Method and apparatus for tracking an object

• US 20050031165A1
• Filed: 08/08/2003
• Published: 02/10/2005
• Est. Priority Date: 08/08/2003
• Status: Active Grant

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Inventors: Olson, Teresa Lorae Pace, Lee, Harry, Stanfill, Robert, Cannon, Jamie, Slaski, James Joseph, Sefcik, Jason
• Granted Patent: US 7,657,059 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51

Abstract

A method and apparatus for tracking an object in an image. An image of a scene comprising an object is received. The image is processed with a first tracker to produce a first estimate of a position of the object in the image and a first confidence value. The image is processed with a second tracker to produce a second estimate of the position of the object in the image and a second confidence value. The second tracker processes the image concurrently with the first tracker. An estimate of the position of the object in the image is selected using the first and second confidence values. The object is then tracked using the selected estimate.

Claims

1. A method of tracking an object comprising:

receiving an image of a scene, the scene comprising the object;

processing the image with a first tracker to produce a first estimate of a position of the object in the image and a first confidence value;

processing the image with a second tracker to produce a second estimate of the position of the object in the image and a second confidence value, the second tracker processing the image concurrently with the first tracker;

selecting an estimate of the position of the object in the image using the first and second confidence values; and

tracking the object using the selected estimate.

2. The method of claim 1 further comprising:

processing the image to produce a third estimate of the position of the object in the image, wherein the third estimate is selected when the first and second confidence values are below a predetermined threshold.

3. The method of claim 1, further comprising:

segmenting the image to produce a fourth estimate of the position of the object in the image and an estimate of a size of the object in the image; and

providing the fourth estimate of the position and the estimate of the size to the first or second tracker respectively based upon the first and second confidence values.

4. The method of claim 3, wherein the fourth estimate is used by the first or second tracker to reinitialize its tracking of the object in the image.

5. The method of claim 1, wherein the step of tracking the object comprises:

providing the selected estimate to a control system; and

adjusting parameters of the control system based on the selected estimate.

6. The method of claim 1, wherein the second tracker processes the image using a first, second or third processing technique based upon a distance between the object and a device comprising the first and second trackers.

7. The method of claim 6, wherein the first processing technique is an unresolved target tracking technique.

8. The method of claim 7, wherein the unresolved tracking technique tracks a single pixel in the image.

9. The method of claim 6, wherein the second processing technique is frame-to-frame correlation technique.

10. The method of claim 6, wherein the third processing technique is a reference template-to-search region correlation process employing a gradient magnitude image.

11. The method of claim 1, wherein the first tracker tracks the object based upon components extracted from the image.

12. The method of claim 11, wherein components are groups of pixels that have a same gradient direction and are connected along horizontal or vertical pixel edges.

13. A system comprising:

a primary tracker which receives an image and outputs a first estimate of a position of an object in the image and an associated first confidence value;

a secondary tracker which receives the image and outputs a second estimate of a position of the object in the image and an associated second confidence value, the secondary tracker processing the image concurrently with the primary tracker;

an output selector which selects an estimate of the position of the object in the image using the first and second confidence values.

14. The system of claim 13, further comprising:

a third tracker which receives the image and outputs a third estimate of the position of the object, wherein the output selector selects the third estimate which the first and second confidence values are below a predetermined threshold.

15. The system of claim 13, further comprising:

an object selector which receives the image and outputs a fourth estimated position of the object in the image, a size of the object in the image, and a fourth confidence value, wherein the fourth estimate of the position and the size of the object are provided to the primary or secondary tracker based on the first and second confidence values, respectively.

16. The system of claim 13, wherein the secondary tracker comprises:

a first secondary tracker;

a second secondary tracker; and

a third secondary tracker, wherein whether the first, second or third secondary tracker provides the second estimate is based on a distance between the system and the object.

17. The system of claim 16, wherein the fist secondary tracker is an unresolved target tracker which tracks a single pixel in the image.

18. The system of claim 16, wherein the second secondary tracker is a frame-to-frame correlation tracker.

19. The system of claim 16, wherein the third secondary tracker is a reference template-to-search region correlation tracker which employs a gradient magnitude image.

20. The system of claim 13, wherein the primary tracker tracks components, wherein components are groups of pixels that have a same gradient direction and are connected along horizontal or vertical pixel edges.

21. A system comprising:

a primary tracker which receives an image and outputs a first estimate of a position of an object in the image and an associated first confidence value;

a secondary tracker which receives the image and outputs a second estimate of a position of the object in the image and an associated second confidence value;

an output selector which selects an estimate of the position of the object in the image using the first and second confidence values when the first and second confidence values are above a predetermined threshold; and

a third tracker which receives the image an outputs a third estimate of a position of the object in the image, wherein the output selector selects the third estimate when the first and second confidence values are below the predetermined threshold.

22. The system of claim 21, further comprising:

an object selector which receives the image and outputs a fourth estimated position of the object in the image and size of the object in the image wherein when the first or second confidence values are below the predetermined threshold for a predetermined amount of time the fourth estimate of the position and the size of the object are provided to the primary or secondary tracker based on the first and second confidence values, respectively.

23. The system of claim 21, wherein the secondary tracker comprises:

a first secondary tracker;

a second secondary tracker; and

a third secondary tracker, wherein whether the first, second or third secondary tracker provides the second estimate is based on a distance between the system and the object.

24. The system of claim 23, wherein the fist secondary tracker is an unresolved target tracker which tracks a single pixel in the image.

25. The system of claim 23, wherein the second secondary tracker is a frame-to-frame correlation tracker.

26. The system of claim 23, wherein the third secondary tracker is a reference template-to-search region correlation tracker which employs a gradient magnitude image.

27. The system of claim 21, wherein the primary tracker tracks components, wherein components are groups of pixels that have a same gradient direction and are connected along horizontal or vertical pixel edges.