SYSTEM AND METHOD FOR MOTION DETECTION IN A SURVEILLANCE VIDEO

• US 20100290710A1
• Filed: 04/22/2010
• Published: 11/18/2010
• Est. Priority Date: 04/22/2009
• Status: Active Grant

• Current Assignee: CheckVideo, LLC (Kastle Systems International LLC)
• Original Assignee: CheckVideo, LLC (Kastle Systems International LLC)
• Inventors: Gritai, Alexei, Gagvani, Nikhil
• Granted Patent: US 8,571,261 B2
• Time in Patent Office: Days
• US Class Current: 382/224
• CPC Class Codes: G06F 2218/14 by matching peak patterns
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: G06T 7/254 involving subtraction of images
• CPC Class Codes: G06V 20/49 Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes
• CPC Class Codes: H04N 5/91 Television signal processing therefor

Abstract

In some embodiments, a non-transitory processor-readable medium stores code representing instructions to cause a processor to smooth a current image of a scene to produce a smoothed image and subtract pixel values of a background image of the scene from corresponding pixel values of the smoothed image to produce an altitude difference image. Pixel values of the altitude difference image are weighted to produce a weighted difference image. The weighted difference image is convolved to produce a convoluted difference image. A threshold is applied to each pixel of the convoluted difference image to produce a thresholded difference image. Pixels having a value less than the threshold are removed from the thresholded difference image and classified as background pixels. Foreground pixels are determined based on the thresholded difference image.

Claims

1. A non-transitory processor-readable medium storing code representing instructions to cause a processor to:

smooth a current image of a scene to produce a smoothed image;

subtract pixel values of a background image of the scene from corresponding pixel values of the smoothed image to produce an altitude difference image;

weight pixel values of the altitude difference image to produce a weighted difference image;

convolve the weighted difference image to produce a convoluted difference image;

apply a threshold to each pixel of the convoluted difference image to produce a thresholded difference image, pixels having a value less than the threshold being removed from the thresholded difference image and classified as background pixels; and

determine foreground pixels based on the thresholded difference image.

2. The non-transitory processor-readable medium of claim 1, wherein the code representing instructions to cause a processor to weight pixel values of the altitude difference image includes code representing instructions to cause a processor to:

smooth the altitude difference image with a first standard of deviation to produce a first smoothed altitude difference image,

smooth the altitude difference image with a second standard of deviation to produce a second smoothed altitude difference image, and

multiply pixel values of the first smoothed altitude difference image with pixel values of the second smoothed altitude difference image to produce the weighted difference image.

3. The non-transitory processor-readable medium of claim 1, wherein the code representing instructions to cause a processor to smooth the current image includes code representing instructions to cause a processor to smooth the current image using a Gaussian kernel.

4. The non-transitory processor-readable medium of claim 1, wherein the code representing instructions to cause a processor to convolve the weighted difference includes code representing instructions to cause a processor to convolve the weighted difference using a Laplacian of Gaussian kernel.

5. The non-transitory processor-readable medium of claim 1, further comprising code representing instructions to cause a processor to:

update the threshold for pixels classified as background pixels for a predetermined number of successive frames.

6. The non-transitory processor-readable medium of claim 1, wherein the convoluted difference image is a non-binary edge image.

7. The non-transitory processor-readable medium of claim 1, further comprising code representing instructions to cause a processor to:

update a pixel counter associated with each pixel classified as a background pixel; and

update the threshold for each pixel classified as a background pixel having a pixel counter with a value greater than a predetermined counter threshold.

8. The non-transitory processor-readable medium of claim 1, wherein the code representing instructions to cause a processor to weight pixel values includes code representing instructions to cause a processor to weight pixel values using a non-linear function.

9. The non-transitory processor-readable medium of claim 1, wherein the code representing instructions to cause a processor to weight pixel values includes code representing instructions cause a processor to weight pixel values using a combination of one or more non-linear functions and one or more linear function.

10. A non-transitory processor-readable medium storing code representing instructions to cause a processor to:

apply a first threshold from a plurality of first thresholds to each value of a plurality of pixels of an image to produce a first threshold image, each pixel from the plurality of pixels being associated with a first threshold from the plurality of first thresholds, each pixel from the plurality of pixels with a value less than its associated first threshold being removed from the first threshold image;

apply a second threshold from a plurality of second thresholds to each value of the plurality of pixels of the image to produce a second threshold image, each pixel from the plurality of pixels being associated with a second threshold from the plurality of second thresholds, each pixel from the plurality of pixels with a value less than its associated second threshold being removed from the second threshold image;

classify each pixel from the plurality of pixels within both the first threshold image and the second threshold image as an initial foreground pixel; and

classify each pixel from the plurality of pixels in the second threshold image contiguous to an initial foreground pixel as a foreground pixel.

