CN107577357B - A kind of automatic matching method and system for inputting information - Google Patents
A kind of automatic matching method and system for inputting information Download PDFInfo
- Publication number
- CN107577357B CN107577357B CN201710709668.7A CN201710709668A CN107577357B CN 107577357 B CN107577357 B CN 107577357B CN 201710709668 A CN201710709668 A CN 201710709668A CN 107577357 B CN107577357 B CN 107577357B
- Authority
- CN
- China
- Prior art keywords
- sequence
- click location
- location sequence
- click
- dummy keyboard
- 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.)
- Active
Links
Landscapes
- Input From Keyboards Or The Like (AREA)
Abstract
The invention discloses a kind of automatic matching method and system for inputting information, wherein, the method includes:The click location sequence being applied on touch screen is obtained, and Primary Location is carried out based on the dummy keyboard described in the click location sequence pair on touch screen;Matching obtains the corresponding target vocabulary of the click location sequence in dictionary;Based on the target vocabulary, the new position of the dummy keyboard after Primary Location and new posture are determined.The technical solution that the application provides can be laid out according to the situation of key position and matching dictionary come adjust automatically.
Description
Technical field
The present invention relates to human-computer interaction technique field, more particularly to a kind of automatic matching method and system for inputting information.
Background technology
Current input method automatic error-correcting mechanism, there are two types of types:
1. the input method error correction based on physical button
Such as the laptop using physical button, search dog input method can realize phonetic automatic error-correcting, as phonetic is defeated
When entering " flat " word, " pnig " can be corrected as " ping " automatically.The major premise of this kind of input method error correction is:The button of user is big absolutely
Part is all correct.A small amount of wrong key need to be only directed to carry out error correction or be adjusted for keystroke sequence.
2. the input method error correction based on virtual key
Such as the SwiftKey English input methods on mobile phone, when inputting English " something ", input by mistake for
" domrrhonf ", input method energy automatic error-correcting.The major premise of this kind of error correction is:User wrong key deviate correct key mapping
Apart from very little.If the deviation from more than one key mapping of distance, then error correction effect is poor.
It is the defects of these automatic error-correctings currently:
It can only be used under key position and the smaller scene of correct position deviation;
Keyboard does not have adaptability, while automatic error-correcting, can not optimize the size and posture of keyboard;
It must show keyboard and keyboard is in fixed position.
Invention content
In order to solve problem of the prior art, an embodiment of the present invention provides it is a kind of input information automatic matching method and
System.The technical solution is as follows:
A kind of automatic matching method for inputting information, the method includes:
The click location sequence being applied on touch screen is obtained, and is based on described in the click location sequence pair on touch screen
Dummy keyboard carry out Primary Location;
Matching obtains the corresponding target vocabulary of the click location sequence in dictionary;
Based on the target vocabulary, the new position of the dummy keyboard after Primary Location and new posture are determined.
Further, Primary Location is carried out to the dummy keyboard on the touch screen to include:
When the total number of the click location sequence is 1, the dummy keyboard of previous frame is put down according to translation matrix
Shifting is handled;Wherein, the translation matrix is determined by the vector of keyboard centerline to the specified coordinate of previous frame dummy keyboard;The finger
Position fixing determines according to the following equation:
Wherein, P be the click location sequence, PiFor the value of i-th of sequence in the click location sequence, | P | for institute
State the total number of sequence in click location sequence.
Further, Primary Location is carried out to the dummy keyboard on the touch screen to include:
When the total number of the click location sequence is more than or equal to 2, to the dummy keyboard of previous frame according to translation
Matrix and spin matrix carry out translation processing and rotation processing respectively;
Wherein, the translation matrix is determined by the vector of keyboard centerline to the specified coordinate of previous frame dummy keyboard;It is described
Specified coordinate determines according to the following equation:
Wherein, P be the click location sequence, PiFor the value of i-th of sequence in the click location sequence, | P | for institute
State the total number of sequence in click location sequence;
The spin matrix determines in the following manner:
All the points in the click location sequence are fitted, to obtain the corresponding straight line of the click location sequence;
The spin matrix is determined by the straight line and horizontal angle.
