Classifying digital ink into a writing or a drawing

• US 20080292190A1
• Filed: 05/21/2007
• Published: 11/27/2008
• Est. Priority Date: 05/21/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Wang, Xin, Ye, Ming, Biswas, Manoj, Raghupathy, Sashi
• Granted Patent: US 7,945,097 B2
• Time in Patent Office: Days
• US Class Current: 382/187
• CPC Class Codes: G06F 3/04883 for inputting data by handwriting, e.g. gesture or text
• CPC Class Codes: G06V 30/1423 the instrument generating sequences of position coordinates corresponding to handwriting preprocessing or recognising digital ink G06V30/32
• CPC Class Codes: G06V 30/32 Digital ink

Abstract

A method for classifying digital ink receives digital ink comprising ink strokes. A plurality of the ink strokes can be classified. A temporal line grouping is performed on a plurality of the classified ink strokes that are grouped to form a temporal line group. The temporal line group is segmented into a cluster. The cluster can be classified.

Claims

1. A computer-implemented method for classifying digital ink, said method comprising:

receiving digital ink comprising ink strokes;

classifying a plurality of said ink strokes;

performing a temporal line grouping on a plurality of said classified ink strokes to form a temporal line group;

segmenting said temporal line group into a cluster; and

classifying said cluster.

2. The computer-implemented method as recited in claim 1, wherein said classifying said cluster comprises:

utilizing a machine learning technique.

3. The computer-implemented method as recited in claim 2, wherein said machine learning technique is selected from the group consisting of a Hidden Markov Model, a decision tree, an AdaBoost decision tree, a neural network, a support vector machine, a conditional random field technique, and a machine learning classification technique.

4. The computer-implemented method as recited in claim 1, wherein said performing a temporal line grouping comprises:

utilizing time information to order said plurality of said classified ink strokes.

5. The computer-implemented method as recited in claim 1, wherein said method comprises a multiple stage classification of said digital ink.

6. The computer-implemented method as recited in claim 1, wherein said classifying said cluster is selected from the group consisting of:

a writing, a drawing, and a composite group.

7. The computer-implemented method as recited in claim 1, wherein said classifying said plurality of said ink strokes is selected from the group consisting of:

a writing and a drawing.

8. A computer-readable medium having computer-executable instructions for performing a method for classifying digital ink, said instructions comprising:

receiving digital ink comprising a plurality of digital ink strokes;

determining if a plurality of said ink strokes is a writing or a drawing;

producing a temporal line group from a plurality of said ink strokes determined to be writing;

producing a cluster from said temporal line group; and

determining what is said cluster.

9. The computer-readable medium of claim 8, wherein said method comprises a multiple stage classification of said digital ink, wherein a later stage uses context information produced by an earlier stage.

10. The computer-readable medium of claim 8, wherein said computer-executable instructions for performing the producing said temporal line group further comprising:

utilizing a dynamic programming technique.

11. The computer-readable medium of claim 10, wherein said computer-executable instructions for performing the determining if said plurality of said ink strokes are said writing or said drawing further comprises computer executable instructions for:

utilizing a machine learning technique.

12. The computer-implemented method as recited in claim 11, wherein said machine learning technique is selected from the group consisting of a decision tree, a Hidden Markov Model, an AdaBoost decision tree, a neural network, a support vector machine, a conditional random field technique, and a machine learning classification technique.

13. The computer-readable medium of claim 8, wherein said computer-executable instructions for performing said determining what is said cluster further comprises computer executable instructions for:

utilizing a machine learning technique.

14. The computer-readable medium of claim 8, wherein said determining what is said cluster is selected from the group consisting of a writing, a drawing, and a composite group.

15. A system for classifying digital ink, said system comprising:

a stroke classification module configured for receiving digital ink comprising ink strokes, said stroke classification module configured for classifying a plurality of said ink strokes as a writing or as a drawing;

a temporal line grouping module coupled with said stroke classification module, said temporal line grouping module configured for receiving any ink strokes classified as said writing by said stroke classification module, said temporal line grouping module configured for performing a temporal line grouping on said any ink strokes and for producing a line group; and

a line-level classification module coupled to receive said line group and configured for segmenting said line group into a plurality of clusters, said line-level classification module configured for classifying a cluster of said plurality of clusters as a writing or a drawing.

16. The system of claim 15, wherein said line-level classification module utilizes a machine learning technique for segmenting said line group.

17. The system of claim 16, wherein said machine learning technique is selected from the group consisting of a classification based on a feature of an elements of said line group, a sequence of classifications of said line group, a Hidden Markov Model, a decision tree, an AdaBoost decision tree, a neural network, a support vector machine, a conditional random field technique, and a machine learning classification technique.

18. The system of claim 15, wherein said system comprises a multiple stage classification of said digital ink, wherein a later stage corrects a classification of produced by an earlier stage.

19. The system of claim 15, wherein said stroke classification module is configured for receiving said digital ink in real-time.

20. The system of claim 15, wherein said line-level classification module is configured for outputting any classification of said cluster.