Method of detecting camera tampering and system thereof

Patent metadata

Field	Value
Publication number	US-9230176-B2
Application number	US-201313796107-A
Country	US
Kind code	B2
Filing date	Mar 12, 2013
Priority date	Mar 14, 2012
Publication date	Jan 5, 2016
Grant date	Jan 5, 2016

Abstract

A method of detecting camera tempering and a system therefor are provided. The method includes: performing at least one of following operations: (i) detecting a size of a foreground in an image, and determining whether a first condition, that the size exceeds a first reference value, is satisfied, (ii) detecting change of a sum of the largest pixel value differences among pixel value differences between adjacent pixels in selected horizontal lines of the image, according to time, and determining whether a second condition, that the change lasts for a predetermined time period, is satisfied, and (iii) adding up a plurality of global motion vectors with respect to a plurality of images, and determining whether a third condition, that a sum of the global motion vectors exceeds a second reference value, is satisfied; and determining occurrence of camera tempering if at least one of the corresponding conditions is satisfied.

First claim

What is claimed is:

1. A method of detecting camera tampering from at least one image captured by at least one camera, the method comprising:
performing at least one of following operations:
(i) detecting a size of a foreground region in an image, and determining whether a first condition, that the size exceeds a first reference value, is satisfied;
(ii) detecting change of a sum of the largest pixel value differences among pixel value differences between adjacent pixels in selected horizontal lines of the image, according to time, and determining whether a second condition, that the change lasts for a predetermined time period, is satisfied; and
(iii) adding up a plurality of global motion vectors with respect to a plurality of images, and determining whether a third condition, that a sum of the plurality of global motion vectors exceeds a second reference value, is satisfied; and

determining that camera tampering has occurred if the first condition is satisfied when operation (i) is performed, if the second condition is satisfied when operation (ii) is performed, and if the third condition is satisfied when operation (iii) is performed.

2. The method of claim 1 , wherein the at least one of the following operations comprises operation (ii), and
wherein operation (ii) comprises:
applying a first weight to an average of sums of the largest pixel value differences in a plurality of images up to a previous image and applying a second weight to a sum of the largest pixel value differences in a current image;
calculating a weighted average of the largest pixel value differences by adding up the first weight applied average and the second weight applied sum; and
determining whether the second condition, that the change lasts for the predetermined time period, is satisfied by using the weighted average.

3. The method of claim 2 , wherein operation (ii) further comprises determining whether the weighted average falls below a third reference value, and
wherein the determining whether the second condition, that the change lasts for the predetermined time period, is satisfied is performed if the weighted average falls below the third reference value.

4. The method of claim 3 , wherein a sum of the first weight and the second weight is one (1).

5. The method of claim 1 , wherein the performing the at least one of the following operations comprises performing operation (iii),
wherein the sum of the global motion vectors comprises at least one of a sum of horizontal directional components of the global motion vectors and a sum of vertical directional components of the global motion vectors.

6. The method of claim 1 , further comprising:
receiving a plurality of images, including the at least one image, of a monitored region captured by a plurality of cameras, including the at least one camera, through a plurality of channels, respectively;
combining the plurality of images into a single image;
pre-processing the single combined image by treating the plurality of images as a single image; and
dividing the pre-processed single combined image into a plurality of pre-processed images,
wherein each of the at least one of the following operations is performed with respect to each of the pre-processed images.

7. The method of claim 6 , wherein the pre-processing comprises at least one of image size conversion, color space conversion and color dimension conversion.

8. The method of claim 7 , wherein the performing the at least one of the following operations comprises performing operation (i), and
wherein the operation (i) further comprises detecting a foreground region of each of the plurality of pre-processed images, on which operation (i) is performed, by detecting a foreground region from the single combined image.

9. The method of claim 6 , wherein the performing the at least one of the following operations comprises performing operations (ii) and (iii).

10. The method of claim 6 , wherein the performing the at least one of the following operations comprises performing operations (i) and (ii), or performing operations (i) and (iii).

11. A camera tampering detection system which detects camera tampering from at least one image captured by at least one camera, the system comprising:
an image analysis unit comprising at least one of the following units:
(i) a foreground ratio detection unit which detects a size of a foreground region in an image, at a given time, and determines whether a first condition, that the size exceeds a first reference value, is satisfied;
(ii) a pixel value difference detection unit which detects change of a sum of the largest pixel value differences among pixel value differences between adjacent pixels in selected horizontal lines of the image, according to time, and determines whether a second condition, that the change lasts for a predetermined time period, is satisfied; and
(iii) a global motion detection unit which adds up a plurality of global motion vectors with respect to a plurality of images, and determines whether a third condition, that a sum of the plurality of global motion vectors exceeds a second reference value, is satisfied,

wherein the image analysis unit determines that camera tampering has occurred if the first condition is satisfied when the image analysis unit comprises the foreground ratio detection unit, if the second condition is satisfied when the image analysis unit comprises the pixel value difference detection unit, and if the third condition is satisfied when the image analysis unit comprises the global motion detection unit.

12. The system of claim 11 , wherein the at least one of the following units comprises the pixel value difference detection unit, and
wherein the pixel value difference detection unit performs:
applying a first weight to an average of sums of the largest pixel value differences in a plurality of images up to a previous image and applying a second weight to a sum of the largest pixel value differences in a current image;
calculating a weighted average of the largest pixel value differences by adding up the first weight applied average and the second weight applied sum; and
determines whether the second condition, that the change lasts for the predetermined time period, is satisfied by using the weighted average.

13. The system of claim 12 , wherein the pixel value difference detection unit further performs determining whether the weighted average falls below a third reference value, and
wherein the pixel value difference detection unit performs the determining whether the second condition, that the change lasts for the predetermined time period, is satisfied if the weighted average falls below the third reference value.

14. The system of claim 13 , wherein a sum of the first weight and the second weight is one (1).

15. The system of claim 11 , wherein the image analysis unit comprises the global motion detection unit,
wherein the sum of the global motion vectors comprises at least one of a sum of horizontal directional components of the global motion vectors and a sum of vertical directional components of the global motion vectors.

16. The system of claim 11 further comprises:
an image input unit which receives a plurality of images, including the at least one image, of a monitored region captured by a plurality of cameras, including the at least one camera, through a plurality of channels, respectively;
an image combination unit which combines the plurality of images into a single image; and
an image unit which performs pre-processing the single combined image by treating the plurality of imag

Assignees

• Hanwha Techwin Co Ltd

Inventors

• Choi Eun-Ji
• Lim Jeong-Eun
• Chang Il-Kwon

Classifications

• G06T2207/20144

  Physics · mapped topic

• G06T7/2006

  Physics · mapped topic

• G06K9/00771 Primary

  Physics · mapped topic

• G06T7/20

  Analysis of motion (motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51) · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title