Further, the method further includes:
When the range residing for the sequence in the click location sequence is less than specified range threshold value or the straight line and water
When the angle of horizontal line is more than specified angle threshold value, rotation processing is not carried out to the previous frame dummy keyboard.
Further, matching obtains the corresponding target vocabulary of the click location sequence and includes in dictionary:
Determine that the constant hard iron of the corresponding consecutive points distance of the click location sequence, consecutive points angle, point set and grouping are estimated
Value function, and the click is determined based on consecutive points distance, consecutive points angle, the constant hard iron of point set and grouping evaluation function
The corresponding comprehensive assessment value of position sequence;
The comprehensive assessment value and the assessed value of vocabulary each in dictionary are compared, and will be with the comprehensive assessment value
The corresponding vocabulary of assessed value of difference minimum is as the corresponding target vocabulary of the click location sequence.
Further, the consecutive points distance determines according to the following equation:
Wherein, piRepresent i-th of sequence in the click location sequence, pi+1Represent i+1 in the click location sequence
A sequence, diRepresent piAnd pi+1The distance between, N represents the quantity of sequence in the click location sequence;
The consecutive points angle determines according to the following equation:
Wherein, pi-1Represent (i-1)-th sequence in the click location sequence, θ represents piAnd pi-1The straight line and p of formationiWith
pi+1Angle between the straight line of formation;
The constant hard iron of point set determines according to the following equation:
I3=(y30+3y)2+(3y21-y03)2
Wherein, I3Represent third class not bending moment,
Wherein, M represents the number of point, xi, yiI-th point of x coordinate and y-coordinate is respectively referred to, the x that x, y respectively refer to point set is sat
Mark average value and y-coordinate average value, f (xi,yi) refer to the weighting function of the point;
The grouping evaluation function determines in the following manner:
The grouping evaluation function is determined according to the following equation:
Wherein, f represents the grouping evaluation function, CiRepresent the number of button in the corresponding piecemeal keyboard of finger;
Or
The grouping evaluation function is determined according to the following equation:
Wherein, F represents the pressure of finger, and S represents the contact area of finger, and P represents the position of finger, and g (*) represents fitting
Function.
Further, it is determined that the new position of dummy keyboard after Primary Location and new posture include:
Ask for the minimum value of following formula:
Wherein, (xi,yi) represent previous frame dummy keyboard in i-th of button coordinate value, θ represent rotation parameter, TxIt represents
The translation parameters of x-axis, TyRepresent the translation parameters of y-axis, S represents zooming parameter, and s (*) is represented in previous frame dummy keyboard i-th
The abscissa of a button converted after value, t (*) represents to carry out the ordinate of i-th button in previous frame dummy keyboard
Value after transformation, N represent the number of button.
Further, the method further includes:
The specified gesture being applied in response to user on the touch screen, transparent window is shown on the touch screen;Its
In, show candidate word in the transparent window;
The click location sequence that user is currently applied on the touch screen is obtained, and based on the current click location
Sequence changes the content of the candidate word.
Further, the content that the candidate word is changed based on the current click location sequence is included:
It horizontally slips if the current click location sequence characterization list refers to, other wait is shown in the transparent window
Select word;
If the current click location sequence characterization clicks current candidate word, the candidate word that will click on is filled in
In information input frame;
If the current click location sequence characterization list is directed toward lower slider, the candidate word of current presentation is deleted;
If the current click location sequence characterization list refers to upward sliding, it is defeated that current candidate word is filled in information
Enter in frame.
A kind of automatic patching system for inputting information, the system comprises:
Click location retrieval unit, for obtaining the click location sequence being applied on touch screen, and based on described
Dummy keyboard described in click location sequence pair on touch screen carries out Primary Location;
Target vocabulary matching unit obtains the corresponding target vocabulary of the click location sequence for being matched in dictionary;
Position and attitude determination unit for being based on the target vocabulary, determines the new position of the dummy keyboard after Primary Location
It puts and new posture.
