SYSTEM AND METHOD FOR IMPLEMENTING SLIDING INPUT OF TEXT BASED UPON ON-SCREEN SOFT KEYBOARD ON ELECTRONIC EQUIPMENT

• US 20160004326A1
• Filed: 09/17/2015
• Published: 01/07/2016
• Est. Priority Date: 04/09/2011
• Status: Active Grant

• Current Assignee: Shanghai Chule (CooTek) Information Technology Co., Ltd.
• Original Assignee: Shanghai Chule (CooTek) Information Technology Co., Ltd.
• Inventors: ZHANG, Qing, XIE, Haichao, HE, Qi
• Granted Patent: US 9,417,709 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 2203/04808 Several contacts: gestures triggering a specific function, e.g. scrolling, zooming, right-click, when the user establishes several contacts with the surface simultaneously; e.g. using several fingers or a combination of fingers and pen
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 3/04883 for inputting data by handwriting, e.g. gesture or text
• CPC Class Codes: G06F 3/04886 by partitioning the display area of the touch-screen or the surface of the digitising tablet into independently controllable areas, e.g. virtual keyboards or menus

Abstract

A system for implementing the sliding input of a text based upon an on-screen soft keyboard on an electronic equipment is provided. The system comprises a feature lexicon module, a feature lexicon establishment module, a user interface module, a sliding trajectory feature extraction module, a rough classifying and filtering module, a rough matching and filtering module and an accurate matching and sorting module. A method for implementing the sliding input using the system is also provided. The system and the method can input a word according to the preorder trajectory of the input portion, and cleverly use a forecasting-between-slidings input, decrease a sliding distance, effectively improve the input speed and accuracy of words, and simplify the operation for the users.

Claims

1. A system for implementing a sliding input of a text based upon an on-screen soft keyboard on an electronic equipment, characterized in that, said system comprises:

a memory device configured to store ideal sliding trajectory features for words; and

a processor coupled to the memory device, the processor being configured to;

record user-sliding trajectories and convert the recorded user-sliding trajectories into a user-sliding trajectory feature set to be processed;

filter in the memory device and originally choose the words, wherein each of the originally chosen words has similar ideal sliding trajectory features with the user-sliding trajectory feature set;

calculate a similarity between the ideal sliding trajectory features of each originally chosen word and said user-sliding trajectory features set according to key points on said trajectory, characterized in that, calculating said similarity including the following steps;

calculating a rough similarity between the ideal sliding trajectory features of each originally chosen word and said user-sliding trajectory features set;

calculating an accurate similarity between the ideal sliding trajectory features of each word obtained from the rough similarity calculation result and said user-sliding trajectory features set;

wherein calculating said accurate similarity including calculating an accurate linear matching distance between the ideal sliding trajectory feature of each word obtained from the rough similarity calculation result and said user-sliding trajectory feature set;

obtain candidate words according to the similarity, wherein the ideal sliding trajectory of each candidate word contains at least one of the key points or at least one of the surrounding points of at least one of the key points on said user-sliding trajectory;

display said candidate words.

2. The system according to claim 1, characterized in that, said processor is further configured to:

calculate a rough matching degree between words stored in said memory device according to the ideal sliding trajectory features and said user-sliding trajectory features set.

3. The system according to claim 2, characterized in that, said processor is further configured to calculate the rough match degree by the following steps:

obtaining a corresponding character string set according to the key point sequence on said user-sliding trajectory;

inquiring in the memory device according to said character string set;

fetching out probabilities of all the inquired-out strings and obtain words corresponding to individual probability;

calculating a rough matching degree between those words and said user-sliding trajectory feature set.

4. The system according to claim 2, characterized in that, said processor is further configured to apply an incremental calculation method or a prefix matching method to calculate the rough match degree.

5. The system according to claim 1, characterized in that, said ideal sliding trajectory features for words are stored in a way of inverted index, wherein said inverted index includes a mapping from a key point to a word node, and said word node includes a word identification and the probability that said key point appears in the ideal sliding trajectory of said word.

6. The system according to claim 5, characterized in that, said inverted index includes Bi-grams inverted index.

7. The system according to claim 1, characterized in that,

said sliding trajectory features include sliding trajectory lengths of words and an word key point sequence, wherein said ideal word key point sequence is an ordered combination of key points on the sliding trajectory of the word.

8. The system according to claim 1, characterized in that, said processor is further configured to calculate a rough similarity between the ideal sliding trajectory features of each originally chosen word and said user-sliding trajectory features set by the following steps:

calculating a linear matching distance between the ideal trajectory of each originally chosen word and said user-sliding trajectory feature set.

