Input pattern detection over virtual keyboard for candidate word identification

• US 10,474,355 B2
• Filed: 01/20/2014
• Issued: 11/12/2019
• Est. Priority Date: 01/21/2013
• Status: Active Grant

• Current Assignee: KeyPoint Technologies India Pvt., Ltd.
• Original Assignee: KeyPoint Technologies India Pvt., Ltd.
• Inventors: Gummadidala, Shailaja, Kurian, Prima Dona, Yelubolu, Sandeep, Goswami, Sumit, Motaparti, Sunil
• Primary Examiner(s): Vu, Kieu D
• Assistant Examiner(s): Borja, Roberto
• Application Number:
  US14/762,355
• Publication Number:
  US 20150355836A1
• Time in Patent Office: 2,122 Days
• CPC Class Codes: G06F 3/023 Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
• 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
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs

Abstract

The present invention relates to a computer-implemented method for inputting text into an electronic device (10). A virtual keyboard (30) having a plurality of keys (40) is displayed on a display screen (20) and one or more characters are associated with each key. An input pattern is generated by a user and the key(s) (40) positioned on or proximal to the input pattern are identified. The keys (40) in the virtual keyboard (30) are allocated to pre-defined groups (50A, 50B, 50C), such that each group comprises a plurality of said keys (40). The input pattern is encoded in dependence on the pre-defined groups in which each of said one or more identified keys (40) are arranged. A candidate word or words is identified to be input into the electronic device (10) based on the encoded input pattern. The present invention also relates to computer-implemented methods for identifying candidate words based on the correlation between input and candidate string lengths; and/or upper/lower input length thresholds; and/or the identification of valid/invalid string combinations. The present invention also relates to a corresponding system, such as a computational apparatus (10).

Claims

1. A computer-implemented method for inputting text into an electronic device, the method comprising:

displaying a virtual keyboard comprising a plurality of keys arranged in a plurality of pre-defined non-overlapping groups on a display screen, each pre-defined non-overlapping group allocated an associated one of a plurality of unique identifier codes, one or more characters being associated with each of said keys;

detecting an input pattern generated by a user on the display screen over the plurality of pre-defined non-overlapping groups of the virtual keyboard;

encoding the input pattern by generating a sequence of the plurality of unique identifier codes corresponding to the input pattern over the plurality of pre-defined non-overlapping groups; and

identifying one or more candidates to be input into the electronic device based on the encoded input pattern; and

for a candidate of the one or more candidates;

identifying a plurality of parts of the candidate that have associated path lengths that fall within an upper path length threshold and a lower path length threshold, wherein the plurality of parts form a sequence, and wherein each next part in the sequence has a same set of characters as a prior part in the sequence and an additional character of the candidate;

determining a plurality of string distances, each string distance being between a respective one of the plurality of parts and the input pattern;

identifying a part from among the plurality of parts that has a best string distance from among the plurality of string distances; and

adding the candidate to a list for output by the electronic device based on a comparison between the best string distance and a string distance threshold.

2. A computer-implemented method as claimed in claim 1, wherein the one or more candidates are identified from one or more sets of words stored in a database, the words in each set comprising a pre-defined string combination;

wherein the method further comprises identifying one or more of said sets based on a string combination corresponding to at least a portion of the input pattern.

3. A computer-implemented method as claimed in claim 2, wherein said portion of the input pattern is an initial portion of the input pattern and the string combination corresponds to a character string at the beginning of the candidate.

4. A computer-implemented method as claimed in claim 1, wherein an input sequence code is stored for each candidate of the one or more candidates, the input sequence code being derived from at least one of the plurality of unique identifier codes allocated to the plurality of pre-defined non-overlapping groups.

5. A computer-implemented method as claimed in claim 1, comprising:

determining a length of the input pattern and calculating the upper path length threshold and the lower path length threshold based on said determined length;

wherein identifying said one or more candidates is based on the upper path length threshold and/or the lower path length threshold.

6. A computer-implemented method as claimed in claim 1, comprising applying a pattern matching algorithm to determine the best string distance.

7. A computer-implemented method as claimed in claim 1, wherein displaying the virtual keyboard, detecting the input pattern, encoding the input pattern, and identifying the one or more candidates are performed in a first phase of the method, the method further comprising:

in a second phase forming one or more character strings consisting of characters associated with each key positioned on or proximal to said input pattern and filtering the one or more character strings to identify one or more candidates for input into the electronic device, wherein filtering the character string comprises identifying valid and/or invalid string combinations within each character string in the second phase; and

in a third phase the keys displayed in the virtual keyboard are arranged in pre-defined groups and one or more characters is associated with each key, a pattern matching algorithm being applied to determine a correlation between said input pattern and a modelled input pattern for a candidate, and identifying one or more of said candidates to be input into the electronic device based on said correlation;

wherein the first, second and third phases are performed in any order.

8. A computer-implemented method as claimed in claim 1, wherein the plurality of pre-defined non-overlapping groups comprises a plurality of rows of the virtual keyboard.

9. A computer-implemented method as claimed in claim 1, wherein the one or more candidates each comprise a word, a phrase, or a sentence.