Therefore the technical solution that the application provides, have following advantage compared with the prior art:
1) independent of keyboard size and posture;
2) according to the input of user, keyboard can adjust size and posture in real time;
3) good error correction effect can be had by deviateing correct key mapping any distance;
4) keyboard can not be shown.
Technical solution provided by the invention can be applicable to the equipment such as mobile phone, tablet computer, VR/AR, microcomputer, projection
On, when inputting word, do not require to show keyboard layout (can also show) in advance, user needs to be familiar with keyboard layout (if not
It is familiar then need to show keyboard in advance), during user inputs word, keyboard can be according to key position and matching dictionary
Situation carrys out adjust automatically layout.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some embodiments of the present invention, for
For those of ordinary skill in the art, without creative efforts, other are can also be obtained according to these attached drawings
Attached drawing.
Fig. 1 is the automatic matching method flow chart that information is inputted in embodiment of the present invention;
Fig. 2 is dummy keyboard schematic diagram in embodiment of the present invention;
Fig. 3 is the functional block diagram for the automatic patching system that information is inputted in embodiment of the present invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.
Referring to Fig. 1, the application provides a kind of automatic matching method for inputting information, the method includes:
S1:The click location sequence being applied on touch screen is obtained, and based on touch described in the click location sequence pair
Dummy keyboard on screen carries out Primary Location;
S2:Matching obtains the corresponding target vocabulary of the click location sequence in dictionary;
S3:Based on the target vocabulary, the new position of the dummy keyboard after Primary Location and new posture are determined.
In the present embodiment, Primary Location is carried out to the dummy keyboard on the touch screen to include:
When the total number of the click location sequence is 1, the dummy keyboard of previous frame is put down according to translation matrix
Shifting is handled;Wherein, the translation matrix is determined by the vector of keyboard centerline to the specified coordinate of previous frame dummy keyboard;The finger
Position fixing determines according to the following equation:
Wherein, P be the click location sequence, PiFor the value of i-th of sequence in the click location sequence, | P | for institute
State the total number of sequence in click location sequence.
In the present embodiment, Primary Location is carried out to the dummy keyboard on the touch screen to include:
When the total number of the click location sequence is more than or equal to 2, to the dummy keyboard of previous frame according to translation
Matrix and spin matrix carry out translation processing and rotation processing respectively;
Wherein, the translation matrix is determined by the vector of keyboard centerline to the specified coordinate of previous frame dummy keyboard;It is described
Specified coordinate determines according to the following equation:
Wherein, P be the click location sequence, PiFor the value of i-th of sequence in the click location sequence, | P | for institute
State the total number of sequence in click location sequence;
The spin matrix determines in the following manner:
All the points in the click location sequence are fitted, to obtain the corresponding straight line of the click location sequence;
The spin matrix is determined by the straight line and horizontal angle.
In the present embodiment, the method further includes:
When the range residing for the sequence in the click location sequence is less than specified range threshold value or the straight line and water
When the angle of horizontal line is more than specified angle threshold value, rotation processing is not carried out to the previous frame dummy keyboard.
In the present embodiment, matching obtains the corresponding target vocabulary of the click location sequence and includes in dictionary:
Determine that the constant hard iron of the corresponding consecutive points distance of the click location sequence, consecutive points angle, point set and grouping are estimated
Value function, and the click is determined based on consecutive points distance, consecutive points angle, the constant hard iron of point set and grouping evaluation function
The corresponding comprehensive assessment value of position sequence;
The comprehensive assessment value and the assessed value of vocabulary each in dictionary are compared, and will be with the comprehensive assessment value
The corresponding vocabulary of assessed value of difference minimum is as the corresponding target vocabulary of the click location sequence.
