[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN102541405A - Method and system for screening candidate area - Google Patents

Method and system for screening candidate area Download PDF

Info

Publication number
CN102541405A
CN102541405A CN2010106024603A CN201010602460A CN102541405A CN 102541405 A CN102541405 A CN 102541405A CN 2010106024603 A CN2010106024603 A CN 2010106024603A CN 201010602460 A CN201010602460 A CN 201010602460A CN 102541405 A CN102541405 A CN 102541405A
Authority
CN
China
Prior art keywords
candidate
candidate regions
candidate word
regions
word
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.)
Pending
Application number
CN2010106024603A
Other languages
Chinese (zh)
Inventor
刘廷超
陈波
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.)
Shenzhen Shiji Guangsu Information Technology Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
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 Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN2010106024603A priority Critical patent/CN102541405A/en
Publication of CN102541405A publication Critical patent/CN102541405A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Character Discrimination (AREA)

Abstract

The invention discloses a method for screening a candidate area. The method comprises the following steps of: recording handwritten note track data which are input in the candidate area; identifying the handwritten note track data by using a handwriting identification engine to obtain an identification result; and according to the identification result, filtering candidate words in the candidate area, re-sorting the remaining candidate words after filtration, and then displaying the candidate words in the candidate area. The invention also provides a system for screening the candidate area. By adoption of the method and the device, the input efficiency of a terminal can be improved, and an input experience of a user is improved.

