Method and apparatus for selecting an object in an image

• US 20050180635A1
• Filed: 07/08/2004
• Published: 08/18/2005
• Est. Priority Date: 02/17/2004
• Status: Active Grant

• Current Assignee: Corel Corporation (Corel Holdings LP)
• Original Assignee: Corel Corporation (Corel Holdings LP)
• Inventors: Trifonov, Mikhail Ivanovich, Sharonova, Olga Vadimovna
• Granted Patent: US 7,840,074 B2
• Time in Patent Office: Days
• US Class Current: 382/199
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/20116 Active contour; Active surface; Snakes
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/149 involving deformable models, e.g. active contour models
• CPC Class Codes: G06V 10/7553 based on shape, e.g. active shape models [ASM]

Abstract

A boundary in an image is identified by identifying a search region within the image. The process continues by determining image gradients in the search region and determining multiple color regions within the search region. An active contour representing the boundary is created based on the image gradients and the multiple color regions.

Claims

1. A method of identifying a boundary in an image, the method comprising:

identifying a search region within the image;

determining image gradients in the search region;

determining a plurality of color regions within the search region; and

creating an active contour based on the image gradients and the plurality of color regions.

2. The method of claim 1, further comprising minimizing an energy associated with the active contour.

3. The method of claim 1, wherein the search region is associated with an edge of an object in the image.

4. The method of claim 1, wherein the search region is identified by at least two points on a boundary of an object in the image.

5. The method of claim 1, wherein the image gradients are associated with at least a portion of an edge of an object in the image.

6. The method of claim 1, wherein determining image gradients includes detecting image gradients at a plurality of different scales.

7. The method of claim 1, further comprising:

creating a plurality of active contours; and

identifying a best active contour from the plurality of active contours.

8. The method of claim 1, wherein the search region is symmetrically disposed about a straight line joining two points on a boundary of an object in the image.

9. The method of claim 1, wherein the search region is in the shape of a rectangle.

10. The method of claim 1, wherein the search region is in the shape of an ellipse.

11. The method of claim 1, further comprising rotating the search region to create a one-dimensional active contour.

12. The method of claim 1, further comprising calculating an error function associated with the active contour, wherein the error function identifies an image energy acting on the active contour.

13. One or more computer-readable media having computer-readable instructions thereon which, when executed, implement the method of claim 1.

14. A computing system embodying the computer-readable media of claim 13.

15. A method of processing an image, the method comprising:

identifying a search region within the image, wherein the search region is identified by at least two points on a boundary of an object in the image;

determining a plurality of image gradients and a plurality of color regions within the search region;

creating a plurality of active contours based on the image gradients and the color regions; and

identifying a best active contour from the plurality of active contours.

16. The method of claim 15, wherein the best active contour most closely matches the boundary of the object in the image.

17. The method of claim 15, further comprising calculating an error function associated with each of the plurality of active contours, wherein identifying a best active contour includes identifying an active contour having a lowest associated error function value.

18. The method of claim 15, wherein the search region is symmetrically disposed about a straight line joining the at least two points on the boundary of the object in the image.

19. The method of claim 15, further comprising rotating the search region to create a one-dimensional active contour.

20. One or more computer-readable media having computer-readable instructions thereon which, when executed, implement the method of claim 15.

21. A computing system embodying the computer-readable media of claim 20.

22. One or more computer-readable media having stored thereon a computer program that, when executed by one or more processors, causes the one or more processors to:

identify a search region associated with an image;

determine a plurality of color regions within the search region;

creating an active contour based on the plurality of color regions; and

minimizing an image energy acting on the active contour.

23. The computer-readable media of claim 22, further comprising determining at least one image gradient in the search region.

24. The computer-readable media of claim 22, wherein the search region is associated with a boundary of an object in the image.

25. The computer-readable media of claim 22, wherein the search region is identified by at least two points on a boundary of an object in the image.

26. The computer-readable media of claim 22, wherein the search region is in the general shape of a rectangle.

27. An apparatus comprising:

an image selection module to identify an image;

a search region selector coupled to the image selection module, the search region selector to identify a search region within the image;

an image gradient identification module coupled to the image selection module, the image gradient identification module to determine image gradients in the search region;

a color region identification module coupled to the image selection module, the color region identification module to determine color regions in the search region; and

an active contour generation module coupled to the image gradient identification module and the color region identification module, the active contour generation module to generate at least one active contour based on the image gradients and the color regions.

28. The apparatus of claim 27, further comprising an active contour optimization module coupled to the active contour generation module, the active contour optimization module to identify a best active contour among the available active contours.

29. The apparatus of claim 27, wherein the search region selector identifies a search region based on at least two points on a boundary of an object in the image.

30. The apparatus of claim 27, wherein the image gradient identification module determines image gradients at a plurality of different scales.