In the present embodiment, the consecutive points distance determines according to the following equation:
Wherein, piRepresent i-th of sequence in the click location sequence, pi+1Represent i+1 in the click location sequence
A sequence, diRepresent piAnd pi+1The distance between, N represents the quantity of sequence in the click location sequence;
The consecutive points angle determines according to the following equation:
Wherein, pi-1Represent (i-1)-th sequence in the click location sequence, θ represents piAnd pi-1The straight line and p of formationiWith
pi+1Angle between the straight line of formation;
The constant hard iron of point set determines according to the following equation:
I3=(y30+3y)2+(3y21-y03)2
Wherein, I3Represent third class not bending moment,
Wherein, M represents the number of point, xi, yiI-th point of x coordinate and y-coordinate is respectively referred to, the x that x, y respectively refer to point set is sat
Mark average value and y-coordinate average value, f (xi,yi) refer to the weighting function of the point;
The grouping evaluation function determines in the following manner:
The grouping evaluation function is determined according to the following equation:
Wherein, f represents the grouping evaluation function, CiRepresent the number of button in the corresponding piecemeal keyboard of finger;
Or
The grouping evaluation function is determined according to the following equation:
Wherein, F represents the pressure of finger, and S represents the contact area of finger, and P represents the position of finger, and g (*) represents fitting
Function.
In the present embodiment, the new position and new posture for determining the dummy keyboard after Primary Location include:
Ask for the minimum value of following formula:
Wherein, (xi,yi) represent previous frame dummy keyboard in i-th of button coordinate value, θ represent rotation parameter, TxIt represents
The translation parameters of x-axis, TyRepresent the translation parameters of y-axis, S represents zooming parameter, and s (*) is represented in previous frame dummy keyboard i-th
The abscissa of a button converted after value, t (*) represents to carry out the ordinate of i-th button in previous frame dummy keyboard
Value after transformation, N represent the number of button.
In the present embodiment, the method further includes:
The specified gesture being applied in response to user on the touch screen, transparent window is shown on the touch screen;Its
In, show candidate word in the transparent window;
The click location sequence that user is currently applied on the touch screen is obtained, and based on the current click location
Sequence changes the content of the candidate word.
In the present embodiment, the content that the candidate word is changed based on the current click location sequence is included:
It horizontally slips if the current click location sequence characterization list refers to, other wait is shown in the transparent window
Select word;
If the current click location sequence characterization clicks current candidate word, the candidate word that will click on is filled in
In information input frame;
If the current click location sequence characterization list is directed toward lower slider, the candidate word of current presentation is deleted;
If the current click location sequence characterization list refers to upward sliding, it is defeated that current candidate word is filled in information
Enter in frame.
In practical application scene, there are two types of input modes for different crowd for self-positioning input method:
Mode one:It is familiar with the crowd of keyboard layout:Using default hidden keyboard, it is defeated to predict that user wants according to key position
The word entered.When needing with keyboard (for example inputting password), dummy keyboard is breathed out by breathing out gesture.
Mode two:It is unfamiliar with the crowd of keyboard layout:Keyboard and common input method input mode one are shown using acquiescence
Sample, What You See Is What You Get, difference are to input any position on the screen.
Assuming that the point sequence P={ p of click location1,p2,...,pn, the scheme of the application may comprise steps of.
1. establish dictionary
Signified dictionary of the invention is the dictionary of broad sense, including the phonetic library being made of phonetic, the word of Chinese word composition
Repertorie, the word library of English word composition, the phrase database of English phrase composition, the word library of other Languages word composition, number
The digital library of sequence composition, the maneuver library of action sequence composition etc., what every available sequences represented is collectively termed as " word ", by word
Composition becomes " dictionary ".
It, cannot be as common input method dependent in keyboard layout since self-positioning input method is independent of keyboard
Key position, and be to rely on dictionary match condition prediction user want input word.Dictionary is made of two parts, and one
Part is local dictionary, records the dictionary produced by user, and edited including user or input method inputs automatic raw according to user
Into word.Another part is high in the clouds dictionary, is the main body of dictionary, this has been unziped to from high in the clouds or installation kit before program startup
Ground.