Description

A kind of screening technique and system to candidate regions
Technical field
The present invention relates to the technical field of terminal input, relate in particular to a kind of screening technique and system candidate regions.
Background technology
In present terminal input operation, no matter be based on the input operation (comprising phonetic input, stroke input etc.) of full keyboard, also be based on the input operation (comprising phonetic input, stroke input etc.) of nine palace lattice, all can in candidate regions, produce many candidate word; Candidate word under association's pattern sometimes also can be many.The scheme of prior art is, the user slides and browses the content in the candidate regions through moving up and down button or finger, and selects required candidate word to submit to and go up screen to get into edit box.As shown in Figure 1, in the input area, by phonetic input " GCF ", then can in candidate regions, produce following candidate word, the user can select required candidate word one by one by last inferior mobile button.As shown in Figure 2 again, in the input field,, then can in candidate regions, produce following candidate word through phonetic input " D ", the user slides through finger and browses the content in the candidate regions, and selects required candidate word.
From above-mentioned existing input operation, can find out; Under the situation of some input; Option in the candidate regions still is a lot; Even have a quite a few page or leaf, and the required position of candidate word after leaning on very much of user sometimes, this just need could realize through page turn over operation many times (through frequent moving up and down the button page turning, perhaps sliding page through frequent finger).The speech operation of selecting of this complicacy can cause the efficient of terminal input to reduce, and brings bad input to experience to the user.
Summary of the invention
In view of this, fundamental purpose of the present invention is to provide a kind of screening technique and system to candidate regions, to improve the efficient of terminal input, improves user's input and experiences.
For achieving the above object, technical scheme of the present invention is achieved in that
The invention provides a kind of screening technique to candidate regions, this method comprises:
Recognition result is discerned and obtained to the hand-written notes track data of importing in the record candidate regions through handwriting recognition engine to said hand-written notes track data;
According to said recognition result, the candidate word in the said candidate regions is filtered, and be presented in the said candidate regions after will filtering remaining candidate word rearrangement.
Said recognition result is discerned and obtained to hand-written notes track data, be specially:
Said handwriting recognition engine carries out feature extraction according to said hand-written notes track data;
Feature database in characteristic of being extracted and the recognition template storehouse is compared, and the stroked character that obtains optimum coupling English alphabet character or coupling is as recognition result.
Saidly candidate word in the candidate regions is filtered, is specially according to recognition result:
The initial of the phonetic of all candidate word is mated in said coupling English alphabet character that comparison is obtained and the said candidate regions, and candidate word that will it fails to match deletes from said candidate regions, keeps the candidate word that the match is successful in the said candidate regions.
Saidly candidate word in the candidate regions is filtered, is specially according to recognition result:
The first stroke of all candidate word mates in the stroked character of the said coupling that comparison is obtained and the said candidate regions, and candidate word that will it fails to match deletes from said candidate regions, keeps the candidate word that the match is successful in the said candidate regions.
This method further comprises:
To filter after remaining candidate word rearrangement is presented in the said candidate regions; Filter once more if desired; Then write down the hand-written notes track data of importing in the candidate regions again, and new recognition result is discerned and obtained to the hand-written notes track data of new input through handwriting recognition engine; According to said new recognition result, the candidate word in the current candidate regions is filtered once more, and be presented in the said candidate regions after will filtering remaining candidate word rearrangement.
The recognition result that said basis is new filters the candidate word in the current candidate regions once more, is specially:
If said new recognition result is the English alphabet character; The initial of then not carrying out the overmatching operation in the phonetic with all candidate word in said English alphabet character and the current candidate regions is mated; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions;
If said new recognition result is a stroked character; The first stroke that does not then carry out the overmatching operation in the stroke with all candidate word in said stroked character and the current candidate regions matees; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions.
The present invention also provides a kind of screening system to candidate regions, and this system comprises:
Data recordin module is used for writing down the hand-written notes track data that candidate regions is imported;
Handwriting recognition engine is used for recognition result is discerned and obtained to said hand-written notes track data;
The candidate word filtering module is used for based on said recognition result the candidate word in the said candidate regions being filtered, and is presented in the said candidate regions after will filtering remaining candidate word rearrangement.
Said handwriting recognition engine further comprises:
Characteristic extracting module is used for carrying out feature extraction according to said hand-written notes track data;
The characteristic comparing module is used for the feature database in characteristic of being extracted and recognition template storehouse is compared, and the stroked character that obtains optimum coupling English alphabet character or coupling is as recognition result.
Said candidate word filtering module is further used for; The initial of the phonetic of all candidate word is mated in said coupling English alphabet character that comparison is obtained and the said candidate regions; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the said candidate regions from said candidate regions.
Said candidate word filtering module is further used for; The first stroke of all candidate word matees in the stroked character of the said coupling that comparison is obtained and the said candidate regions; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the said candidate regions from said candidate regions.
Filtering after remaining candidate word rearrangement is presented in the said candidate regions, filter once more if desired, then said data recordin module writes down the hand-written notes track data of importing in the candidate regions again; New recognition result is discerned and obtained to said handwriting recognition engine to the hand-written notes track data of new input; Said candidate word filtering module filters the candidate word in the current candidate regions based on said new recognition result once more, and is presented in the said candidate regions after will filtering remaining candidate word rearrangement.
Said candidate word filtering module is further used for,
When said new recognition result is the English alphabet character; The initial of not carrying out the overmatching operation in the phonetic with all candidate word in said English alphabet character and the current candidate regions is mated; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions;
