Method of modifying classification scheme stored in processing system

• US 7,760,947 B2
• Filed: 11/26/2008
• Issued: 07/20/2010
• Est. Priority Date: 10/15/2001
• Status: Expired due to Fees

• Current Assignee: Silverbrook Research Pty Limited
• Original Assignee: Silverbrook Research Pty Limited
• Inventors: Napper, Jonathon Leigh
• Primary Examiner(s): Mariam; Daniel G
• Assistant Examiner(s): WOLDEMARIAM, AKLILU K
• Application Number:
  US12/324,646
• Publication Number:
  US 20090074295A1
• Time in Patent Office: 601 Days
• Field of Search: 382/186
• US Class Current: 382/186
• CPC Class Codes: G06V 30/244 using graphical properties, e.g. alphabet type or font

Abstract

The present invention provides a method of modifying a classification scheme for classifying hand-written characters to thereby make the classification scheme user dependent. The classification scheme includes a number of allographs, each allograph representing a respective style of a respective letter. The method includes obtaining at least one hand-written character, and then selecting an allograph representing the character. One or more of the allographs are then modified in accordance with the selection, to thereby reflect the fact that the allograph corresponds to a style of letter used by the user.

Claims

1. A method of modifying a classification scheme stored by a processing system for classifying hand-written characters to thereby make the classification scheme user dependent, the processing system storing the classification scheme to include a plurality of classes and user dependent weighted allographs, each class representing a respective letter and containing one or more of the allographs, each allograph including an associated weighting and representing a respective style of a respective letter and containing one or more prototypes, each prototype representing a variation in the respective allograph and being represented as a prototype vector, each prototype vector being formed from a number of values, each value quantifying a respective feature of the respective prototype vector, the method including the steps of:

obtaining at least one hand-written character at an input of the processing system;

determining, in the processing system, a selection value based on the similarity of the obtained character and a respective prototype vector of the stored classification scheme by;

determining a feature vector representing the character, the feature vector being formed from a number of values, each value quantifying a respective feature of the character; and

determining a distance value representing the distance between the determined feature vector and said respective prototype vector;

selecting, in the processing system, a respective one of the prototype vectors of the stored classification scheme in accordance with the determined selection values;

selecting, in the processing system, a class of the stored classification scheme in accordance with the selected prototype representing the character to thereby determine the letter represented by the character;

selecting, in the processing system, one of the stored user dependent weighted allographs in accordance with the weightings and in accordance with said selected prototype vector representing the character; and

modifying, in the processing system, one or more of the allographs in the respective class in accordance with the selection by modifying the weighting of one or more of the allographs;

storing a modified classification scheme in the processing system which includes said modified allographs.

2. The method according to claim 1, the method including modifying the allographs to reflect that the selected allograph corresponds to a style of the respective letter used by the user.

3. The method according to claim 1, the selection value being determined in accordance with the determined distance value and the weighting associated with the respective allograph.

4. The method according to claim 3, the method including modifying the weighting of at least the allograph containing the selected prototype.

5. The method according to claim 1, the method including selecting the prototype having the lowest selection value.

6. The method according to claim 1, the step of modifying the weightings including at least one of:

decreasing the weighting for the allograph containing the selected prototype; and

,

increasing the weighting for one or more of the allographs that do not contain the selected prototype.

7. The method according to claim 6, including increasing and/or decreasing the weighting of the allographs by a predetermined amount.

8. The method according to claim 1, including:

selecting a reference prototype for each allograph to be modified;

determining the distance value for each reference prototype;

determining an average of the determined distance values; and

,

modifying the weighting of each allograph in accordance with the distance value of the respective reference prototype and the average distance value.

9. The method according to claim 8, the reference prototype being the prototype having the lowest distance value for the respective allograph.

10. The method according to claim 1, wherein the processing system includes:

a store for storing allograph data representing the allographs; and

a processor configured to obtain the hand-written character at the input;

select the allograph, and modify the one or more allographs.