9. The system according to claim 1, characterized in that, said processor is further configured to calculate a rough similarity between the ideal sliding trajectory features of each originally chosen word and said user-sliding trajectory features set by the following steps:

performing trajectory segmentation on the ideal trajectories of the originally chosen words according to said user-sliding trajectory feature set.

10. The system according to claim 1, characterized in that, said processor is further configured to obtain candidate words according to the similarity by the following steps:

sorting the words obtained by the accurate similarity calculation according to at least one of the accurate similarity, word frequency, user input history information and the system-configured number of candidate words.

11. A method for implementing sliding input of a text based upon an on-screen soft keyboard on an electronic equipment having a memory device and a processor, characterized in that, said method comprises the following steps:

recording user-sliding trajectories;

converting the recorded user-sliding trajectories into a user-sliding trajectory feature set to be processed;

filtering in the memory device and originally choosing the words, wherein each of the originally chosen words has similar ideal sliding trajectory features with the user-sliding trajectory feature set;

calculating a similarity between the ideal sliding trajectory features of each originally chosen word and said user-sliding trajectory features set according to key points on said trajectory, characterized in that, calculating said similarity including the following steps;

calculating a rough similarity between the ideal sliding trajectory features of each originally chosen word and said user-sliding trajectory features set;

calculating an accurate similarity between the ideal sliding trajectory features of each word obtained from the rough similarity calculation result and said user-sliding trajectory features set;

wherein calculating said accurate similarity including calculating an accurate linear matching distance between the ideal sliding trajectory feature of each word obtained from the rough similarity calculation result and said user-sliding trajectory feature set;

obtaining candidate words according to the similarity, wherein the ideal sliding trajectory of each candidate word contains at least one of the key points or at least one of the surrounding points of at least one of the key points on said user-sliding trajectory;

feedbacking the candidate words.

12. The method according to claim 11, characterized in that, said sliding trajectory features include trajectory lengths of words and an word key point sequence, wherein said word key point sequence is an ordered combination of key points on the sliding trajectory of the word.

13. The method according to claim 12, characterized in that, said key points include starting and ending key points as well as ordinary key points of the trajectory, wherein said ordinary key points are inflection points on the trajectory.

14. The method according to claim 11, characterized in that, said ideal sliding trajectory features for words are stored in a way of inverted index.

15. The method according to claim 11, characterized in that, said ideal sliding trajectory features for words are stored in a way of inverted index, wherein said inverted index is a Bi-grams inverted index.

16. The method according to claim 11, characterized in that, said ideal sliding trajectory features for words are stored in a way of inverted index, wherein said inverted index includes a mapping from a key point to a word node, and said word node includes word identification and the probability that said key point appears in the ideal sliding trajectory of said word.

17. The method according to claim 11, characterized in that, said step of converting the user-sliding trajectories into a user-sliding trajectory feature set to be processed includes substeps of:

performing a trajectory preprocessing operation on the user-sliding trajectory by the system;

performing a key point extraction processing operation on the preprocessed user-sliding trajectory by the system.

18. The method according to claim 11, characterized in that, said step of filtering in the memory device and originally choosing the words includes a substep of calculating a rough matching degree between words stored in said memory device according to the ideal sliding trajectory features and said user-sliding trajectory features set.

19. The method according to claim 11, characterized in that, said step of filtering in the memory device and originally choosing the words utilizes an incremental calculation method or a prefix matching method.

20. The method according to claim 11, characterized in that, said step of filtering in the memory device and originally choosing the words includes following substeps:

obtaining a corresponding character string set according to the key point sequence on said user-sliding trajectory;

inquiring in the memory device according to said character string set;

fetching out probabilities of all the inquired-out strings and obtain words corresponding to individual probability;

calculating a rough matching degree between those words and said user-sliding trajectory feature set.

21. The method according to claim 11, characterized in that, said calculating a rough similarity between the ideal sliding trajectory feature of each originally chosen word and said user-sliding trajectory feature set, includes a substep of calculating a linear matching distance between the ideal trajectory of each originally chosen word and said user-sliding trajectory feature set.

22. The method according to claim 11, characterized in that, said step of calculating a rough similarity between the ideal sliding trajectory feature of each originally chosen word and

said user-sliding trajectory feature set further includes a substep of performing trajectory segmentation on the ideal trajectories of the original chosen words according to said user-sliding trajectory feature set.

23. The method according to claim 11, wherein said step of obtaining candidate words includes a substep of sorting the words obtained by the accurate similarity calculation in the processor according to at least one of the accurate similarity, word frequency, user input history information and the system-configured number of candidate words.