When said new recognition result is stroked character; The first stroke that does not carry out the overmatching operation in the stroke with all candidate word in said stroked character and the current candidate regions matees; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions.
A kind of screening technique and system provided by the present invention to candidate regions; The situation that candidate word when importing to the terminal in the candidate regions is too much, through the initial of handwriting recognition candidate word phonetic or the beginning stroke of candidate word, can the rapid screening candidate word; Make the candidate word quantity in the candidate regions reduce; The user can find required candidate word fast, thereby improves the efficient of terminal input, improves user's input and experiences.
Description of drawings
Fig. 1 is for selecting the synoptic diagram of candidate word through moving up and down button in the prior art;
Fig. 2 selects the synoptic diagram of candidate word through pointing to slide in the prior art;
Fig. 3 is the process flow diagram of a kind of screening technique to candidate regions of the present invention;
Fig. 4 is for streaking the synoptic diagram one of filtering candidate regions through input pen in the embodiment of the invention;
Fig. 5 is for streaking the synoptic diagram two of filtering candidate regions through input pen in the embodiment of the invention;
Fig. 6 passes through the synoptic diagram one that input alphabet filters candidate regions in the embodiment of the invention;
Fig. 7 passes through the synoptic diagram two that input alphabet filters candidate regions in the embodiment of the invention;
Fig. 8 passes through the synoptic diagram three that input alphabet filters candidate regions in the embodiment of the invention;
Fig. 9 is the composition structural representation of a kind of screening system to candidate regions of the present invention.
Embodiment
Below in conjunction with accompanying drawing and specific embodiment technical scheme of the present invention further is discussed in more detail.
A kind of screening technique provided by the present invention to candidate regions, as shown in Figure 3, mainly may further comprise the steps:
Step 301, recognition result is discerned and obtained to the hand-written notes track data of importing in the record candidate regions through handwriting recognition engine to hand-written notes track data.
When carrying out input operations such as phonetic, stroke the user; If produced the expansion of candidate word, then the user can be at the initial (several letters before also can being) or the first stroke (several strokes before also can being) of the quick hand-written candidate word phonetic that will select of candidate's spreading area (abbreviation candidate regions); Realize the hand-written notes track data imported in the terminal records candidate regions of input operation, and recognition result is discerned and obtained to hand-written notes track data through handwriting recognition engine.
Wherein, the handwriting recognition engine at terminal is discerned hand-written notes track data, and the operation that obtains recognition result is specially:
Handwriting recognition engine carries out feature extraction according to the hand-written notes track data that the user imports in candidate regions; Feature database in characteristic of being extracted and the pre-configured recognition template storehouse is compared, obtain the optimum coupling English alphabet character or the stroked character of coupling, and elect the coupling English alphabet character of said optimum or the stroked character of coupling as recognition result.
Step 302 according to recognition result, is filtered the candidate word in the candidate regions, and is presented in the candidate regions after will filtering remaining candidate word rearrangement.
If the recognition result that obtains in the step 301 is coupling English alphabet character; Then the initial of the phonetic of all candidate word is mated in terminal said coupling English alphabet character that comparison is obtained and the candidate regions; And the candidate word that will mate failure is deleted from candidate regions; Keep the successful candidate word of coupling in the candidate regions, and be presented in the candidate regions after the candidate word that the remains rearrangement;
If the recognition result that obtains in the step 301 is the stroked character of coupling; Then the first stroke of all candidate word matees in the stroked character of terminal said coupling that comparison is obtained and the candidate regions; And the candidate word that will mate failure is deleted from candidate regions; Keep the successful candidate word of coupling in the candidate regions, and be presented in the candidate regions after the candidate word that the remains rearrangement.
In addition; To filter after remaining candidate word rearrangement is presented in the candidate regions; Filter once more if desired; Then the terminal need be write down the hand-written notes track data of importing in the candidate regions again, and through handwriting recognition engine new recognition result is discerned and obtained to the hand-written notes track data of new input; According to said new recognition result, the candidate word in the current candidate regions is filtered once more, and be presented in the candidate regions after will filtering remaining candidate word rearrangement.Wherein, The operation of filtering once more is specially: if said new recognition result is the English alphabet character; The initial of then not carrying out the overmatching operation in the phonetic with all candidate word in said English alphabet character and the current candidate regions is mated; And the candidate word that will mate failure deletes from current candidate regions, keeps the successful candidate word of coupling in the current candidate regions, and the candidate word that remains is presented in the candidate regions after resequencing; If said new recognition result is a stroked character; The first stroke that does not then carry out the overmatching operation in the stroke with all candidate word in said stroked character and the current candidate regions matees; And the candidate word that will mate failure is deleted from current candidate regions; Keep the successful candidate word of coupling in the current candidate regions, and be presented in the candidate regions after the candidate word that the remains rearrangement.
Be example below with the portable terminal, and combine specific embodiment that the screening technique of above-mentioned candidate regions is further set forth in detail.
First shown in Figure 4, when the user on the mobile terminal through squares Pinyin input "D", it will appear in the candidate region many candidate words; Suppose you want to enter the word "but" because "but" the word first stroke is "Pie", that is written, then the user can quickly enter in the candidate region strokes "Pie" (Figure 4).Handwriting recognition engine of the mobile terminal in the candidate area based on the user input handwritten notes trajectory data for feature extraction, the extracted feature with a pre-configured in the library recognition template feature database for comparison, the recognition result is obtained stroke " Pie "; Subsequently, the mobile terminal of the stroke than the resulting" Pie "all the candidate and the candidate region of the first stroke to match words and failure to match the candidate word deleted from the candidate area, the candidate successfully retain match The candidate word, and retained the reordered candidate word displayed in the candidate region, the candidate region after filtering as shown in Figure 5, only "a", "more", "get", "hungry", "minute", "but" these candidate words.Thus, the user can from candidate regions, find very fast " but " word, thereby fast will " but " word submits to and goes up screen to get into edit box.
