Method of recognizing handwritten input

• US 5,600,735 A
• Filed: 05/10/1994
• Issued: 02/04/1997
• Est. Priority Date: 05/10/1994
• Status: Expired due to Term

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Seybold, John L. C.
• Primary Examiner(s): Mancuso, Joseph
• Application Number:
  US08/240,405
• Time in Patent Office: 1,001 Days
• Field of Search: 382/174, 382/177, 382/178, 382/179, 382/186, 382/185
• US Class Current: 382/178
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/15 Cutting or merging image elements, e.g. region growing, watershed or clustering-based techniques

Abstract

The present invention determines whether two discrete continuous segments of handwritten input S₁ (210) and S₂ (220) form part of the same handwritten input or are part of more than one, separate handwritten inputs. The present method calculates one or more substantially parallel distance disposed substantially parallel to the writing access (210) and compares these distances to one or more predefined thresholds. The predefined thresholds specify minimum distance measures which must be exceeded by the substantially parallel distances for the discrete continuous segments to be judged as belonging to separate segments of handwritten input.

Claims

1. A method comprising the steps of:

receiving handwritten character input, which handwritten character input is comprised of at least a first and a second discrete continuous segment;

calculating of at least one distance between the first and second discrete continuous segments, the distance being substantially parallel to a writing axis of the first and second discrete continuous segments including the steps of;

calculating a plurality of distances between the first and second discrete continuous segments which are substantially parallel to the writing axis of the first and second discrete continuous segments;

identifying which of the plurality of distances has a smallest value; and

using the distance to determine whether the first and second discrete continuous segments belong to separate handwritten character inputs.

2. The method of claim 1, wherein the step of using the parallel distance to determine whether the first and second discrete continuous segments belong to separate handwritten inputs includes the step of determining whether the first and second discrete continuous segments belong to separate handwritten inputs based upon the smallest value of the plurality of distances.

3. A method comprising the steps of:

receiving handwritten character input, which handwritten character input is comprised of at least a first and a second discrete continuous segment;

calculating at least one distance between the first and second discrete continuous segments, the distance being substantially parallel to a writing axis of the first and second discrete continuous segments;

identifying a boundary for each of the first and second discrete continuous segments, which boundary for each of the first and second discrete continuous segments is substantially perpendicular to the writing axis;

calculating a distance between the boundaries for the first and second discrete continuous segments; and

using the distance between the boundaries for the first and second discrete continuous segments to determine whether the first and second discrete continuous segments belong to separate handwritten inputs.

4. A method comprising the steps of:

receiving handwritten input comprising first and second continuous segments, the handwritten input having a first axis defined along a direction of writing of the handwritten input;

determining a boundary for each of the first and second continuous segments, the boundary for each of the first and second continuous segments being defined along a second axis perpendicular to the first axis through a point along the first axis for each of the first and second continuous segments nearest the boundary of the other of the first and second continuous segments;

calculating a distance parallel to the first axis between the boundaries; and

determining from the distance whether the first and second continuous segments are separate handwritten inputs, including the steps of;

comparing the distance to a first predetermined threshold;

determining that the first and second continuous segments are separate handwritten inputs when the distance is greater than the first predetermined threshold;

separating each of the first and second continuous segments into a plurality of sections when the distance is not greater than the first predetermined threshold;

analyzing each of the plurality of sections to determine which of the plurality of sections has a portion of each of the first and second continuous segments;

for each of the plurality of sections where the portion of each of the first and second continuous segments has been found, calculating a segment distance between the portions of the first and second continuous segments that are nearest to each other; and

determining from the distance and the segment distance whether the first and second continuous segments are separate handwritten input.

5. A method according to claim 4 wherein the step of determining from the distance and the section distance whether at least first and second continuous segment are separate handwritten input comprises the steps of:

calculating a weighted average of the distance and the segment distance;

comparing the weighted average to a second predetermined threshold;

determining that the at least first and second continuous segment are separate handwritten input if the weighted average is greater than the second predetermined threshold; and

determining that the at least first and second continuous segment are not separate handwritten input if the weighted average is not greater than the second predetermined threshold.

6. A method according to claim 4 wherein the sections are distributed in a direction perpendicular to the writing axis.

7. A method according to claim 6 wherein the sections are of equal breadth in a direction perpendicular to the writing axis.

8. A method, comprising the steps of:

processing first and second discrete continuous segments representing handwritten input entered along a writing axis to identify a first boundary for the first discrete continuous segment and a second boundary for the second discrete continuous segment, the first boundary representing a greatest displacement of the first discrete continuous segment from a reference and the second boundary representing a least displacement of the second discrete continuous segment from the reference;

calculating a first distance between the first and second boundary, the first distance being substantially parallel to the writing axis;

determining that the first and second discrete continuous segments represent a separate handwritten character when the first distance exceeds a first threshold;

determining a second distance when the first distance does not at least exceed the first threshold, the second distance being substantially parallel to the writing axis and selected as a minimum distance from a plurality of distances between left-most points of the first discrete continuous segment and right-most points of the second discrete continuous segment in a like plurality of bands being substantially parallel to the writing axis; and

determining that the first and second discrete continuous segments represent a separate handwritten character when the second distance exceeds a second threshold.

9. A method comprising the steps of:

receiving handwritten character input, which handwritten character input is comprised of at least a first and a second discrete continuous segment positioned along a writing axis;

defining a first and a second boundary which are substantially perpendicular to the writing axis;

calculating the first boundary by finding the point in the first discrete continuous segment that has the largest displacement from a reference point along the writing axis;

calculating the second boundary by finding the point in the second discrete continuous segment that has the smallest displacement from the reference point along the writing axis;

determining that the largest displacement of the point in the first discrete continuous segment is greater than the smallest displacement of the point in the second discrete continuous segment;

calculating a negative value by subtracting the smallest displacement of the point of the second discrete continuous segment from the largest displacement of the point of the first discrete continuous segment; and

using the negative value to determine whether the first and second discrete continuous segments belong to separate handwritten inputs.