Method, device and computer program for recognition of a handwritten character

• US 7,349,576 B2
• Filed: 01/11/2002
• Issued: 03/25/2008
• Est. Priority Date: 01/15/2001
• Status: Expired due to Fees

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: Zi Decuma AB
• Inventors: Holtsberg, Anders
• Primary Examiner(s): CHAWAN, SHEELA C
• Application Number:
  US10/466,087
• Publication Number:
  US 20040096105A1
• Time in Patent Office: 2,265 Days
• Field of Search: 382/187, 382/186, 382/185, 382/229, 382/197, 382/214, 382/179, 382/189, 382/310, 382/227, 382/192, 382/204, 382/218, 382/198, 382/201, 382/209, 382/200, 382/203, 382/242, 382/224, 382/159, 382/228, 382/190, 382/298, 382/202, 704/251, 704/256, 704/256.1, 704237-244
• US Class Current: 382/187
• CPC Class Codes: G06V 30/1423 the instrument generating sequences of position coordinates corresponding to handwriting preprocessing or recognising digital ink G06V30/32

Abstract

A method for recognition of a handwritten character comprises the steps of determining a plurality of position features defining the handwritten character, and comparing the handwritten character to reference characters stored in a database in order to find the closest matching reference character. The step of comparing comprises the steps of computing a difference between one of the plurality of position features of the handwritten character and a corresponding position feature of one of the reference characters, determining, by lookup in a predefined table, a distance measure based on the computed difference and determining a distance measure for each of the plurality of position features of the handwritten character, and computing a cost function based on the determined distance measures. A device and a computer program for implementing the method are also described.

Claims

1. A method for recognition of a handwritten character, which method comprises the steps of:

determining a plurality of position features defining the handwritten character, and

comparing the handwritten character to reference characters stored in a database in order to find the closest matching reference character, said step of comparing comprising the steps of;

computing a difference between one of the plurality of position features of the handwritten character and a position feature of one of the reference characters,

determining, by lookup in a predefined table, a distance measure based on the computed difference,

repeating the steps of computing a difference and determining a distance measure for each of the plurality of position features of the handwritten character, and

computing a cost function based on the determined distance measures.

2. The method according to claim 1, wherein each position feature is defined by at least two characteristics and wherein the step of computing a difference between a position feature of the handwritten character and a position feature of the reference character comprises the computing of a difference between each characteristic of the position feature of the handwritten character and the corresponding characteristics of the position feature of the reference character.

3. The method according to claim 2, wherein the step of computing a difference for each characteristic is performed in a single operation.

4. The method according to any one of the preceding claims, wherein the step of determining position features comprises the step of determining a location of the position features in a grid.

5. The method according to claim 4, further comprising the step of centering the handwritten character in the grid before the step of determining the position features of the handwritten character is performed.

6. The method according to any one of claims 1 to 3, wherein the position features of the handwritten character and the reference characters comprise coordinates of comparison points.

7. The method according to claim 6, wherein the step of determining position features comprises the step of determining an x-coordinate and a y-coordinate of the comparison point in a grid.

8. The method according to claim 7, wherein the grid is of a size of 16×

16 positions and the coordinates of one comparison point are represented by one data byte.

9. The method according to claim 6, wherein the distance measure received from the table lookup in the predefined table is a function of the Euclidean distance between the comparison point of the handwritten character and a comparison point of the reference character.

10. The method according to any one of claims 1 to 3, wherein the position features of the handwritten character and the reference characters comprise positions and directions of comparison vectors.

11. The method according to claim 10, wherein the distance measure received from the table lookup in the predefined table is a function of a difference between curvatures of the handwritten character and the reference characters.

12. The method according to any one of claims 1 to 3, further comprising the step of scaling the handwritten character to a normalized size in accordance with the size of the reference characters, before the step of determining the position features of the handwritten character is performed.

13. The method according to any one of claims 1 to 3, further comprising the step of determining the number of strokes that forms the handwritten character.

14. The method according to claim 13, wherein the step of comparing the handwritten character to reference characters comprises the step of initially comparing the handwritten character to reference characters that are formed by the same number of strokes.

15. The method according to claim 13, wherein the step of comparing further comprises the step of comparing the handwritten character to reference characters that are formed by a few strokes fewer than the handwritten character.

16. The method according to claim 13 wherein the step of comparing further comprises the step of comparing the handwritten character to reference characters that are formed by a few strokes more than the handwritten character.

17. The method according to claim 13, wherein a stroke of the handwritten character is compared to a corresponding stroke of the reference character, wherein the corresponding strokes are determined by the order in which the strokes of the handwritten character are drawn.

18. The method according to claim 17, further comprising the step of reorganizing the order of the strokes of the handwritten character.

19. A device for recognition of a handwritten character and for transforming it into a digital format, comprising:

means for determining a plurality of position features defining the handwritten character,

a database, comprising reference characters,

means for comparing the handwritten character to said reference characters of said database in order to find the closest, matching reference character, said means for comparing comprising;

means for computing a difference between one of the plurality of position features of the handwritten character and a position feature of one of the reference characters,

a predefined table, defining a distance measure based on differences in position features,

means for determining a distance measure by lookup in said predefined table, and

means for computing a cost function based on the determined distance measures.

20. The device according to claim 19, wherein a position feature comprises two characteristics defining it and said characteristics are stored in one data register in the database.

21. A handheld device for data handling comprising a pressure-sensitive area, which is arranged for manual writing of a character and a device according to claim 19 for recognizing a character written in the pressure-sensitive area for transforming it into a digital format.

22. A computer-readable medium storing instructions, which when executed by a computer, perform a method for recognition of a handwritten character, the method comprising:

determining a plurality of position features defining the handwritten character, and

comparing the handwritten character to reference characters stored in a database in order to find the closest matching reference character, said code for comparing comprising;

computing a difference between one of the plurality of position features of the handwritten character and a position feature of one of the reference characters,

determining, by lookup in a predefined table, a distance measure based on the computed difference,

repeating the steps of computing a difference and determining a distance measure for each of the plurality of position features of the handwritten character, and

computing a cost function based on the determined distance measures.