SYSTEM AND METHOD FOR AUTOMATIC CAMERA HEALTH MONITORING

• US 20070126869A1
• Filed: 12/06/2006
• Published: 06/07/2007
• Est. Priority Date: 12/06/2005
• Status: Active Grant

• Current Assignee: March Networks Corporation (Infinova (Canada) Ltd.)
• Original Assignee: March Networks Corporation (Infinova (Canada) Ltd.)
• Inventors: SUN, Man-Yee, MONTMINY, Christian, MAUGHAN, Kenneth, MACLEAN, Roger
• Granted Patent: US 7,683,934 B2
• Time in Patent Office: Days
• US Class Current: 348/143
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G08B 13/196 using television cameras
• CPC Class Codes: G08B 29/04 Monitoring of the detection circuits
• CPC Class Codes: G08B 29/26 by updating and storing reference thresholds
• CPC Class Codes: H04N 17/002 for television cameras
• CPC Class Codes: H04N 23/60 Control of cameras or camera modules
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A system and method are provided for automatic camera health monitoring, such as for a camera in a video surveillance system. The system preferably provides substantially continuous monitoring and detection of camera malfunction, due to either external or internal conditions. Camera malfunction is detected when a computed camera health measurement exceeds a malfunction threshold. The camera health measurement is computed based on a comparison of a current camera health record to a plurality of stored camera health records obtained in a learning mode, which characterize known states of normal camera operation. Aside from monitoring camera health, methods of the present invention can also be used to detect an object being added to or removed from a scene, or to detect a change in lighting in a scene, possibly caused by a defective light fixture.

Claims

1. A method of detecting camera malfunction in a digital video surveillance system, the method comprising:

generating a current camera health record based on current image data;

computing a camera health measurement based on a comparison of the current camera health record to a plurality of stored camera health records characterizing known states of normal camera operation; and

detecting a camera malfunction when the camera health measurement exceeds a malfunction threshold.

2. The method of claim 1 wherein the malfunction threshold is exceeded when the camera health measurement exceeds a malfunction severity threshold.

3. The method of claim 1 further comprising incrementing a malfunction counter value when the current camera health record is different from each of the plurality of stored camera health records.

4. The method of claim 3 wherein the malfunction threshold is exceeded when the malfunction counter value exceeds a malfunction persistence threshold.

5. The method of claim 1 further comprising sending a malfunction indication to a user after the camera malfunction has been detected, the malfunction indication including malfunction information.

6. The method of claim 1 wherein the stored camera health records are based on sets of stored image data obtained in a learning mode.

7. The method of claim 6 further comprising, in the learning mode:

obtaining a current learning camera health record based on current image data in the learning mode;

computing a learning camera health measurement based on a comparison of the current learning camera health record to a plurality of stored learning camera health records; and

storing the current learning camera health record as part of the stored learning camera health records when the learning camera health measurement exceeds a similarity threshold.

8. The method of claim 7 further comprising incrementing a match counter for a selected stored learning camera health record of the plurality of stored learning camera health records when the current learning camera health record is substantially similar to the selected stored learning camera health record.

9. The method of claim 8 further comprising retaining the stored learning camera health records having the highest match counter values as the stored camera health records.

10. The method of claim 1 further comprising performing the steps of obtaining, comparing, computing and detecting for a plurality of cameras and with respect to a plurality of camera-specific health records for each camera.

11. The method of claim 1 wherein the step of generating the current camera health record comprises generating an average frame by computing an average luminance of the image data obtained during a reference data generation period.

12. The method of claim 11 wherein the step of generating the average frame comprises computing an average chrominance of the video frames obtained during the reference data generation period.

13. The method of claim 11 wherein the step of computing the camera health measurement comprises computing a number of different pixels by comparing pixels in a current average frame to pixels in a stored average frame, a pixel being determined to be different when an absolute difference of the pixel at the same position in the current average frame and the stored average frame exceeds a pixel comparison threshold.

14. The method of claim 13 wherein the step of detecting the camera malfunction is performed when a percentage of different pixels exceeds a pixel difference percentage threshold.

15. The method of claim 11 wherein the step of generating the current camera health record comprises generating an average energy value by calculating a sum of square of all pixels in the average frame.

16. The method of claim 15 wherein the step of detecting the camera malfunction is performed when a comparison of the average energy value of a current average frame to average energy values of a plurality of stored average frames yields a percentage of different average energy values that exceeds an average energy percentage difference threshold.

17. The method of claim 1 wherein the step of generating the current camera health record comprises generating an anchor region by determining a region having a uniquely identifiable energy of frequency components within an average frame.

18. The method of claim 17 wherein generating the anchor region comprises:

computing an average of all pixels in a candidate anchor region, the candidate anchor region being located in a spatial region;

subtracting the average from each pixel value to obtain a pixel error for each pixel in the candidate anchor region;

taking a sum of square values of all pixel errors in the candidate anchor region to obtain a sum of squared errors (SSE); and

assigning the candidate anchor region to be the anchor region when the SSE is a spatial region maximum SSE and is among the highest SSEs in the average frame.

19. The method of claim 17 wherein the step of computing the camera health measurement comprises:

calculating, for each possible position of an anchor-region sized window in a search area, a sum of square errors between reference image pixels in an anchor region and current image pixels in the position in the search area; and

storing each sum of square errors in an entry in an anchor region correlation array.

20. The method of claim 19 wherein the step of detecting the camera malfunction is performed when the smallest value of the anchor region correlation array expressed as a percentage of a sum of all elements in the array exceeds a correlation threshold.

21. The method of claim 1 wherein the step of generating the current camera health record comprises generating an image focus level by performing an orthogonal transform on each frame and summing a square value of all non-DC components of the transformed signal.

22. The method of claim 21 wherein the step of computing the camera health measurement comprises computing an average image focus level for frames acquired during a reference data generation period.

23. The method of claim 22 wherein the step of detecting the camera malfunction is performed when a comparison of the generated image focus level to the average image focus level exceeds an image focus threshold.

24. A system for automatic camera health monitoring comprising:

a camera health record generator to generate a camera health record characterizing camera operation;

a camera health record comparator to compare two camera health records;

a camera health record manager to store a plurality of camera health records and arbitrate access to the stored camera health records; and

a camera monitor in communication with the camera health record generator, the camera health record comparator and the camera health record manager, to initiate a comparison of a current camera health record with a stored camera health record and to detect a camera malfunction when a camera health measurement exceeds a malfunction threshold.

25. The system of claim 24 wherein the camera monitor computes the camera health measurement based on the comparison of the current camera health record with the stored camera health record.

26. The system of claim 24 further comprising:

a user interface in communication with the camera monitor to receive a camera malfunction indication in response to detection of the camera malfunction.

27. The system of claim 24 wherein the camera health record manager comprises a camera health record memory to store the plurality of camera health records.

28. The system of claim 24 wherein the camera health manager comprises a camera health profile memory to store a plurality of camera health profiles, each camera health profile including a plurality of camera health records.