2. Primary Location
If Q is the position of keyboard and the function of state of posture, Q0It is current state, Q ' is the state of previous frame, MRIt is rotation
Matrix, MTIt is translation matrix, MSIt is scaled matrix, it is clear that Q0It is that Q ' is obtained by rotation, Pan and Zoom, then has:
Q0=f (Q ', MR,MT,MS)
It needs to use Primary Location there are three types of situation:
1) when key position sequence input by user is fewer;
2) characters matching returns to undesirable result;
3) other abnormal conditions need default location;
Primary Location is in two kinds of situation:
When | P |=1
Keyboard is without the variation for rotating and scaling, i.e. Q0=f (Q ', E, MT, E), MTIt is then the keyboard centerline by Q ' to coordinateVector determine.
When | P | >=2
Keyboard does not scale variation, only rotation and translation at this time.
MTIt is then the keyboard centerline by Q ' to coordinateVector determine.
MRThen by P={ p1,p2,...,pnAll the points fitting (such as least square method) obtain straight line, by straight line with
Horizontal angle determines that it is more than threshold value more to concentrate or rotate such as point, can not also take rotation amount, this season MR=E.
3. characters matching
By P={ p1,p2,...,pnPoint composition cyclic sequence, then can obtain following characteristics:
The distance of consecutive points:Have after normalized,
The angle of consecutive points has after normalized,
The Hu of point set not bending moment (without loss of generality, by taking third moment as an example):I3=(y30+3y)2+(3y21-y03)2
It is grouped evaluation function:Refer to that user currently presses the probability of some letter.According to QWERTY keyboard button, each hand
Referring to the letter of control can be grouped, such as left hand little finger of toe control tri- keys of QAZ (probability that little finger of toe presses Q at this time is 1/3), third finger control
Tri- keys of WSX, middle finger control tri- keys of EDC, forefinger control six keys of RFVTGB;Right hand forefinger controls six keys of YHNUJM, middle finger control IK
Two keys, third finger control two keys of OL, little finger of toe control mono- keys of P.
Some equipment, if VR/AR equipment can learn the corresponding triggering finger of each coordinate, thenWherein CiRefer to
Be every group of alphabetical number
Some equipment are then needed according to the corresponding triggering of the characteristic estimatings coordinate such as finger pressure F, contact area S, position P
Finger, then evaluation function:
By above several characteristics, comprehensive score function can define:
S(d,θ,I,f)
Its score S is calculated to each word i in dictionaryi, then SiWith the score S of position sequence0Difference determine its correlation
Property, difference reckling is best match person, i.e.,:
Min Δs S=| Si-S0|
4. calculate keyboard position and posture
After a upper section obtains best match, as shown in Fig. 2, after user's sequence presses four positions (left figure), obtain most
The word of good adaptation is take (right figure).
If P '={ p1′,p′2,...,p′nIt is the coordinate that new keyboard corresponds to button, in order to ask the position of keyboard and posture,
It needs to acquire values below minimum
Since new keyboard is old keyboard through rotating, translating, scaling, if
Due to Sx=Sy=S, therefore compound matrice have totally four unknown parameters, i.e. θ, Tx,Ty, S, then (1-1) available parameter
It is expressed as:
Rational step-length is set, four parameters can obtain by iteration convergence, so as to obtain the position of keyboard and posture.
Referring to Fig. 3, the application also provides a kind of automatic patching system for inputting information, the system comprises:
Click location retrieval unit 100, for obtaining the click location sequence being applied on touch screen, and based on institute
It states the dummy keyboard described in click location sequence pair on touch screen and carries out Primary Location;
Target vocabulary matching unit 200 obtains the corresponding target word of the click location sequence for being matched in dictionary
It converges;
Position and attitude determination unit 300 for being based on the target vocabulary, determines the new of the dummy keyboard after Primary Location
Position and new posture.