11. The non-transitory processor-readable medium of claim 10, further comprising code representing instructions to cause a processor to:

reset each first threshold from the plurality of first thresholds associated with a foreground pixel or an initial foreground pixel to a default first threshold; and

update each first threshold from the plurality of first thresholds associated with a pixel from the plurality of pixels not classified as a foreground pixel or an initial foreground pixel for a predetermined number of successive frames.

12. The non-transitory processor-readable medium of claim 10, further comprising code representing instructions to cause a processor to:

automatically update each first threshold from the plurality of first thresholds and each second threshold from the plurality of second thresholds based on whether a pixel associated with each first threshold and each second threshold is classified as a foreground pixel or an initial foreground pixel, a brightness of the pixel, and a background difference of each pixel.

13. The non-transitory processor-readable medium of claim 10, further comprising code representing instructions to cause a processor to:

classify each pixel from the plurality of pixels not classified as a foreground pixel or an initial foreground pixel as a background pixel;

update a pixel counter associated with each pixel classified as a background pixel; and

update the first threshold and the second threshold associated with each pixel classified as a background pixel having a pixel counter with a value greater than a predetermined counter threshold.

14. The non-transitory processor-readable medium of claim 10, wherein the image is a convoluted difference image, the code further comprising code representing instructions to cause a processor to:

subtract pixel values of a background image of a scene from corresponding pixel values of a smoothed image of the scene to produce an altitude difference image;

weight each pixel value of the altitude difference image to produce a weighted difference image; and

convolve the weighted difference image to produce the convoluted difference image.

15. A non-transitory processor-readable medium storing code representing instructions to cause a processor to:

classify each pixel from a plurality of pixels associated with an image as a foreground pixel or a background pixel using a plurality of first thresholds and a plurality of second thresholds, each pixel from the plurality of pixels being associated with a first threshold from the plurality of first thresholds and a second threshold from the plurality of second thresholds;

reset a value of a first first threshold from the plurality of first thresholds associated with a first pixel from the plurality of pixels to a default first threshold value, the first pixel being classified as a foreground pixel;

update a value of a second first threshold from the plurality of first thresholds associated with a second pixel from the plurality of pixels, the second pixel being classified as a background pixel, a first counter configured to count the number of successive frames the second pixel is classified as a background pixel having a value greater than a counter threshold; and

incrementing a second counter configured to count the number of successive frames a third pixel from the plurality of pixels is classified as a background pixel, the third pixel being classified as a background pixel, the second counter having a value less than the counter threshold, a third first threshold from the plurality of first thresholds associated with the third pixel maintaining its value.

16. The non-transitory processor-readable medium of claim 15, further comprising code representing instructions to cause a processor to:

reset a value of a second threshold from the plurality of second thresholds associated with the first pixel from the plurality of pixels to a default second threshold value.

17. The non-transitory processor-readable medium of claim 15, further comprising code representing instructions to cause a processor to:

reset a value of a first second threshold from the plurality of second thresholds associated with the first pixel to a default second threshold value; and

update a value of a second second threshold from the plurality of second thresholds associated with the second pixel.

18. The non-transitory processor-readable medium of claim 15, wherein the code representing instructions to cause a processor to update the value of the second first threshold includes code representing instructions to cause a processor to update the value of the second first threshold based on a brightness of the second pixel and a background difference of the second pixel.

19. The non-transitory processor-readable medium of claim 15, wherein the image is a convoluted difference image, the code further comprising code representing instructions to cause a processor to:

subtract pixel values of a background image of a scene from corresponding pixel values of a smoothed image of the scene to produce an altitude difference image;

weight each pixel value of the altitude difference image to produce a weighted difference image; and

convolve the weighted difference image to produce the convoluted difference image.

20. The non-transitory processor-readable medium of claim 15, wherein the value of the first first threshold is greater than a value of a second threshold from the plurality of second thresholds associated with the first pixel.

21. The non-transitory processor-readable medium of claim 15, wherein the default first threshold value is associated with the first pixel, the code further comprising code representing instructions to cause a processor to:

update a default first threshold value associated with the second pixel; and

maintain a default first threshold value associated with the third pixel.