It should be noted that, in the results shown in Figure 5 on the basis of the candidate, if the user needs to be filtered again, the candidate region can be continued rapid input strokes "Shu."Handwriting recognition engine of the mobile terminal in the candidate area based on the user input handwritten notes trajectory data for feature extraction, the extracted feature with a pre-configured in the library recognition template feature database for comparison, a new recognition result is obtained stroke "Shu"; Subsequently, the mobile terminal of the stroke than the resulting "Shu" and the current candidate area strokes all the candidate words matching operations are not carried out the first stroke (ie, "the" word "Shu" strokes, "many" characters "a" stroke, "get" the word "Pie" strokes, "hungry" character "a" stroke, "minute" word
Figure BDA0000040157470000061
strokes, "but" the word "Shu" stroke) matching and fails to match the candidate word deleted from the current candidate region, candidate region to retain the current matching successful candidate word and preserved reordered candidate words shown in the candidate area.At this moment, only remaining " ", " but " these two candidate word, thus, just be more convenient for searching of user.
In addition, users can enter multiple strokes, for example: an input strokes "Ren", then the mobile terminal needs to be "Ren" and the candidate regions for all candidate words to match the first two strokes, resulting in only after a filter after left "and", "but" this two candidate words, the input efficiency is further improved.
As shown in Figure 7 again, when the user passes through the phonetic input on portable terminal, can in candidate regions, show many candidate word; What suppose that the user wants to import is " you " word, because the first letter of pinyin of " you " word is " n ", this moment, the user can input " n " or " N " (shown in Figure 7) fast on candidate regions.The handwriting recognition engine of portable terminal carries out feature extraction according to the hand-written notes track data that the user imports in candidate regions, the feature database in characteristic of being extracted and the pre-configured recognition template storehouse is compared, and the recognition result that obtains is letter " n "; Subsequently; The initial of all candidate word phonetics is mated in the said letter " n " that portable terminal obtains comparison and the candidate regions; And the candidate word that will mate failure deletes from candidate regions, keeps the successful candidate word of coupling in the candidate regions, and the candidate word that remains is presented in the candidate regions after resequencing; Candidate regions through after filtering is as shown in Figure 8, only remaining " you ", " ", " plan ", " Buddhist nun ", " mud ", " contrary ", " girl ", " Ni ", these several candidate word.Thus, the user can find " you " word very fast from candidate regions, goes up screen to get into edit box thereby fast " you " word is submitted to.
It should be noted that, in the results shown in Figure 8 on the basis of the candidate, if the user needs to be filtered again, the candidate region can be continued rapid input strokes "Pie."Handwriting recognition engine of the mobile terminal in the candidate area based on the user input handwritten notes trajectory data for feature extraction, the extracted feature with a pre-configured in the library recognition template feature database for comparison, a new recognition result is obtained stroke "Pie"; Subsequently, the mobile terminal of the stroke than the resulting "Pie" and the current candidate area strokes all the candidate words matching operations are not carried out the first stroke (ie, "you" word "Pie" strokes, "mean" word "Shu" strokes, "to be" the word of the "Shu" strokes, "Benny" the word "one" stroke, "mud" the word "dish" strokes, "reverse" the word "dish" stroke, etc. etc.) match, and the failure to match the candidate words from the current candidate region deleted retained in the current candidate area successfully matched candidate word, and retained the reordered candidate words displayed in the candidate region.At this moment, only remaining " you ", " Ni " these two candidate word, thus, just be more convenient for searching of user.
Corresponding above-mentioned screening technique to candidate regions, the present invention also provides a kind of screening system to candidate regions, and is as shown in Figure 9, and this system comprises: data recordin module 11, handwriting recognition engine 12 and candidate word filtering module 13 are formed.Wherein, data recordin module 11 is used for writing down the hand-written notes track data that candidate regions is imported.Handwriting recognition engine 12 is used for recognition result is discerned and obtained to hand-written notes track data.Candidate word filtering module 13 is used for based on recognition result, the candidate word in the candidate regions is filtered, and be presented in the candidate regions after will filtering remaining candidate word rearrangement.
Preferable, handwriting recognition engine 12 further comprises: characteristic extracting module 121 and characteristic comparing module 122.Characteristic extracting module 121 is used for carrying out feature extraction according to hand-written notes track data.Characteristic comparing module 122 is used for the feature database in characteristic of being extracted and recognition template storehouse is compared, and the stroked character that obtains optimum coupling English alphabet character or coupling is as recognition result.
Accordingly; Candidate word filtering module 13 is further used for; The initial of the phonetic of all candidate word is mated in coupling English alphabet character that comparison is obtained and the candidate regions, and the candidate word that will mate failure deletes from candidate regions, keeps the successful candidate word of coupling in the candidate regions; Perhaps, the first stroke of all candidate word matees in the stroked character of the said coupling that comparison is obtained and the candidate regions, and the candidate word that will mate failure deletes from candidate regions, keeps the successful candidate word of coupling in the candidate regions.
Preferable, filtering after remaining candidate word rearrangement is presented in the candidate regions, filter once more if desired, then data recordin module 11 writes down the hand-written notes track data of importing in the candidate regions again; New recognition result is discerned and obtained to the hand-written notes track data of 12 pairs of new inputs of handwriting recognition engine; Candidate word filtering module 13 filters the candidate word in the current candidate regions based on new recognition result once more, and is presented in the candidate regions after will filtering remaining candidate word rearrangement.Concrete; When new recognition result is the English alphabet character; The initial of not carrying out the overmatching operation in the phonetic of candidate word filtering module 13 with all candidate word in said English alphabet character and the current candidate regions is mated; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions; When new recognition result is stroked character; The first stroke that does not carry out the overmatching operation in the stroke of candidate word filtering module 13 with all candidate word in said stroked character and the current candidate regions matees; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions.
In sum; The present invention is directed to the too much situation of candidate word in when input candidate regions of terminal, through the initial of handwriting recognition candidate word phonetic or the beginning stroke of candidate word, can the rapid screening candidate word; Make the candidate word quantity in the candidate regions reduce; The user can find required candidate word fast, thereby improves the efficient of terminal input, improves user's input and experiences.
The above is merely preferred embodiment of the present invention, is not to be used to limit protection scope of the present invention.

