SELECTION EDITING USING A LOCALIZED LEVEL SET ALGORITHM

• US 20140258941A1
• Filed: 03/08/2013
• Published: 09/11/2014
• Est. Priority Date: 03/08/2013
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Kim, Byungmoon, Lim, JaeGwang, Hadap, Sunil
• Granted Patent: US 9,547,410 B2
• Time in Patent Office: Days
• US Class Current: 715/862
• CPC Class Codes: G06F 2203/04805 Virtual magnifying lens, i.e. window or frame movable on top of displayed information to enlarge it for better reading or selection
• CPC Class Codes: G06F 3/04812 Interaction techniques based on cursor appearance or behaviour, e.g. being affected by the presence of displayed objects
• CPC Class Codes: G06F 3/04842 Selection of displayed objects or displayed text elements G06F3/0482 takes precedence
• CPC Class Codes: G06F 3/04845 for image manipulation, e.g. dragging, rotation, expansion or change of colour
• CPC Class Codes: G06T 2207/20104 Interactive definition of region of interest [ROI]
• CPC Class Codes: G06T 7/13 Edge detection

Abstract

Techniques are disclosed for selecting a targeted portion of a digital image. In one embodiment, a selection cursor having central and peripheral regions is provided. The central region is used to force a selection or a deselection, and therefore moving the central region over a portion of the image causes that portion of the image to be selected or deselected, respectively. The peripheral region of the cursor surrounds the central region and defines an area where a localized level set algorithm for boundary detection is performed. This provides more accurate boundary detection within the narrowly-focused peripheral region and eliminates the need to apply the level set algorithm across the entire image. Thus moving the peripheral region of the selection cursor over a boundary of the targeted portion of the image applies the level set algorithm in that boundary region and increases the likelihood that the boundary will be detected accurately.

Claims

1. A method for defining a selection mask associated with a source image, the method comprising:

displaying a selection cursor over the source image, the selection cursor having an interior region and a peripheral region; and

locating an initial position of a selection mask boundary in a portion of the source image that is overlaid by the peripheral region of the selection cursor, wherein the initial position of the selection mask boundary is at least partially based on a boundary that is detected as a result of selectively applying a boundary detection algorithm to a sub-region of the source image.

2. The method of claim 1, further comprising expanding the selection mask boundary so as to expand the selection region within the peripheral region of the selection cursor, wherein the expansion is constrained by the boundary that is detected as a result of applying the boundary detection algorithm.

3. The method of claim 1, further comprising contracting the selection mask boundary so as to shrink the selection region within the peripheral region of the selection cursor, wherein the contraction is constrained by the boundary that is detected as a result of applying the boundary detection algorithm.

4. The method of claim 1, further comprising snapping the selection mask boundary to the boundary that is detected as a result of applying the boundary detection algorithm.

5. The method of claim 1, wherein:

the selection cursor has a positive polarity mode and a negative polarity mode;

the interior region of the selection cursor defines a region within the selection mask when the selection cursor is in the positive polarity mode; and

the interior region of the selection cursor defines a region outside the selection mask when the selection cursor is in the negative polarity mode.

6. The method of claim 1, wherein:

the selection cursor includes a positive polarity mode in which the interior region of the selection cursor defines a region within the selection mask;

the selection cursor includes a negative polarity mode in which the interior region of the selection cursor defines a region outside the selection mask; and

the positive and negative polarity modes are selected using a control provided on a user interface device.

7. The method of claim 1, wherein the boundary detection algorithm is applied to that sub-region of the source image that is overlaid by the peripheral region of the selection cursor.

8. The method of claim 1, wherein the boundary detection algorithm is a level set algorithm.

9. The method of claim 1, wherein the boundary detection algorithm is a level set algorithm applied using an adaptive quad tree arid.

10. The method of claim 1, further comprising moving the selection mask boundary to a modified position in response to the interior region of the selection cursor passing over the initial position of the selection mask boundary.

11. An image segmentation system, comprising:

a user interface module configured to overlay on a displayed source image a selection cursor having an interior region and a peripheral region; and

an boundary detection module configured to detect a boundary within a region of the source image that is overlaid by the peripheral region, wherein the boundary is detected by applying a level set algorithm to a portion of the source image.

12. The image segmentation system of claim 11, wherein:

the selection cursor includes a positive polarity mode in which the interior region of the selection cursor defines a region within a selection region that corresponds to the boundary detected by the boundary detection module; and

the selection cursor includes a negative polarity mode in which the interior region of the selection cursor defines a region outside a selection region that corresponds to the boundary detected by the boundary detection module.

13. The image segmentation system of claim 11, wherein the level set algorithm is applied to that portion of the source image that is overlaid, by the peripheral region.

14. The image segmentation system of claim 11, further comprising a display module configured to display an edge of a selection region that corresponds to the boundary detected by the edge detection module.

15. The image segmentation system of claim 11, further comprising a boundary smoothing module configured to smoothen the detected boundary.

16. The image segmentation system of claim 11, wherein the boundary is detected by applying a level set algorithm using an adaptive quad tree grid.

17. A non-transient computer readable medium having instructions encoded thereon that, when executed by one or more processors, causes an image segmentation process to he carried out, the process comprising:

displaying a selection cursor over the source image, the selection cursor having an interior region and a peripheral region; and

setting an initial position of a selection mask boundary in a portion of the source image that is overlaid by the peripheral region of the selection cursor, wherein the initial position of the selection mask boundary is at least partially based on a boundary that is detected as a result of selectively applying a boundary detection algorithm to a sub-region of the source image.

18. The non-transient computer readable medium of claim 17, wherein:

the selection cursor includes a positive polarity mode in which the interior region of the selection cursor defines a region within the selection mask; and

the selection cursor includes a negative polarity mode in which the interior region of the selection cursor defines a region outside the selection mask.

19. The non-transient computer readable medium of claim 17, wherein the boundary detection algorithm is a level set algorithm applied using an adaptive quad tree grid.

20. The non-transient computer readable medium of claim 17, wherein the image segmentation process further comprises moving the selection mask boundary to a modified position in response to the interior region of the selection cursor passing over the initial position of the selection mask boundary.