Video image segmentation

• US 8,249,152 B2
• Filed: 08/25/2008
• Issued: 08/21/2012
• Est. Priority Date: 08/25/2008
• Status: Expired due to Fees

• Current Assignee: Hankuk University of Foreign Studies Research and Industry-University Cooperation Foundation
• Original Assignee: Hankuk University of Foreign Studies Research and Industry-University Cooperation Foundation
• Inventors: Yun, Il Dong
• Primary Examiner(s): Sefcheck, Gregory
• Assistant Examiner(s): SMITH, JOSHUA Y
• Application Number:
  US12/198,061
• Publication Number:
  US 20100046605A1
• Time in Patent Office: 1,457 Days
• Field of Search: None
• US Class Current: 375/240.07
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/174 involving the use of two or more images
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation

Abstract

Techniques for segmenting an image of an object are provided. In some embodiments, an image segmentation apparatus includes an image input unit and an image processing unit. The image input unit is configured to receive as input a video image having a first image frame and a second image frame that is consecutive to the first image frame. The image processing unit is configured to segment the second image frame based at least in part on information on the first image frame.

Claims

1. An apparatus comprising:

an image input unit configured to receive as input a video image having a first image frame and a second image frame, the second image frame being consecutive to the first image frame; and

an image processing unit configured to segment the second image frame based at least in part on information on the first image frame,

wherein the image processing unit comprises a foreground segmentation unit configured to segment the first image frame to determine a boundary between a foreground and a background of the first image frame, an attenuation vector generating unit configured to generate an attenuation vector from the segmented first image frame, and a contrast attenuating unit configured to attenuate a contrast of the second image frame by using the attenuation vector, and

the attenuation vector generating unit is configured to generate the attenuation vector based on a difference between the contrast for a pixel in the second image frame and the contrast for a pixel of the boundary of the first image frame,

the attenuation vector generating unit is configured to measure a difference between a contrast at a pixel of the second image frame and a contrast at a certain point of the boundary of the first image frame to generate a component of the attenuation vector, and

the attenuation vector generating unit is configured to measure distances between the pixel of the second image frame and two different points of the boundary, and generates the component of the attenuation vector according to the distances and the inner product of components of the contrast at the pixel of the second image frame.

2. The apparatus of claim 1, further comprising a display configured to display the segmented second image frame.

3. The apparatus of claim 1, wherein the foreground segmentation unit is configured to segment the second image frame.

4. The apparatus of claim 1, wherein the attenuation vector generating unit calculates a Gradient Vector Flow (GVF) from the boundary.

5. The apparatus of claim 1, wherein the foreground segmentation unit applies a segmentation technique to segment the first image frame into a foreground and a background.

6. The apparatus of claim 5, wherein the segmentation technique includes a graph cut and a grab cut methods.

7. The apparatus of claim 1, wherein the apparatus is a portable device.

8. The apparatus of claim 7, wherein the apparatus is a cellular phone.

9. A method comprising:

receiving a video image having a first image frame and a second image frame, the second image frame being consecutive to the first image frame; and

processing the video image to segment the second image frame based at least in part on information on the first image frame,

wherein processing the video image comprises segmenting the first image frame to determine a boundary between a foreground and a background of the first image frame, generating an attenuation vector from the segmented first image frame, and attenuating the contrast of the second image frame by using the attenuation vector,

the attenuation vector is generated based on a difference between the contrast for a pixel in the second image frame and the contrast for a pixel of the boundary of the first image frame,

the generating an attenuation vector includes measuring a difference between a contrast at a pixel of the second image frame and a contrast at a certain point of the boundary of the first image frame to generate the component of the attenuation vector, and

the generating an attenuation vector includes measuring distances between the pixel of the second image frame and two different points of the boundary, and generating the component of the attenuation vector according to the distances and the inner product of components of the contrast at the pixel of the second image frame.

10. The method of claim 9, further comprising displaying the segmented second image frame.

11. The method of claim 9, wherein processing the video image further comprises segmenting the second image frame.

12. The method of claim 9, wherein generating an attenuation vector includes calculating a Gradient Vector Flow (GVF) from the boundary to generate an attenuation vector map.

13. The method of claim 11, wherein segmenting the first image frame and segmenting the second image frame include applying a segmentation technique to segment the first image frame and the second image frame into a foreground and a background respectively.

14. The method of claim 13, wherein the segmentation technique includes a graph cut and a grab cut methods.

15. A non-transitory computer-readable storage medium having stored thereon computer instructions that, when executed by a computer, cause the computer to perform a method comprising:

receiving a video image having a first image frame and a second image frame, the second image frame being consecutive to the first image frame; and

processing the video image to segment the second image frame based on information on the first image frame,

wherein processing the video image comprises segmenting the first image frame to determine a boundary between a foreground and a background of the first image frame, generating an attenuation vector from the segmented first image frame, and attenuating the contrast of the second image frame by using the attenuation vector,

the attenuation vector is generated based on a difference between the contrast for a pixel in the second image frame and the contrast for a pixel of the boundary of the first image frame,

the generating an attenuation vector includes measuring a difference between a contrast at a pixel of the second image frame and a contrast at a certain point of the boundary of the first image frame to generate the component of the attenuation vector, and

the generating an attenuation vector includes measuring distances between the pixel of the second image frame and two different points of the boundary, and generating the component of the attenuation vector according to the distances and the inner product of components of the contrast at the pixel of the second image frame.