Claims (12)

1. the screening technique to candidate regions is characterized in that, this method comprises:
Recognition result is discerned and obtained to the hand-written notes track data of importing in the record candidate regions through handwriting recognition engine to said hand-written notes track data;
According to said recognition result, the candidate word in the said candidate regions is filtered, and be presented in the said candidate regions after will filtering remaining candidate word rearrangement.
2. according to the said screening technique of claim 1, it is characterized in that, said recognition result is discerned and obtained to hand-written notes track data, be specially candidate regions:
Said handwriting recognition engine carries out feature extraction according to said hand-written notes track data;
Feature database in characteristic of being extracted and the recognition template storehouse is compared, and the stroked character that obtains optimum coupling English alphabet character or coupling is as recognition result.
3. according to the said screening technique of claim 2, it is characterized in that, saidly candidate word in the candidate regions filtered, be specially according to recognition result to candidate regions:
The initial of the phonetic of all candidate word is mated in said coupling English alphabet character that comparison is obtained and the said candidate regions, and candidate word that will it fails to match deletes from said candidate regions, keeps the candidate word that the match is successful in the said candidate regions.
4. according to the said screening technique of claim 2, it is characterized in that, saidly candidate word in the candidate regions filtered, be specially according to recognition result to candidate regions:
The first stroke of all candidate word mates in the stroked character of the said coupling that comparison is obtained and the said candidate regions, and candidate word that will it fails to match deletes from said candidate regions, keeps the candidate word that the match is successful in the said candidate regions.
5. according to each said screening technique of claim 1 to 4, it is characterized in that this method further comprises to candidate regions:
To filter after remaining candidate word rearrangement is presented in the said candidate regions; Filter once more if desired; Then write down the hand-written notes track data of importing in the candidate regions again, and new recognition result is discerned and obtained to the hand-written notes track data of new input through handwriting recognition engine; According to said new recognition result, the candidate word in the current candidate regions is filtered once more, and be presented in the said candidate regions after will filtering remaining candidate word rearrangement.
6. according to the said screening technique of claim 5, it is characterized in that the recognition result that said basis is new filters the candidate word in the current candidate regions once more, is specially to candidate regions:
If said new recognition result is the English alphabet character; The initial of then not carrying out the overmatching operation in the phonetic with all candidate word in said English alphabet character and the current candidate regions is mated; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions;
If said new recognition result is a stroked character; The first stroke that does not then carry out the overmatching operation in the stroke with all candidate word in said stroked character and the current candidate regions matees; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions.
7. the screening system to candidate regions is characterized in that, this system comprises:
Data recordin module is used for writing down the hand-written notes track data that candidate regions is imported;
Handwriting recognition engine is used for recognition result is discerned and obtained to said hand-written notes track data;
The candidate word filtering module is used for based on said recognition result the candidate word in the said candidate regions being filtered, and is presented in the said candidate regions after will filtering remaining candidate word rearrangement.
8. according to the said screening system of claim 7, it is characterized in that said handwriting recognition engine further comprises to candidate regions:
Characteristic extracting module is used for carrying out feature extraction according to said hand-written notes track data;
The characteristic comparing module is used for the feature database in characteristic of being extracted and recognition template storehouse is compared, and the stroked character that obtains optimum coupling English alphabet character or coupling is as recognition result.
9. said according to Claim 8 screening system to candidate regions; It is characterized in that; Said candidate word filtering module is further used for; The initial of the phonetic of all candidate word is mated in said coupling English alphabet character that comparison is obtained and the said candidate regions, and the candidate word that will mate failure deletes from said candidate regions, keeps the successful candidate word of coupling in the said candidate regions.
10. said according to Claim 8 screening system to candidate regions; It is characterized in that; Said candidate word filtering module is further used for; The first stroke of all candidate word matees in the stroked character of the said coupling that comparison is obtained and the said candidate regions, and the candidate word that will mate failure deletes from said candidate regions, keeps the successful candidate word of coupling in the said candidate regions.
11. based on each said screening system of claim 7 to 10 to candidate regions; It is characterized in that; To filter after remaining candidate word rearrangement is presented in the said candidate regions; Filter once more if desired, then said data recordin module writes down the hand-written notes track data of importing in the candidate regions again; New recognition result is discerned and obtained to said handwriting recognition engine to the hand-written notes track data of new input; Said candidate word filtering module filters the candidate word in the current candidate regions based on said new recognition result once more, and is presented in the said candidate regions after will filtering remaining candidate word rearrangement.
12., it is characterized in that said candidate word filtering module is further used for according to the said screening system of claim 11 to candidate regions,
When said new recognition result is the English alphabet character; The initial of not carrying out the overmatching operation in the phonetic with all candidate word in said English alphabet character and the current candidate regions is mated; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions;
When said new recognition result is stroked character; The first stroke that does not carry out the overmatching operation in the stroke with all candidate word in said stroked character and the current candidate regions matees; And the candidate word that will mate failure is deleted the successful candidate word of reservation coupling in the current candidate regions from current candidate regions.
CN2010106024603A 2010-12-23 2010-12-23 Method and system for screening candidate area Pending CN102541405A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2010106024603A CN102541405A (en) 2010-12-23 2010-12-23 Method and system for screening candidate area

