Image processing system and method for improving repeatability

• US 7,839,431 B2
• Filed: 10/19/2006
• Issued: 11/23/2010
• Est. Priority Date: 10/19/2006
• Status: Expired due to Fees

• Current Assignee: Bosch Security Systems Incorporated (Robert Bosch GmbH), Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Swarr, Lonnel J., Sablak, Sezai
• Primary Examiner(s): Ho; Tuan
• Assistant Examiner(s): Cowan; Euel K
• Application Number:
  US11/550,836
• Publication Number:
  US 20080094480A1
• Time in Patent Office: 1,496 Days
• Field of Search: 348/143, 348/144, 348/148, 348/149, 348/154, 348/155, 348/161
• US Class Current: 348/95
• CPC Class Codes: G06T 2207/30244 Camera pose
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: G06T 7/80 Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
• CPC Class Codes: G08B 13/19689 Remote control of cameras, e.g. remote orientation or image zooming control for a PTZ camera

Abstract

Embodiments of the invention relate to a dome camera assembly that is able to ensure repeatability. The dome camera assembly includes a nonvolatile storage mechanism including a stored video image and a corresponding stored camera position. The assembly may further include an image capturing device for capturing a current video image after the camera moves to the stored camera position. An image processing component compares the current video image with the stored video image and determines an offset amount between the current video image and the stored video image. An adjustment mechanism adjusts a current camera position in accordance with the determined offset amount to facilitate repeatability.

Claims

1. A system for improving camera repeatability based on a stored video image and a stored camera position corresponding to the stored video image, the system comprising:

a drive mechanism configured to move the camera to a plurality of different positions during a surveillance operation;

a memory for storing a plurality of video images as a camera moves to the plurality of different positions and for storing a camera position corresponding to each stored video image, the drive mechanism being configured to return the camera to a selected stored camera position;

image processing components for comparing a current video image captured at the selected stored camera position with the stored video image previously captured at the selected stored camera position by using a stationary object in the stored video image and the current video image to determine a position offset; and

an adjustment mechanism for adjusting a current camera position based on the determined position offset to align the current video image with the stored video image.

2. The system of claim 1, wherein the stored video image is a portion of an originally captured video image identified as a region of interest.

3. The system of claim 1, wherein the image processing components compare vertical and horizontal edges of the stationary object in the stored video image with vertical and horizontal edges of the stationary object in the current video image.

4. The system of claim 3, wherein the image processing components determine pan, tilt, and zoom position offsets based on the horizontal and vertical edge comparisons.

5. The system of claim 1, wherein the stored camera position comprises a pan, tilt, and zoom position.

6. The system of claim 1, wherein the adjustment mechanism comprises a processor for adjusting a camera motor.

7. The system of claim 1, further comprising nonvolatile memory for storing the plurality of stored camera positions and the plurality of stored video images.

8. The system of claim 1, wherein the adjustment mechanism aligns the images to an alignment accuracy of one pixel.

9. A camera assembly comprising:

a nonvolatile storage mechanism including a plurality of stored video images and a corresponding stored camera position for each of the plurality of stored video images;

an image capturing device for capturing a current video image upon movement of the camera to a selected one of the stored camera positions;

an image processing component for comparing the current video image with the stored video image previously captured at the selected stored camera position and determining an offset amount between the current video image and the stored video image; and

an adjustment mechanism for adjusting a current camera position in accordance with the determined offset amount to facilitate repeatability.

10. The system of claim 9, further comprising a dome camera assembly having selectable pan, tilt, and zoom positions.

11. The system of claim 9, wherein each stored video image is a portion of an originally captured video image identified as a region of interest.

12. The system of claim 9, wherein the image processing components compare vertical and horizontal edges of the stationary object in the stored video image with vertical and horizontal edges of the stationary object in the current video image.

13. The system of claim 10, wherein the image processing components determine pan, tilt, and zoom position offsets based on the horizontal and vertical edge comparisons.

14. The system of claim 10, wherein each stored camera position comprises a pan, tilt, and zoom position.

15. The system of claim 9, wherein the adjustment mechanism comprises a processor for adjusting a camera motor.

16. The system of claim 9, wherein the adjustment mechanism aligns the images to an alignment accuracy of one pixel.

17. A method for improving camera repeatability, the method comprising:

moving a camera to a plurality of different positions during a surveillance operation;

storing a plurality of video images as a camera moves to the plurality of different positions;

storing a camera position corresponding to each stored video image;

returning the camera to a selected one of the stored camera positions and capturing a current video image at the selected stored camera position;

comparing the current video image to the stored video image corresponding to the selected stored camera position and determining an offset amount; and

adjusting a current camera position in accordance with the offset amount in order to align the current video image with the stored video image.

18. The method of claim 17, further comprising storing the plurality of video images and corresponding camera positions in nonvolatile memory.

19. The method of claim 18, further comprising storing the corresponding camera positions in terms of pan, tilt, and zoom values.

20. The method of claim 18, further comprising storing only a portion of an original video image corresponding to a region of interest of the original video image.

21. The method of claim 17, further comprising returning the camera to another stored camera position upon receiving an operator command.

22. The method of claim 17, wherein comparing the current video image to the stored video image comprises comparing vertical and horizontal edges of a stationary object in the stored video image with vertical and horizontal edges of the stationary object in the current video image.

23. The method of claim 22, wherein comparing the current video image to the stored video image comprises determining pan, tilt, and zoom position offsets based on the horizontal and vertical edge comparisons.

24. The method of claim 17, wherein adjusting the current camera position comprises causing a processor to move a camera motor.

25. The method of claim 17, wherein adjusting the current camera position comprises aligning the images to an alignment accuracy of one pixel.