determining the writing order in a dictation process |
| CN111079758B (en) * | 2019-07-11 | 2023-10-31 | 广东小天才科技有限公司 | A method and terminal device for determining writing order during dictation |
| CN114880063A (en) * | 2022-06-09 | 2022-08-09 | 北京火山引擎科技有限公司 | A text display method, device, equipment, medium and product |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2004111921A1 (en) | 2004-12-23 |
| CN1303564C (en) | 2007-03-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1303564C (en) | Identification of character input in improved electronic device | |
| JP4519381B2 (en) | Keyboard system with automatic correction | |
| JP4527731B2 (en) | Virtual keyboard system with automatic correction function | |
| US9250805B2 (en) | Methods and systems for a linear character selection display interface for ambiguous text input | |
| CN101292214B (en) | Improved text entry in the shorthand interface on the keyboard | |
| US9557916B2 (en) | Keyboard system with automatic correction | |
| JP3141015B2 (en) | Handwritten kanji recognition method and apparatus | |
| US20080150910A1 (en) | Handwritten charater input device | |
| KR101753625B1 (en) | The method for preventing incorrect input in potable terminal and device thereof | |
| EP2698692A1 (en) | System and method for implementing sliding input of text based upon on-screen soft keyboard on electronic equipment | |
| JPH08506444A (en) | Handwriting recognition method of likely character strings based on integrated dictionary | |
| JP2007133884A5 (en) | ||
| EP1513053A2 (en) | Apparatus and method for character recognition | |
| CN100416593C (en) | Apparatus and method for handwriting recognition | |
| WO2008119244A1 (en) | Method and system for handwriting recognition | |
| US6035062A (en) | Character recognition method and apparatus | |
| KR101919841B1 (en) | Method and system for calibrating touch error | |
| JP4936650B2 (en) | Similar word search device, method thereof, program thereof, and information search device | |
| JP2014191355A (en) | Character input device, and character input method | |
| JP2018112931A (en) | Character recognition device, character recognition program, and character recognition method | |
| EP0519713A2 (en) | Character input method and apparatus | |
| US5991440A (en) | Method and apparatus for recognizing input handwritten characters associated with title information | |
| JP3675511B2 (en) | Handwritten character recognition method and apparatus | |
| JP2000155642A (en) | Software keyboard | |
| JP2006163830A (en) | Character recognizing device, character recognizing method, and character recognizing program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: MOTOROLA MOBILE CO., LTD. Free format text: FORMER OWNER: MOTOROLA INC. Effective date: 20110110 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: ILLINOIS, USA TO: ILLINOIS STATE, USA |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20110110 Address after: Illinois State Patentee after: MOTOROLA MOBILITY, Inc. Address before: Illinois Instrunment Patentee before: Motorola, Inc. |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| C56 | Change in the name or address of the patentee | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Illinois State Patentee after: MOTOROLA MOBILITY LLC Address before: Illinois State Patentee before: MOTOROLA MOBILITY, Inc. |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20160304 Address after: California, USA Patentee after: Google Technology Holdings LLC Address before: Illinois State Patentee before: MOTOROLA MOBILITY LLC |
|
| CX01 | Expiry of patent term |
Granted publication date: 20070307 |
|
| CX01 | Expiry of patent term |