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2010106024603A CN102541405A (en) 2010-12-23 2010-12-23 Method and system for screening candidate area

Publications (1)

Publication Number Publication Date
CN102541405A true CN102541405A (en) 2012-07-04

Family

ID=46348413

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2010106024603A Pending CN102541405A (en) 2010-12-23 2010-12-23 Method and system for screening candidate area

Country Status (1)

Country Link
CN (1) CN102541405A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104765526A (en) * 2015-03-18 2015-07-08 百度在线网络技术(北京)有限公司 Method and device for filtering candidate entries
CN113641253A (en) * 2021-07-09 2021-11-12 北京搜狗科技发展有限公司 Method, apparatus and medium for screening candidate items

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1997020284A1 (en) * 1995-12-01 1997-06-05 Motorola Inc. Method and system for velocity-based handwriting recognition
CN101071338A (en) * 2007-02-07 2007-11-14 腾讯科技(深圳)有限公司 Word input method and system
EP2088536A1 (en) * 2008-02-08 2009-08-12 ExB Asset Management GmbH Text input system and method involving finger-based handwriting recognition and word prediction
CN101620480A (en) * 2008-06-30 2010-01-06 戴静芬 Method for realizing handwritten stroke input on touch screen
CN101794398A (en) * 2010-04-02 2010-08-04 华为终端有限公司 Method, system and device for recognizing handwriting input characters

