Method and apparatus for recording of electronic handwriting

• US 20030122802A1
• Filed: 12/27/2002
• Published: 07/03/2003
• Est. Priority Date: 12/28/2001
• Status: Active Grant

• Current Assignee: Anoto Group AB
• Original Assignee: Anoto Group AB
• Inventors: Bryborn, Mattias
• Granted Patent: US 7,162,087 B2
• Time in Patent Office: Days
• US Class Current: 345/173
• CPC Class Codes: G06V 10/19 by sensing codes defining pattern positions
• CPC Class Codes: G06V 30/1423 the instrument generating sequences of position coordinates corresponding to handwriting preprocessing or recognising digital ink G06V30/32

Abstract

According to a method for recording of electronic handwriting, a handwriting movement is digitized into a sequence of points along a writing curve that corresponds to the handwriting movement. The sequence of points is resampled by eliminating those points that are judged not to be necessary for being able to reconstruct the writing curve later with the required visual quality. A resampled representation of the writing curve is created from the points remaining—that is not eliminated—after the resampling. The resampling is carried out on successive and partially overlapping partial sequences of the points, the resampled representation of the writing curve being extended successively after resampling of a respective partial sequence.

Claims

1. A method for recording of electronic handwriting, in which

a handwriting movement is digitized into a sequence of points along a writing curve that corresponds to the handwriting movement;

the sequence of points is resampled by eliminating those points that are judged not to be necessary for being able to reconstruct the writing curve later with the required visual quality; and

a resampled representation of the writing curve is created from the points remaining—

that is, not eliminated—

after the resampling, wherein

the resampling is carried out on successive and partially overlapping partial sequences of the points, and

the resampled representation of the writing curve is extended successively after resampling of a respective partial sequence.

2. The method as claimed in claim 1, in which the resampling is carried out according to an algorithm for polygonal approximation of the writing curve, the curve segment between two consecutive remaining points being represented by a straight line.

3. The method as claimed in claim 2, in which said algorithm is the Douglas-Peucker'"'"'s algorithm.

4. The method as claimed in claim 1, in which a partial sequence comprises n points, the overlap between two consecutive partial sequences is w points, and 1<

w<

=n/2, an individual partial sequence being stored during the resampling in a memory buffer with room for n points, and the last w points in this individual partial sequence being left in the memory buffer and being included in the resampling of a next following partial sequence.

5. The method as claimed in claim 4, in which, after the resampling of said individual partial sequence, the remaining—

not eliminated—

points in the memory buffer are added to the resampled representation of the writing curve, with the exception of the remaining points from a subset of said last w points in said individual partial sequence, the subset consisting of x points, which instead are included in the resampling of a next following partial sequence as regards addition to the resampled representation of the writing curve.

6. The method as claimed in claim 5, comprising the steps of

a) filling the memory buffer with points;

b) applying a line-generalization algorithm to the points in the memory buffer, wherein such remaining points are marked which are not eliminated by the algorithm;

c) letting the marked points from the points 1 to n−

x in the memory buffer be incorporated in the resampled representation of the writing curve;

d) deleting the first n−

w points in the memory buffer;

e) placing the remaining w points at the start of the memory buffer;

f) as long as points are available for resampling;

f1) filling up the memory buffer with n−

w new points;

f2) applying the line-generalization algorithm to the points in the memory buffer, in accordance with step b);

f3) letting the marked points from the points w−

x to n−

x in the memory buffer be incorporated in the resampled representation of the writing curve;

f4) deleting the first n−

w points in the memory buffer; and

f5) placing the remaining w points at the start of the memory buffer; and

g) letting the marked points from the remaining points n−

x+1 to n in the memory buffer be incorporated in the resampled representation of the writing curve.

7. The method as claimed in claim 5, in which x=w/2.

8. The method as claimed in claim 5, in which n=128 and w≈

n/10.

9. The method according to claim 1, comprising the final steps, after resampling, of

based on the resampled representation of the writing curve, relative coding the coordinates for the points remaining therein;

and

compressing the relative-coded coordinates using statistical coding.

10. The method as claimed in claim 9, in which the coordinates for an individual remaining point are relative coded by approximating a polynomial of order N to the N+1 preceding remaining points;

determining a prediction of said individual point using said polynomial;

determining a deviation between said individual point and its prediction; and

using this deviation as a relative coding for said individual point.

11. The method as claimed in claim 1, in which the resampled representation of the writing curve is stored as a data set in a memory.

12. The method as claimed in claim 1, in which the resampling is carried out as soon as the handwriting movement has been digitized into a predetermined number of points.

13. The method as claimed in claim 4, in which points remaining—

not eliminated—

in the memory buffer after the resampling of said individual partial sequence are added to the resampled representation of the writing curve, specifically in such a way that the remaining points from the overlapping w points between two consecutive partial sequences are added to the resampled representation of the writing curve, if they have not been eliminated during resampling of at least one of these two consecutive partial sequences.

15. The apparatus as claimed in claim 14, in which the resampling means is arranged to carry out the method according to claim 1.

17. A computer program product that can be directly read into a memory associated with a processor, comprising program code for carrying out the steps according to claim 1.

18. A method as claimed in claim 6, wherein said line-generalization algorithm is the Douglas-Peucker'"'"'s algorithm.

14. An apparatus for recording of electronic handwriting, comprising digitizing means, resampling means and a memory, said digitizing means being arranged to digitize a handwriting movement into a sequence of points along a writing curve corresponding to the handwriting movement and said resampling means being arranged to eliminate those points in the sequence of points that are judged not to be necessary for being able to reconstruct the writing curve later with the required visual quality, and to create a resampled representation of the writing curve from the points remaining—

that is, not eliminated—

after the resampling, wherein

said resampling means is arranged to carry out the elimination of points for successive and partially overlapping partial sequences of points, and to extend the resampled representation of the writing curve successively after resampling of a respective partial sequence.

16. The apparatus as claimed in claim 14, in the form of an electronic pen.