Method and system for identifying objects in an image

• US 20060062458A1
• Filed: 06/27/2005
• Published: 03/23/2006
• Est. Priority Date: 03/06/2003
• Status: Active Grant

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Inventors: Lee, Harry C.
• Granted Patent: US 7,817,859 B2
• Time in Patent Office: Days
• US Class Current: 382/181
• CPC Class Codes: G06T 2207/10048 Infrared image
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes

Abstract

Methods and apparatus for identifying objects in an image include processing the image with a gradient operator to produce a gradient magnitude and direction for each pixel. A number of different gradient directions in a portion of the processed image is determined. The portion of the processed image is identified as an object if the number of different gradient directions exceeds a threshold number of gradient directions. The determination of the gradient directions can be simplified by employing a lookup table.

Claims

1. A method for identifying objects in an image comprising:

processing the image with a gradient operator to produce a gradient magnitude and direction for each pixel;

determining a number of different gradient directions in a portion of the processed image; and

identifying the portion of the processed image as an object if the number of different gradient directions exceeds a threshold number of gradient directions.

5. The method of claim 1, wherein the image is a monochrome image with image intensities on various gray levels.

6. The method of claim 1, wherein the image is processed using a Sobel operator.

7. The method of claim 1, wherein the step of processing the image comprises:

determining a gradient magnitude for each pixel in the image; and

determining a gradient direction for each pixel in the image, the gradient direction being determined using a look up table.

2. (canceled)

8. A method for identifying objects in an image comprising:

determining a gradient magnitude for each pixel in the image;

determining a gradient direction for each pixel in the image, the gradient direction being determined using a look up table.

9. The method of claim 8, wherein the step of determining a gradient direction comprises:

calculating a first and second gradient vector for each pixel;

determining whether the first and second gradient vectors are positive integers; and

determining whether an absolute value of the first gradient vector is greater than an absolute value of the second gradient vector.

10. The method of claim 9, further comprising:

setting a first binary value of a first binary word to equal one if the first gradient vector is a positive integer;

setting a second binary value of a second binary word equal to one if the second gradient vector is a positive integer; and

setting a third binary value of a third binary word equal to one if the absolute value of the first gradient vector is greater than the absolute value of the second gradient vector.

11. The method of claim 10, further comprising:

performing a logic OR operation on the first, second and third binary words;

determining a gradient direction based on a result of the logical OR operation.

12. A computer readable-medium containing a program which executes the steps of:

processing an image with a gradient operator to produce a gradient magnitude and direction for each pixel;

determining a number of different gradient directions in a portion of the processed image; and

identifying the portion of the processed image as an object if the number of different gradient directions exceeds a threshold number of gradient directions.

14. The computer-readable medium of claim 12, wherein each possible gradient direction is associated with a unique binary value, and wherein the gradient direction for each pixel is encoded into one of the unique binary values.

15. The computer-readable medium of claim 14, wherein a binary OR operation of any number of the unique binary values results in a binary value which uniquely identifies a number of different gradient directions employed in the binary OR operation.

16. The computer-readable medium of claim 12, wherein the image is a monochrome image with image intensities on various gray levels.

17. The computer-readable medium of claim 12, wherein the image is processed using a Sobel operator.

18. The computer-readable medium of claim 12, wherein the step of processing the image comprises:

determining a gradient magnitude for each pixel in the image; and

determining a gradient direction for each pixel in the image, the gradient direction being determined using a look up table.

13. (canceled)

19. A computer-readable medium containing a program which executes the steps of:

determining a gradient magnitude for each pixel in an image; and

determining a gradient direction for each pixel in the image, the gradient direction being determined using a look up table.

3. The method of claim 21, wherein each possible gradient direction is associated with a unique binary value, and wherein the gradient direction for each pixel is encoded into one of the unique binary values.

4. The method of claim 3, wherein a binary OR operation of any number of the unique binary values results in a binary value which uniquely identifies a number of different gradient directions employed in the binary OR operation.

20. The computer-readable medium of claim 19, wherein the step of determining a gradient direction comprises:

calculating a first and second gradient vector for each pixel;

determining whether the first and second gradient vectors are positive integers; and

determining whether an absolute value of the first gradient vector is greater than an absolute value of the second gradient vector.

21. A computer-readable medium of claim 20, wherein the computer program performs the steps of:

setting a first binary value of a first binary word to equal one if the first gradient vector is a positive integer;

setting a second binary value of a second binary word equal to one if the second gradient vector is a positive integer; and

setting a third binary value of a third binary word equal to one if the absolute value of the first gradient vector is greater than the absolute value of the second gradient vector.

22. The computer-readable medium of claim 21, wherein the computer program performs the steps of:

performing a logic OR operation on the first, second and third binary words; and

determining a gradient direction based on a result of the logical OR operation.

23. A system comprising:

a processor which processes the image with a gradient operator to produce a gradient magnitude and direction for each pixel, determines a number of different gradient directions in a portion of the processed image, and identifies the portion of the processed image as an object if the number of different gradient directions exceeds a threshold number of gradient directions; and

means for outputting the identified portion of the processed image.

25. The system of claim 23, wherein each possible gradient direction is associated with a unique binary value, and wherein the gradient direction for each pixel is encoded into one of the unique binary values.

26. The system of claim 25, wherein a binary OR operation of any number of the unique binary values results in a binary value which uniquely identifies a number of different gradient directions employed in the binary OR operation.

27. The system of claim 23, wherein the image is a monochrome image with image intensities on various gray levels.

28. The system of claim 23, wherein the image is processed using a Sobel operator.

29. The system of claim 23, wherein the processor processes the image with the gradient operator by determining a gradient magnitude for each pixel in the image, and by determining a gradient direction for each pixel in the image, the gradient direction being determined using a look up table.

24. (canceled)