Efficient detection of constant regions of an image

• US 8,224,096 B1
• Filed: 01/28/2011
• Issued: 07/17/2012
• Est. Priority Date: 03/06/2006
• Status: Expired due to Fees

• Current Assignee: Marvell Asia Pte Limited (Marvell Technology Group Limited)
• Original Assignee: Marvell International Technology, Ltd. (Marvell Technology Group Limited)
• Inventors: Shoen, Jay
• Primary Examiner(s): BHATNAGAR, ANAND P
• Application Number:
  US13/016,454
• Time in Patent Office: 536 Days
• Field of Search: 382/100, 382/144, 382/145, 382/147, 382/181, 382/190, 382/192, 382/193, 382/194, 382/195, 382/199, 382/201, 382/205, 382/206, 382/209, 382/216, 382/217, 382/221, 382/225, 382260-266, 382/271, 35842601-42616
• US Class Current: 382/195
• CPC Class Codes: G06T 7/11 Region-based segmentation

Abstract

A technique that improves image analysis efficiency by reducing the number of computations needed to detect constant regions. Constant region detection according to the present techniques includes determining whether an image analysis window at a current position contains a constant region by analyzing a new line of pixels in the image analysis window if a pixel at a predetermined location in the image analysis window in the current position has a value equal to a pixel at the predetermined location from a previous position of the image analysis window. Analyzing only the new line of pixels saves the computational time that would otherwise go into analyzing all of the pixels in the image analysis window.

Claims

1. A method for detecting a constant region of an image frame, comprising:

determining, with at least one processor, that a first region of pixels in an image analysis window at a first position is a constant region;

in response to determining that the first region is a constant region;

determining, with the at least one processor, that a second region in the image analysis window at a second position is a constant region by analyzing a number of pixels in the second region that is less than a total number of pixels in the second region.

2. The method of claim 1, further in response to determining that the first region is the constant region:

shifting, with the at least one processor, the image analysis window to the second position, wherein the second region in the image analysis window at the second position comprises a new line of pixels from the image frame that is different from any line of pixels in the image analysis window at the first position.

3. The method of claim 2, wherein analyzing the number of pixels that is less than the total number of pixels in the second region comprises:

determining that a value of a first pixel in the second region is equal to a value of a second pixel in the first region, wherein the first pixel and the second pixel are located at a predetermined location, and wherein the predetermined location is relative to a boundary of the image analysis window; and

analyzing only the new line of pixels after determining that the value of the first pixel is equal to the value of the second pixel.

4. The method of claim 3, wherein the predetermined location is a center of the image analysis window relative to the boundary of the image analysis window.

5. The method of claim 2, wherein analyzing the new line of pixels comprises comparing a set of values of the new line of pixels to a value of the second pixel at the predetermined location when the image analysis window was at the first position.

6. The method of claim 2, wherein shifting the image analysis window comprises shifting the image analysis window by one pixel.

7. The method of claim 1, further comprising analyzing all of the pixels in the image analysis window at the second position when a value of a first pixel in the second region is not equal to a value of a second pixel in the first region, wherein the first pixel and the second pixel are located at a predetermined location, and wherein the predetermined location is relative to a boundary of the image analysis window.

8. The method of claim 1, wherein determining that the second region in the image analysis window at the second position is a constant region comprises determining that all of the pixels in the image analysis window are substantially similar in value.

9. A device comprising:

an image buffer that holds an image frame; and

an image processor that is configured to detect a constant region at a current position of the image frame by;

determining that an entire area of an image analysis window at a previous position is a constant region; and

in response to determining that the entire area of the image analysis window at the previous position is a constant region,

determining that an entire area of the image analysis window at the current position is a constant region by analyzing a number of pixels in the image analysis window that is less than a total number of pixels in the image analysis window at the current position.

10. The device of claim 9, wherein the image processor is further configured to detect a constant region of a current position of the image frame by:

shifting the image analysis window from the previous position to the current position so that the image analysis window at the current position includes a new line of pixels from the image frame that is different from any line of pixels in the image analysis window at the previous position.

11. The device of claim 10, wherein the image processor is configured to analyze a number of pixels in the image analysis window that is less than the total number of pixels in the image analysis window at the current position by:

determining that a value of a first pixel in the image analysis window at the current position is equal to a value of a second pixel in the image analysis window at the previous position, wherein the first pixel and the second pixel are located at a predetermined location, the predetermined location being relative to a boundary of the image analysis window; and

analyzing only the new line of pixels after determining that the value of the first pixel is equal to the value of the second pixel.

12. The device of claim 11, wherein the predetermined location is a center of the image analysis window relative to the boundary of the image analysis window.

13. The device of claim 11, wherein the image processor is configured to analyze the new line of pixels by comparing a set of values of the new line of pixels to a value of the second pixel at the predetermined location when the image analysis window was at the previous position.

14. The device of claim 10, wherein the image processor is configured to shift the image analysis window by shifting the image analysis window by one pixel.

15. The device of claim 9, wherein the image processor is further configured to detect a current constant region of an image frame by analyzing all of the pixels in the image analysis window;

wherein the image processor is configured to analyze all of the pixels in the image analysis window when a value of a first pixel in the area of the image analysis window in the current position is not equal to a value of a second pixel in the area of the image analysis window in the previous position,

wherein the first pixel and the second pixel are located at a predetermined location, and

wherein the predetermined location is relative to a boundary of the image analysis window.

16. The device of claim 9, wherein the image processor is configured to determine that an entire area of the image analysis window at the current position is a constant region by determining that all of the pixels in the image analysis window are substantially similar in value.

17. An image processing system comprising:

means for holding an image frame;

means for determining that a first region of pixels in an image analysis window at a first position is a constant region of pixels;

means for shifting the image analysis window from the first position to the second position, wherein a second region in the image analysis window at the second position comprises a new line of pixels from the image frame that is different from any line of pixels in the image analysis window at the first position;

means for detecting that the second region of the image analysis window at the second position is a constant region by analyzing a number of pixels in the second region that are less than a total number of pixels in the second region.

18. The image processing system of claim 17, wherein analyzing a number of pixels in the second region that is less than the total number of pixels in the second region comprises:

determining that a value of a first pixel in the second region is equal to a value of a second pixel in the first region, wherein the first pixel and the second pixel are located at a predetermined location, and wherein the predetermined location is relative to a boundary of the image analysis window; and

analyzing only the new line of pixels after determining that the value of the first pixel is equal to the value of the second pixel.

19. The image processing system of claim 18, wherein analyzing the new line of pixels comprises comparing a set of values of the new line of pixels to a value of the second pixel at the predetermined location when the image analysis window was at the first position.

20. The image processing system of claim 18, wherein the predetermined location is a center of the image analysis window relative to the boundary of the image analysis window.