Therefore the technical solution that the application provides, have following advantage compared with the prior art:
1) independent of keyboard size and posture;
2) according to the input of user, keyboard can adjust size and posture in real time;
3) good error correction effect can be had by deviateing correct key mapping any distance;
4) keyboard can not be shown.
Technical solution provided by the invention can be applicable to the equipment such as mobile phone, tablet computer, VR/AR, microcomputer, projection
On, when inputting word, do not require to show keyboard layout (can also show) in advance, user needs to be familiar with keyboard layout (if not
It is familiar then need to show keyboard in advance), during user inputs word, keyboard can be according to key position and matching dictionary
Situation carrys out adjust automatically layout.
The apparatus embodiments described above are merely exemplary, wherein the unit illustrated as separating component can
To be or may not be physically separate, the component shown as unit may or may not be physics list
Member, you can be located at a place or can also be distributed in multiple network element.It can be selected according to the actual needs
In some or all of module realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying creativeness
Labour in the case of, you can to understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on such understanding, on
Technical solution is stated substantially in other words to embody the part that the prior art contributes in the form of software product, it should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including several fingers
It enables and (can be personal computer, server or the network equipment etc.) so that computer equipment is used to perform each implementation
Method described in certain parts of example or embodiment.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention.
Claims (9)
1. a kind of automatic matching method for inputting information, which is characterized in that the method includes:
The click location sequence being applied on touch screen is obtained, and based on the void described in the click location sequence pair on touch screen
Intend keyboard and carry out Primary Location;
Matching obtains the corresponding target vocabulary of the click location sequence in dictionary;
Based on the target vocabulary, the new position of the dummy keyboard after Primary Location and new posture are determined;
Matching obtains the corresponding target vocabulary of the click location sequence and includes in dictionary:
Determine the constant hard iron of the corresponding consecutive points distance of the click location sequence, consecutive points angle, point set and grouping valuation letter
Number, and the click location is determined based on consecutive points distance, consecutive points angle, the constant hard iron of point set and grouping evaluation function
The corresponding comprehensive assessment value of sequence;
The comprehensive assessment value and the assessed value of vocabulary each in dictionary are compared, and will be different with the comprehensive assessment value difference
The minimum corresponding vocabulary of assessed value is as the corresponding target vocabulary of the click location sequence.
2. according to the method described in claim 1, it is characterized in that, Primary Location is carried out to the dummy keyboard on the touch screen
Including:
When the total number of the click location sequence is 1, the dummy keyboard of previous frame is carried out according to translation matrix at translation
Reason;Wherein, the translation matrix is determined by the vector of keyboard centerline to the specified coordinate of previous frame dummy keyboard;The specified seat
Mark determines according to the following equation:
Wherein, P be the click location sequence, PiFor the value of i-th of sequence in the click location sequence, | P | it is the click
The total number of sequence in position sequence.
3. according to the method described in claim 1, it is characterized in that, Primary Location is carried out to the dummy keyboard on the touch screen
Including:
When the total number of the click location sequence is more than or equal to 2, to the dummy keyboard of previous frame according to translation matrix
Carry out translation processing and rotation processing respectively with spin matrix;
Wherein, the translation matrix is determined by the vector of keyboard centerline to the specified coordinate of previous frame dummy keyboard;It is described specified
Coordinate determines according to the following equation:
Wherein, P be the click location sequence, PiFor the value of i-th of sequence in the click location sequence, | P | it is the click
The total number of sequence in position sequence;
The spin matrix determines in the following manner:
All the points in the click location sequence are fitted, to obtain the corresponding straight line of the click location sequence;
The spin matrix is determined by the straight line and horizontal angle.
4. according to the method described in claim 3, it is characterized in that, the method further includes:
When the range residing for the sequence in the click location sequence is less than specified range threshold value or the straight line and horizontal line
Angle when being more than specified angle threshold value, rotation processing is not carried out to the previous frame dummy keyboard.