10. A computer-implemented method for inputting text into an electronic device, the method comprising:

displaying a virtual keyboard comprising a plurality of keys arranged in a plurality of pre-defined non-overlapping groups on a display screen, each pre-defined non-overlapping group allocated an associated one of a plurality of unique identifier codes, one or more characters being associated with each of said keys;

detecting an input pattern generated by a user on the display screen over the plurality of pre-defined non-overlapping groups of the virtual keyboard;

generating an initial list of candidates based on a sequence of the plurality unique identifier codes corresponding to the input pattern over the plurality of pre-defined non-overlapping groups;

generating a character string corresponding to the input pattern; and

filtering the character string to identify valid and/or invalid combinations of characters within the character string; and

generating a filtered list of candidates by;

matching the initial list of candidates to the valid combinations of characters; and

for a candidate in the initial list;

identifying a plurality of parts of the candidate that have associated path lengths that fall within an upper path length threshold and a lower path length threshold, wherein the plurality of parts form a sequence, and wherein each next part in the sequence has a same set of characters as a prior part in the sequence and an additional character of the candidate;

determining a plurality of string distances, each string distance being between a respective one of the plurality of parts and the input pattern;

identifying a part from among the plurality of parts that has a best string distance from among the plurality of string distances; and

adding the candidate to the filtered list based on a comparison between the best string distance and a string distance threshold.

11. A computer-implemented method as claimed in claim 10, wherein filtering the character string comprises determining if a particular character in the character string in combination with each subsequent character in the character string forms a valid combination.

12. A computer-implemented method as claimed in claim 11, wherein filtering the character string comprises initiating the filtering by checking for a valid combination with a first character in the character string which corresponds to a start point of the input pattern.

13. A computer-implemented method as claimed in claim 12, wherein filtering the character string further comprises checking each character in the character string in combination with each subsequent character in the character string for one or more valid combinations of characters.

14. A computer-implemented method as claimed in claim 10, wherein the plurality of pre-defined non-overlapping groups comprises a plurality of rows of the virtual keyboard.

15. A computer-implemented method as claimed in claim 10, wherein the candidates in the initial list each comprise a word, a phrase, or a sentence.

16. A computer-implemented method for inputting text into an electronic device, the method comprising:

displaying a virtual keyboard comprising a plurality of keys arranged in a plurality of pre-defined non-overlapping groups on a touchscreen display, each pre-defined non-overlapping group allocated an associated one of a plurality of unique identifier codes one or more characters being associated with each key;

receiving a user input that contacts the touchscreen display and moves over the plurality of pre-defined non-overlapping groups of the virtual keyboard to form a continuous input pattern over the plurality of pre-defined non-overlapping groups;

collecting X,Y coordinates of a plurality of contact points in response to the movement of the user input over the plurality of pre-defined non-overlapping groups;

detecting the continuous input pattern and registering a respective one of the plurality of unique identifier codes to a sequence when the input pattern traces over a key in a group of the plurality of pre-defined non-overlapping groups associated with the respective one of the plurality of unique identifiers based on collecting the X,Y coordinates;

identifying one or more candidates to be input into the electronic device based on the continuous input pattern and the sequence; and

for a candidate of the one or more candidates;

identifying a plurality of parts of the candidate that have associated path lengths that fall within an upper path length threshold and a lower path length threshold, wherein the plurality of parts form a sequence, and wherein each next part in the sequence has a same set of characters as a prior part in the sequence and an additional character of the candidate;

determining a plurality of string distances, each string distance being between a respective one of the plurality of parts and the input pattern;

identifying a part from among the plurality of parts that has a best string distance from among the plurality of string distances; and

adding the candidate to a list for output by the electronic device based on a comparison between the best string distance and a string distance threshold.

17. A computer-implemented method as claimed in claim 16, comprising identifying a first set of corner points in said continuous input pattern;

and

identifying a second set of corner points in a modelled input pattern.

18. A computer-implemented method as claimed in claim 17, further comprising applying a pattern matching algorithm to determine a correlation between the first and second sets of corner points.

19. A computer-implemented method as claimed in claim 18, wherein a weighting is applied to each corner point in said first set and/or said second set based on an included angle and/or curvature at each corner point.

20. A computer-implemented method as claimed in claim 16, wherein said pattern matching algorithm is a dynamic time warping algorithm.

21. A computer-implemented method as claimed in claim 16, wherein the plurality of pre-defined non-overlapping groups comprises a plurality of rows of the virtual keyboard.

22. A computer-implemented method as claimed in claim 16, wherein the one or more candidates each comprise a word, a phrase, or a sentence.

23. A computing device comprising an electronic processor configured to execute instructions to:

display a virtual keyboard comprising a plurality of keys arranged in a plurality of pre-defined non-overlapping groups on a display screen, each pre-defined non-overlapping group allocated an associated one of a plurality of unique identifier codes, one or more characters being associated with each of said keys;

detect an input pattern generated by a user on the display screen over the plurality of pre-defined non-overlapping groups of the virtual keyboard;

encode the input pattern by generating a sequence of the plurality of unique identifier codes corresponding to the input pattern over the plurality of pre-defined non-overlapping groups;

identify one or more candidates to be input into the electronic device based on the encoded input pattern; and

for a candidate of the one or more candidates;

identify a plurality of parts of the candidate that have associated path lengths that fall within an upper path length threshold and a lower path length threshold, wherein the plurality of parts form a sequence, and wherein each next part in the sequence has a same set of characters as a prior part in the sequence and an additional character of the candidate;

determine a plurality of string distances, each string distance being between a respective one of the plurality of parts and the input pattern;

identify a part from among the plurality of parts that has a best string distance from among the plurality of string distances; and

add the candidate to a list for output by the electronic device based on a comparison between the best string distance and a string distance threshold.

24. A computing device as claimed in claim 23, wherein the plurality of pre-defined non-overlapping groups comprises a plurality of rows of the virtual keyboard.

25. A computing device as claimed in claim 23, wherein the one or more candidates each comprise a word, a phrase, or a sentence.