CN102426483B - Multi-channel accurate target positioning method for touch equipment - Google Patents
Multi-channel accurate target positioning method for touch equipment Download PDFInfo
- Publication number
- CN102426483B CN102426483B CN 201110445164 CN201110445164A CN102426483B CN 102426483 B CN102426483 B CN 102426483B CN 201110445164 CN201110445164 CN 201110445164 CN 201110445164 A CN201110445164 A CN 201110445164A CN 102426483 B CN102426483 B CN 102426483B
- Authority
- CN
- China
- Prior art keywords
- touch
- phonetic entry
- control
- confidence level
- target
- 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
Images
Landscapes
- User Interface Of Digital Computer (AREA)
- Position Input By Displaying (AREA)
Abstract
The invention belongs to the field of human-machine interaction and particularly relates to a multi-channel accurate target positioning method for touch equipment. According to the multi-channel accurate target positioning method, touch input and voice input are used for positioning the target, wherein the touch input is used for providing a possible target object candidate set; the voice input is also used for providing a possible target object candidate set; and the possible target object candidate sets are positioned to an accurate target object by a multi-channel fusion algorithm. According to the multi-channel accurate target positioning method disclosed by the invention, two natural input channels, such as the touch input and the voice input as well as two natural interaction manners are supported; more accurate target positioning can be carried out; and the efficiency of positioning relevant tasks on the touch equipment is increased.
Description
Affiliated technical field
The invention belongs to field of human-computer interaction, be specifically related to a kind of precision target of the hyperchannel towards touch control device localization method.
Background technology
On touch control device, some tasks often need the user first to navigate to after destination object to carry out subsequent operation.For example, in text editing operation, often need first cursor to be navigated to target location, and then inserted, the operation such as deletion; In the navigation process of map, need to first navigate to after target location is amplified and check; To the operation of single object, need to first select destination object in a plurality of small objects.Yet, due to the contact area large (list of references: Fat finger worries:How older and younger users physically interact with PDAs.In Proc.INTERACT 2005 of finger with touch screen, 267-280.), often can not accurately select or navigate to destination object.Therefore, for the user provides effectively, accurate Technology for Target Location becomes particularly important.
High-precision touch screen interaction technology is widely studied.(list of references: Shift:A technique for operating pen-based interfaces using touch.In Proc.CHI 2007 657-666.) utilizes the cursor migration technology to solve the larger problem of the mutual middle finger input of touch-control for Vogel and Baudisch.(the list of references: Rubbing and tapping for precise and rapid selection on touch-screen displays.In Proc.CHI 2008 such as Olwal, 295-304.), (the list of references: TapTap and MagStick:Improving one-handed target acquisition on small touch-screens.In Proc.AVI 2008 such as Roudaut, 146-153.), (the list of references: TouchCuts and TouchZoom:Enhanced target selection for touch displays using finger proximity sensing.In Proc.CHI 2011 such as Yang, 2585-2594.) and
deng (list of references: FingerGlass:Efficient multiscale interaction on multitouch screens.In Proc.CHI 2011 1343-1352.) utilizes the scaling technology to support accurate mutual.(list of references: Escape:A target selection technique using visually-cued gestures.In Proc.CHI 2008, the speed while 285-294.) adopting the gesture with visual cues to carry out the accelerating selection destination object such as Yatani.(the list of references: Precise selection techniques for multi-touch screens.In Proc.CHI 2006 such as Benko, 1263-1272.) and Olwal etc. (list of references: Rubbing and tapping for precise and rapid selection on touch-screen displays.In Proc.CHI 2008 295-304.) makes with the hands to refer to select to help accurate location.Albinsson and Zhai (list of references: High precision touch screen interaction.In Proc.CHI2003,105-112.) have designed some interactive tools and have supported accurate mutual.(list of references: Barrier pointing:using physical edges to assist target acquisition on mobile device touch screens.In Proc.ASSETS 2007,19-26.) utilize the physics frame of mobile device to carry out obtaining of auxiliary mark to Froehlich etc.Although more accurate mutual on these technical support touch control devices, after once touching clicking operation, they often need extra adjusting operation.
A plurality of input channels can provide complementary and information redundancy, can be used to design new interaction technique.For example, Hinckley and Song (list of references: Sensor Synaesthesia:Touch in Motion, and Motion in Touch.In Proc.CHI 2011,801-810.) in connection with the advantage of multiple point touching and motion perception, support that abundanter touch is mutual.Wigdor and Balakrishman (list of references: TiltText:using tilt for text input to mobile phones.In Proc.UIST 2003,81-90.) designed TiltText, support the user in button the inclination mobile phone with the input target character.(the list of references: Multimodal Chinese text entry with speech and keypad on mobile devices.In Proc.IUI 2008 such as Jiang, 341-344.) in conjunction with voice and keyboard input, support Chinese text input more efficiently on mobile device.Aspect is very useful providing abundanter interaction technique and raise the efficiency etc. for these methods, but they are not to design for the precision target orientation problem, can not be by the simple solution that must be applied to this problem.
Summary of the invention
The object of the invention is to utilize the hyperchannel input, the object localization method on a kind of touch control device is provided, thereby improve the accuracy of touch-control input target localization.The method has merged touch-control input and phonetic entry: non-accurate touch-control input provides a series of possible positions of destination object, and phonetic entry is for determining the exact position of target; By being carried out to Multichannel fusion, touch-control input and phonetic entry navigate to final destination object, as shown in Figure 1.
Specifically, the technical solution used in the present invention is as follows:
A kind of precision target of the hyperchannel towards touch control device localization method, its step comprises:
1) gather touch-control input and the phonetic entry on touch control device, obtain touch-control and input corresponding target candidate set and target candidate set corresponding to phonetic entry;
2) calculate the confidence level of described target candidate by the Multichannel fusion algorithm, by confidence level, the location positioning of the highest target candidate is final position location.
Further, described touch control device need to both be supported the touch-control input, also support voice input.Can be touch-control mobile phone, as iPhone, Google Phone etc., can be all kinds of touch-screen tablet computers, as iPad, happy Pad etc., can be also touch-control desktop etc.Object localization method of the present invention is not limited to the form for location, as cursor, little button etc.
Further, the position of destination object is pointed in described touch-control input, and the interface that the collection of touch-control input can provide based on existing touch control operation system (as Android, iOS, Windows Phone etc.) obtains.The information obtained comprises the position of the current touch point that system-computed obtains, the area of touch area etc.The finger touch overlay area has comprised possible target location, and they form the destination object candidate collection.The size of the distance of the touch point that the confidence level of each target candidate and this target location and system detect is inversely proportional to, and apart from touch point, the confidence level of nearer target candidate is larger.Meeting on the basis of this principle, the specific definition of candidate collection and confidence level can be defined by the developer.For example, when cursor is located, the cursor position that can be detected by system obtains peripheral cursor position, forms the cursor position candidate collection, and the confidence level of cursor position is along with the change of the cursor position that system of distance obtains reduces greatly.
Further, the content of described phonetic entry is the voice that destination object is relevant.For example, phonetic entry can be title or other association attributes of destination object.With touch-control on input time the most contiguous phonetic entry be that current touch-control is inputted corresponding phonetic entry.Phonetic entry can obtain the candidate of some recognition results and their confidence level by speech recognition engine identification.Speech recognition engine can be the speech recognition engine (as the Speech API of Google) that existing touch control operation system provides, and can be also third-party speech recognition engine (as iSpeech etc.).
Further, the computing formula of described Multichannel fusion algorithm is:
p(c
i)＝a*p(c
i|T)+(1-a)*p(c
i|S) (1)
Wherein, T is that touch-control is inputted corresponding target candidate set; S is target candidate set corresponding to phonetic entry; c
ii candidate in T; p(c
i| T) be target location c under the touch-control input condition
iconfidence level; p(c
i| S) be target location c in the phonetic entry situation
iconfidence level; p(c
i) be c
iconfidence level as the final goal object; A and (1-a) be touch-control input and phonetic entry shared proportion when the decision destination object, can be adjusted according to dissimilar user.The position of the target candidate that the confidence level calculated according to formula (1) is the highest is final position location, as shown in Figure 2.
Compared with prior art, the present invention has advantage and good effect are as follows:
1) the more accurate target localization of support of the present invention, thus the efficiency of target localization on touch control device can be improved.
2) the present invention adopts the input channel of touch-control and two kinds of natures of voice, thereby can support the mutual of nature.
3) the present invention adopts phonetic entry and touch-control input, and due to most of touch control device support voice input, this technology can be adopted by touch control device easily.
The accompanying drawing explanation
Fig. 1 is the schematic diagram of the present invention towards the hyperchannel precision target localization method of touch control device.
The schematic diagram of the Multichannel fusion algorithm that Fig. 2 is touch-control input of the present invention and phonetic entry.
The schematic diagram of cursor positioning method in the text editing that Fig. 3 is the embodiment of the present invention.
The schematic diagram that Fig. 4 is the example (navigating between the character ' h ' and ' s ' in " touchscreen ") that adopts cursor to position in the embodiment of the present invention, wherein: (a) touch target location say the first character ' s ' of back, target location with voice; (b) the larger zone that finger covers; (c) possible cursor position; (d) determine accurate cursor position by phonetic entry.
Embodiment
For purpose of the present invention, feature and advantage can be become apparent, hereinafter by specific embodiment, and by reference to the accompanying drawings, be described in detail.
Current very common at present alternately based on touch-control on mobile device, on these equipment, the user usually uses the application such as note, mail, notepad.In these application, cursor usually is used to the particular location of specify text input and editor's tasks carrying.Because the less finger mutual for touch-control of the text on mobile device is relatively large, the touch area of finger may cover a plurality of characters of different row and columns, and this makes the efficiency that cursor is positioned to target location become low and be easy to make mistakes.In addition, the often less and mutual vicinity of the character on touch-screen, be positioned cursor between the narrow character (as " ij ") of two vicinities to become particularly difficulty.The inventive method provides a kind of solution for this problem.
The present embodiment supports that the user carries out more effectively on touch-control mobile phone, accurate cursor positioning action; Be embodied on GoogleNexus S mobile phone, its screen size is 4.0 cun, and resolution is 480 * 800, and operating system is Android 2.3.
Fig. 3 provides the schematic diagram of this embodiment.When positioning cursor, the user says the character of target light cursor position back in the target light cursor position of touch screen with voice.In this embodiment, the first character that the content setting of phonetic entry is target light cursor position back, but the present invention is not limited to this, other is set (as the character of target light cursor position front etc.) and has similar effect.The Android system acquisition obtains the cursor position of finger touch screen, obtains the character of (upper and lower, left and right) around this cursor, forms the candidate collection of cursor position; Speech recognition engine based on Google Speech API obtains the character candidates collection of phonetic entry.By merging candidate collection and the voice Candidate Set (in the computing formula of p (ci), the value of a is set as 0.5) of touch-control input, obtain the confidence level of each cursor position in finger touch cursor candidate collection.The cursor position that confidence level is the highest is final cursor position.
Fig. 4 provides the example in above-described embodiment, cursor positioned.When the user attempts positioning cursor between ' h ' and ' s ' in " touchscreen " time, the user is by this cursor position of finger click, and say ' s ' with voice (Fig. 4 a) simultaneously.User's finger has covered a plurality of characters (Fig. 4 b and Fig. 4 c).Now, the cursor position that the touch-control input that the Android system is pointed according to the user obtains is arranged between ' c ' and ' h ' of " touchscreen ", not identical with user's intention.By utilizing phonetic entry and carrying out Multichannel fusion, cursor correctly can be navigated to (Fig. 4 d) between ' h ' and ' s '.