5. according to the method described in claim 1, it is characterized in that, consecutive points distance determines according to the following equation:
Wherein, piRepresent i-th of sequence in the click location sequence, pi+1Represent i+1 sequence in the click location sequence
Row, diRepresent piAnd pi+1The distance between, N represents the quantity of sequence in the click location sequence;
The consecutive points angle determines according to the following equation:
Wherein, pi-1Represent (i-1)-th sequence in the click location sequence, θ represents piAnd pi-1The straight line and p of formationiAnd pi+1
Angle between the straight line of formation;
The constant hard iron of point set determines according to the following equation:
I3=(η30-3η12)2+(3η21-η03)2
Wherein, I3Represent third class not bending moment,
Wherein, M represents the number of point, xi, yiI-th point of x coordinate and y-coordinate is respectively referred to,Respectively refer to the x coordinate of point set
Average value and y-coordinate average value, f (xi,yi) refer to the weighting function of the point;
The grouping evaluation function determines in the following manner:
The grouping evaluation function is determined according to the following equation:
Wherein, f represents the grouping evaluation function, CiRepresent the number of button in the corresponding piecemeal keyboard of finger;
Or
The grouping evaluation function is determined according to the following equation:
Wherein, F represents the pressure of finger, and S represents the contact area of finger, and P represents the position of finger, and g (*) represents fitting letter
Number.
6. the according to the method described in claim 1, it is characterized in that, new position of dummy keyboard after determining Primary Location and new
Posture includes:
Ask for the minimum value of following formula:
Wherein, (xi,yi) represent previous frame dummy keyboard in i-th of button coordinate value, θ represent rotation parameter, TxRepresent x-axis
Translation parameters, TyRepresent the translation parameters of y-axis, S represents zooming parameter, and s (*) is represented to i-th in previous frame dummy keyboard
The abscissa of button converted after value, t (*) expression the ordinate of i-th of button in previous frame dummy keyboard is become
Value after changing, N represent the number of button.
7. according to the method described in claim 1, it is characterized in that, the method further includes:
The specified gesture being applied in response to user on the touch screen, transparent window is shown on the touch screen;Wherein, institute
It states and candidate word is shown in transparent window;
The click location sequence that user is currently applied on the touch screen is obtained, and based on the current click location sequence
Change the content of the candidate word.
8. the method according to the description of claim 7 is characterized in that the time is changed based on the current click location sequence
The content of word is selected to include:
It horizontally slips, is shown in the transparent window other candidate if the current click location sequence characterization list refers to
Word;
If the current click location sequence characterization clicks current candidate word, the candidate word that will click on is filled in information
In input frame;
If the current click location sequence characterization list is directed toward lower slider, the candidate word of current presentation is deleted;
If the current click location sequence characterization list refers to upward sliding, current candidate word is filled in information input frame
In.
9. a kind of automatic patching system for inputting information, which is characterized in that the system comprises:
Click location retrieval unit, for obtaining the click location sequence being applied on touch screen, and based on the click
Position sequence carries out Primary Location to the dummy keyboard on the touch screen;
Target vocabulary matching unit obtains the corresponding target vocabulary of the click location sequence for being matched in dictionary;
Position and attitude determination unit, for being based on the target vocabulary, determine the dummy keyboard after Primary Location new position and
New posture;
Matching obtains the corresponding target vocabulary of the click location sequence and includes in dictionary:
Determine the constant hard iron of the corresponding consecutive points distance of the click location sequence, consecutive points angle, point set and grouping valuation letter
Number, and the click location is determined based on consecutive points distance, consecutive points angle, the constant hard iron of point set and grouping evaluation function
The corresponding comprehensive assessment value of sequence;
The comprehensive assessment value and the assessed value of vocabulary each in dictionary are compared, and will be different with the comprehensive assessment value difference
The minimum corresponding vocabulary of assessed value is as the corresponding target vocabulary of the click location sequence.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710709668.7A CN107577357B (en) | 2017-08-18 | 2017-08-18 | A kind of automatic matching method and system for inputting information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710709668.7A CN107577357B (en) | 2017-08-18 | 2017-08-18 | A kind of automatic matching method and system for inputting information |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107577357A CN107577357A (en) | 2018-01-12 |
CN107577357B true CN107577357B (en) | 2018-07-06 |
Family
ID=61035521
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710709668.7A Active CN107577357B (en) | 2017-08-18 | 2017-08-18 | A kind of automatic matching method and system for inputting information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107577357B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108536037B (en) * | 2018-07-12 | 2023-08-11 | 四川虹美智能科技有限公司 | Control equipment and method of intelligent household appliance and intelligent household appliance |
CN111580663A (en) * | 2020-05-11 | 2020-08-25 | 清华大学 | Information input method, system, electronic equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1668994A (en) * | 2002-07-19 | 2005-09-14 | 索尼株式会社 | Information display input device and information display input method, and information processing device |
WO2012140883A1 (en) * | 2011-04-11 | 2012-10-18 | パナソニック株式会社 | Display processing device |
CN104156148A (en) * | 2014-07-18 | 2014-11-19 | 百度在线网络技术(北京)有限公司 | Method and device for providing virtual keyboards for mobile equipment |
-
2017
- 2017-08-18 CN CN201710709668.7A patent/CN107577357B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1668994A (en) * | 2002-07-19 | 2005-09-14 | 索尼株式会社 | Information display input device and information display input method, and information processing device |
WO2012140883A1 (en) * | 2011-04-11 | 2012-10-18 | パナソニック株式会社 | Display processing device |
CN104156148A (en) * | 2014-07-18 | 2014-11-19 | 百度在线网络技术(北京)有限公司 | Method and device for providing virtual keyboards for mobile equipment |
Also Published As
Publication number | Publication date |
---|---|
CN107577357A (en) | 2018-01-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10156981B2 (en) | User-centric soft keyboard predictive technologies | |
CN100437739C (en) | System and method for continuous stroke word-based text input | |
Dunlop et al. | Multidimensional pareto optimization of touchscreen keyboards for speed, familiarity and improved spell checking | |
RU2504820C2 (en) | Dynamic programmed keyboard | |
US20140198047A1 (en) | Reducing error rates for touch based keyboards | |
JP2022003539A (en) | Method, apparatus, electronic device and storage medium for correcting text errors | |
US10884613B2 (en) | Method and device for input with candidate split mode | |
CN105074643B (en) | The gesture keyboard of non-dictionary character string inputs | |
WO2017005207A1 (en) | Input method, input device, server and input system | |
CN110083254A (en) | More gesture text input predictions | |
US11409374B2 (en) | Method and device for input prediction | |
US9910589B2 (en) | Virtual keyboard with adaptive character recognition zones | |
CN107577357B (en) | A kind of automatic matching method and system for inputting information | |
JP7133002B2 (en) | Punctuation prediction method and apparatus | |
CN104461056B (en) | A kind of information processing method and electronic equipment | |
CN107797676B (en) | Single character input method and device | |
CN111026322A (en) | Keyboard layout method and device, electronic equipment and storage medium | |
CN111124200B (en) | Touch key response method and device, electronic equipment and storage medium | |
KR102128415B1 (en) | Method and system for correcting keyboard typo based on deep learning model | |
CN111090341A (en) | Input method candidate result display method, related equipment and readable storage medium | |
WO2022233275A1 (en) | Input correction method and apparatus | |
CN110738261A (en) | Image classification and model training method and device, electronic equipment and storage medium | |
CN111367459B (en) | Text input method using pressure touch pad and intelligent electronic device | |
CN114639158A (en) | Computer interaction method, apparatus and program product | |
KR20200078454A (en) | Method and system for correcting keyboard typo based on deep learning model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |