# US 2008/0304748 A1 · Method and system for line extraction in digital ink

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2008/0304748 A1
- Document ID: US20080304748A1
- Status: Published application
- Kind: A1
- Application number: 11/944,404
- Filed: 2007-11-21
- Published: 2008-12-11
- Earliest claimed: 2003-02-26
- Assignees: Silverbrook Research Pty Ltd
- Inventors: Dimitrios Koubaroulis; Jonathon Leigh Napper; Paul Lapstun
- Web page: https://patentgrep.com/patent/US20080304748A1

## CPC

- G06V 30/333: {Preprocessing; Feature extraction}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A method and system of line extraction in a digital ink sequence of handwritten text data points is provided in which a stroke sequence comprised of a sequence of strokes is obtained, the strokes are segmented into a sequence of substrokes by applying a stroke segmentation algorithm, angular differences between neighbouring groups of substrokes, in the sequence of substrokes, are calculated, and the positions of the extrema of the angular differences are determined, thereby identifying the substrokes at line breaks and enabling line extraction.

## Claims

### Claim 1

A method of line extraction in a digital ink sequence of handwritten text, the method including the steps of:
obtaining a stroke sequence comprised of a sequence of strokes;
segmenting the strokes into a sequence of substrokes by applying a stroke segmentation algorithm;
calculating angular differences between neighbouring groups of substrokes, in the sequence of substrokes; and,
determining the positions of the extrema of the angular differences, thereby identifying the substrokes at line breaks and enabling line extraction.

### Claim 2 (depends on 1)

The method as claimed in claim 1, wherein a centroid is calculated for each substroke, the centroids being used to calculate angular differences between neighbouring groups of substrokes, in the sequence of substrokes.

### Claim 3 (depends on 1)

The method as claimed in claim 1, wherein substrokes are grouped about a selected substroke into a preceding group of N substrokes and a subsequent group of N substrokes, the centroid of the preceding group and the following group used to calculate the angular difference at the selected substroke.

### Claim 4 (depends on 1)

The method as claimed in claim 1, wherein the extrema are the maxima or minima of the sequence of angular differences as a function of substroke sequence order.

### Claim 5 (depends on 1)

The method as claimed in claim 1, wherein the stroke segmentation algorithm includes the steps of:
smoothing stroke data points;
computing a relative angle for each line segment connecting successive stroke data points;
segmenting the stroke into candidate substrokes if the difference between successive relative angles is greater than a preset threshold; and,
merging candidate substrokes into a substroke if the angular direction of successive candidate substrokes is less than a threshold angle.

### Claim 6 (depends on 1)

The method as claimed in claim 1, wherein the stroke sequence undergoes pre-processing.

### Claim 7 (depends on 6)

The method as claimed in claim 6, wherein pre-processing includes removing data points that are less than a threshold distance from a preceding data point in the digital ink sequence.

### Claim 8 (depends on 6)

The method as claimed in claim 6, wherein pre-processing includes removing strokes that are less than a threshold length dimension compared to neighbouring strokes.

### Claim 9 (depends on 1)

The method as claimed in claim 1, wherein the stroke sequence undergoes post-processing.

### Claim 10 (depends on 9)

The method as claimed in claim 9, wherein post-processing includes removing any line breaks between substrokes of a parent stroke.

### Claim 11

A method of line extraction in an on-line digital ink sequence of handwritten text data points, the method including the steps of:
(1) obtaining a stroke sequence comprised of a sequence of strokes;
(2) removing small strokes less than a threshold length;
(3) segmenting the strokes into a sequence of substrokes by:
(A) computing a relative angle for each line segment connecting successive stroke data points; and
(B) segmenting the stroke into substrokes if the difference between successive relative angles is greater than a preset threshold; and
(4) identifying the substrokes at line breaks, thereby facilitating line extraction, by:
(A) computing N-neighbourhood centroids;
(B) computing angular differences between N-neighbouring substrokes, or groups of substrokes; and
(C) computing local maxima of the angular differences in the sequence of substrokes.

### Claim 12

A system for line extraction in a digital ink sequence of handwritten text data points, the system comprising:
(1) an input device to obtain a stroke sequence comprised of a sequence of strokes;
(2) a processor in communication with the input device and adapted to:
(A) segment the strokes into a sequence of substrokes by applying a stroke segmentation algorithm;
(B) calculate angular differences between neighbouring substrokes, or groups of substrokes, in the sequence of substrokes; and,
(C) determine the positions of the extrema of the angular differences, thereby identifying the substrokes at line breaks enabling line extraction.