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1997020284A1 (en) * 1995-12-01 1997-06-05 Motorola Inc. Method and system for velocity-based handwriting recognition
CN101071338A (en) * 2007-02-07 2007-11-14 腾讯科技(深圳)有限公司 Word input method and system
EP2088536A1 (en) * 2008-02-08 2009-08-12 ExB Asset Management GmbH Text input system and method involving finger-based handwriting recognition and word prediction
CN101620480A (en) * 2008-06-30 2010-01-06 戴静芬 Method for realizing handwritten stroke input on touch screen
CN101794398A (en) * 2010-04-02 2010-08-04 华为终端有限公司 Method, system and device for recognizing handwriting input characters

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104765526A (en) * 2015-03-18 2015-07-08 百度在线网络技术(北京)有限公司 Method and device for filtering candidate entries
CN113641253A (en) * 2021-07-09 2021-11-12 北京搜狗科技发展有限公司 Method, apparatus and medium for screening candidate items

Similar Documents

Publication Publication Date Title
CN101430623B (en) Input-handwriting automatic transformation system and method
CN103268363B (en) A kind of Chinese calligraphy's image search method based on elastic HOG feature and DDTW coupling
CN103577843B (en) A kind of aerial hand-written character string recognition methods
US10296170B2 (en) Electronic apparatus and method for managing content
EP2698692A1 (en) System and method for implementing sliding input of text based upon on-screen soft keyboard on electronic equipment
JP5774558B2 (en) Handwritten document processing apparatus, method and program
CN111857332A (en) Method and device for acquiring note information
JP2007317022A (en) Handwritten character processor and method for processing handwritten character
CN103440299A (en) Quick information input method based on focus context associational words
CN100555312C (en) Utilize charcter topology information to carry out the method and apparatus of the handwriting recognition of aftertreatment
US20160321238A1 (en) Electronic device, method and storage medium
CN102419645A (en) Method and device for inputting Chinese characters in combination and palm terminal
CN104598148B (en) A kind of computer contact screen handwritten Chinese character method
Garz et al. Creating ground truth for historical manuscripts with document graphs and scribbling interaction
CN102073884A (en) Handwriting recognition method, system and handwriting recognition terminal
CN101149806B (en) Method and device for hand writing identification post treatment using context information
KR102373884B1 (en) Image data processing method for searching images by text
CN103019590A (en) Method, client side and system for inputting handwriting characters and character strings
CN101281449A (en) Hand-written character recognizing method and system
CN106953913A (en) A kind of information-pushing method and mobile terminal
CN103455527A (en) Handwritten document retrieval apparatus, handwritten document retrieval method and recording medium
CN107219935B (en) Chinese character input system and method for continuously writing Chinese characters and supporting interaction
CN102193718A (en) Chinese character handwriting input method and Chinese character handwriting input system
CN104077011B (en) Correlating method and terminal device between a kind of same type document
CN104978577B (en) Information processing method, device and electronic equipment

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
ASS Succession or assignment of patent right

Owner name: SHENZHEN SHIJI LIGHT SPEED INFORMATION TECHNOLOGY

Free format text: FORMER OWNER: TENGXUN SCI-TECH (SHENZHEN) CO., LTD.

Effective date: 20131017

C41 Transfer of patent application or patent right or utility model
COR Change of bibliographic data

Free format text: CORRECT: ADDRESS; FROM: 518044 SHENZHEN, GUANGDONG PROVINCE TO: 518057 SHENZHEN, GUANGDONG PROVINCE

TA01 Transfer of patent application right

Effective date of registration: 20131017

Address after: A Tencent Building in Shenzhen Nanshan District City, Guangdong streets in Guangdong province science and technology 518057 16

Applicant after: Shenzhen Shiji Guangsu Information Technology Co., Ltd.

Address before: Shenzhen Futian District City, Guangdong province 518044 Zhenxing Road, SEG Science Park 2 East Room 403

Applicant before: Tencent Technology (Shenzhen) Co., Ltd.

RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20120704