For the example in Fig. 4, below further introduce its Multichannel fusion algorithm.The touch-control input has generated 12 candidates' cursor position, and cursor can be positioned at ' h ', ' c ', ' s ', ' u ', ' c ', ' ', ' d ', ' l ', ' e ', ' e ', ' e ', ' e ' before, and they sort from high to low according to confidence level.The change of the distance of the cursor position that this confidence level obtains along with candidate's cursor position and system is large and reduce.Identify candidate ' s ', ' f ' and ' a ' that obtains voice identification result by speech recognition engine, and arrange from high to low by confidence level.A kind of mode of the confidence level of phonetic entry is directly to be provided by speech recognition engine; Another mode is the order definition confidence level of the identification candidate result returned according to speech recognition engine by the technological development personnel, and need to meet following principle: in candidate result sequences, more forward identification candidate's confidence level is higher.Then pass through blending algorithm, adopt formula (1), the cursor position that can calculate the touch-control input of sorting from high to low according to confidence level is: ' s ', ' h ', ' c ', ' u ', ' c ', ' ', ' d ', ' l ', ' e ', ' e ', ' e ', ' e '.Therefore, having the highest believable cursor position is that character ' s ' is front, and it will be as final cursor position, i.e. the cursor position of user's expectation.
In the above-described embodiments, during application of formula (1), a and (1-a) be touch-control input and phonetic entry shared proportion when the decision destination object, can be adjusted according to dissimilar user.When user's phonetic entry is more lack of standardization or environmental noise when larger, the speech recognition accuracy reduces, and can set largerly by the value of a, and the dependence that localization method is inputted touch-control like this is just larger; Otherwise, when phonetic entry standard and environment peace and quiet, can heighten the effect of phonetic entry in location, what soon the value of a was set is smaller.In the present embodiment, touch-control input and phonetic entry are considered on an equal basis, a is made as to 0.5.
Above method of the precision target of the hyperchannel towards touch control device of the present invention being located by embodiment has been described in detail, but specific implementation form of the present invention is not limited to this.One of ordinary skill in the art, can carry out various apparent variations and modification to it in the situation that do not deviate from spirit and the principle of the method for the invention.Protection scope of the present invention should be as the criterion so that claims are described.
Claims (10)
1. the precision target of the hyperchannel towards a touch control device localization method, its step comprises:
1) gather touch-control input and the phonetic entry on touch control device, obtain described touch-control and input corresponding target candidate set and target candidate set corresponding to described phonetic entry;
2) calculate the confidence level of described target candidate by the Multichannel fusion algorithm, by confidence level, the location positioning of the highest target candidate is final position location; The computing formula of described Multichannel fusion algorithm is:
Wherein, T is that touch-control is inputted corresponding target candidate set, S is target candidate set corresponding to phonetic entry, ci is i candidate in T, p (ci|T) is the confidence level of target location ci under the touch-control input condition, p (ci|S) is the confidence level of target location ci in the phonetic entry situation, and p (ci) is the confidence level of ci as the final goal object, a and (1-a) be touch-control input and phonetic entry shared proportion when the decision destination object.
2. the method for claim 1, is characterized in that, described touch control device is touch-control mobile phone, touch-screen tablet computer or the touch-control desktop of supporting touch-control input and phonetic entry.
3. method as claimed in claim 2, is characterized in that, described touch control device adopts cursor or little button to carry out described location.
4. method as claimed in claim 2, is characterized in that, the interface that described touch-control input provides based on the touch control operation system obtains, and described touch control operation system comprises Android, iOS and Windows Phone.
5. method as claimed in claim 4, is characterized in that, the information obtained by described touch-control input comprises: the position of touch point, the area of touch area.
6. the method for claim 1, is characterized in that, the confidence level that described touch-control is inputted each target candidate in corresponding target candidate set is inversely proportional to the size of the distance of the touch point detected from this target location to system.
7. the method for claim 1, is characterized in that, the content of described phonetic entry is relevant to destination object.
8. method as claimed in claim 7, is characterized in that, the title that the content of described phonetic entry is destination object.
9. the method for claim 1, is characterized in that, by speech recognition engine, identifies described phonetic entry and obtain the confidence level of described phonetic entry.
10. method as claimed in claim 9, is characterized in that, described speech recognition engine is the speech recognition engine that existing touch control operation system provides, and comprises the Speech API of Google; Or third-party speech recognition engine, comprise iSpeech.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110445164 CN102426483B (en) | 2011-12-27 | 2011-12-27 | Multi-channel accurate target positioning method for touch equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110445164 CN102426483B (en) | 2011-12-27 | 2011-12-27 | Multi-channel accurate target positioning method for touch equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102426483A CN102426483A (en) | 2012-04-25 |
| CN102426483B true CN102426483B (en) | 2013-12-25 |
Family
ID=45960479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110445164 Active CN102426483B (en) | 2011-12-27 | 2011-12-27 | Multi-channel accurate target positioning method for touch equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102426483B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9250730B2 (en) * | 2014-03-18 | 2016-02-02 | City University Of Hong Kong | Target acquisition system for use in touch screen graphical interface |
| CN104199594B (en) * | 2014-09-28 | 2018-02-23 | 厦门幻世网络科技有限公司 | A kind of target location localization method and its device based on touch-screen |
| CN106990843B (en) * | 2017-04-01 | 2021-01-08 | 维沃移动通信有限公司 | Parameter calibration method of eye tracking system and electronic equipment |
| CN108829248B (en) * | 2018-06-01 | 2020-11-20 | 中国科学院软件研究所 | Moving target selection method and system based on user performance model correction |
| CN109121081B (en) * | 2018-09-11 | 2020-12-29 | 电子科技大学 | Indoor positioning method based on position candidate set and EM algorithm |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1203390C (en) * | 2002-08-02 | 2005-05-25 | 北京南山高科技有限公司 | Method and device for inputting Chinese characters by speech and equipment with simplified keyboard |
| CN1300662C (en) * | 2003-12-29 | 2007-02-14 | 中国科学院自动化研究所 | Audio control device as computer peripheral |
| US20090182562A1 (en) * | 2008-01-14 | 2009-07-16 | Garmin Ltd. | Dynamic user interface for automated speech recognition |
-
2011
- 2011-12-27 CN CN 201110445164 patent/CN102426483B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN102426483A (en) | 2012-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9665276B2 (en) | Character deletion during keyboard gesture | |
| RU2635285C1 (en) | Method and device for movement control on touch screen | |
| US8237665B2 (en) | Interpreting ambiguous inputs on a touch-screen | |
| US11340755B2 (en) | Moving a position of interest on a display | |
| US20120188164A1 (en) | Gesture processing | |
| EP2551759A2 (en) | Gesture recognition method and touch system incorporating the same | |
| US20130215018A1 (en) | Touch position locating method, text selecting method, device, and electronic equipment | |
| US20150242114A1 (en) | Electronic device, method and computer program product | |
| CN103914196B (en) | Electronic equipment and the method for determining the validity that the touch key-press of electronic equipment inputs | |
| CN102426483B (en) | Multi-channel accurate target positioning method for touch equipment | |
| CN102272699A (en) | Gesture detection zones | |
| US10416868B2 (en) | Method and system for character insertion in a character string | |
| US9778780B2 (en) | Method for providing user interface using multi-point touch and apparatus for same | |
| CN104199607A (en) | Candidate selection method and device based on input method | |
| US20160147436A1 (en) | Electronic apparatus and method | |
| KR20100028465A (en) | The letter or menu input method which follows in drag direction of the pointer | |
| CN105335089A (en) | Page switch method and device based on intelligent terminal | |
| CN106227449A (en) | Input control method based on sense of touch vision technique and system | |
| US20120050184A1 (en) | Method of controlling driving of touch panel | |
| US20200142582A1 (en) | Disambiguating gesture input types using multiple heatmaps | |
| US20150355769A1 (en) | Method for providing user interface using one-point touch and apparatus for same | |
| US20120050171A1 (en) | Single touch process to achieve dual touch user interface | |
| CN101794182B (en) | Method and equipment for touch input | |
| US20140035876A1 (en) | Command of a Computing Device | |
| KR20110058047A (en) | Method for controlling screen using pointing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |