System and method for anonymizing content to protect privacy

Patent metadata

Field	Value
Publication number	US-10885606-B2
Application number	US-201916378379-A
Country	US
Kind code	B2
Filing date	Apr 8, 2019
Priority date	Apr 8, 2019
Publication date	Jan 5, 2021
Grant date	Jan 5, 2021

Abstract

A video management system may identify moving objects in a scene, and may obscure the moving object. Additionally, the video management system may identify parts of the scene that are not moving, and may identify pixels having a color falling within a range which may be associated with human skin, and may obscure such pixels. The video management system may present an image to a monitor including the obscured moving object and the obscured pixels having the color falling within the range associated with human skin.

First claim

What is claimed is:

1. A method for displaying security images on a security monitor while maintaining privacy of humans shown in the security images, the security images provided by a video camera, the method comprising:
receiving an image from the video camera;
storing a background image from the video camera;
comparing the image to the background image to identify one or more regions of the image that are different from the background image and one or more regions of the image that are the same as the background image;
identifying one or more objects that are moving in the image as corresponding to those regions of the image that are different from the background image;
obscuring the one or more objects that are identified as moving in the image;
for at least those parts of the image that are not identified as containing any moving objects, identifying pixels that have a color falling within one or more defined color ranges that are associated with human skin;
obscuring pixels that are identified as having a color falling within the one or more defined color ranges;
displaying a resulting image on a display, including any obscured moving objects and any obscured pixels that are identified as having a color falling within the one or more of the defined color ranges; and
repeatedly updating the background image with a new background image so that objects previously moving in the image that then become stationary in the image when the background image is updated with the new background image become part of the new background image.

2. The method of claim 1 , wherein comparing the image to the background image comprises performing a pixel-by-pixel subtraction between the image and the background image.

3. The method of claim 1 , wherein the one or more defined color ranges comprises two or more defined color ranges.

4. The method of claim 3 , wherein each of at least two defined color ranges correspond to a different human race.

5. The method of claim 1 , wherein obscuring of the one or more objects that are identified as moving in the image comprises pixelating the one or more moving objects.

6. The method of claim 1 , wherein obscuring of the one or more objects that are identified as moving in the image comprises fuzz balling the one or more moving objects.

7. The method of claim 1 , wherein obscuring pixels comprises pixelating a group of pixels around each pixel that is identified as having a color falling within the one or more defined color ranges.

8. The method of claim 1 , wherein obscuring pixels comprises applying a fuzz ball that includes the pixels that are identified as having a color falling within the one or more defined color ranges.

9. The method of claim 1 , wherein obscuring pixels comprises setting the pixels that are identified as having a color falling within the one or more defined color ranges to a predetermined color.

10. The method of claim 9 , wherein obscuring pixels comprises adjusting a pixelation block size in accordance with a detected screen size.

11. A method for displaying security images on a security monitor while maintaining privacy of humans shown in the security images, the security images provided by a video camera, the method comprising:
receiving an image from the video camera;
comparing the image to a background image to identify one or more regions of the image that are different from the background image and one or more regions of the image that are the same as the background image;
associating the one or more regions of the image that are different from the background image as one or more foreground regions;
associating the one or more regions of the image that are the same as the background image as one or more background regions;
obscuring the one or more foreground regions;
identifying one or more sub-regions within the one or more background regions that have a color that falls within one or more defined color ranges that are associated with human skin;
obscuring the one or more sub-regions within the one or more background regions that have a color that falls within one or more defined color ranges that are associated with human skin;
displaying a resulting image on a display that includes the obscured one or more foreground regions and the obscured one or more sub-regions within the one or more background regions that have a color that falls within one or more defined color ranges that are associated with human skin; and
repeatedly updating the background image with a new background image so that when one or more regions that differ from the background image remain as the background image is updated with the new background image, those one or more regions become part of the background image.

12. The method of claim 11 , wherein comparing the image to the background image comprises performing a pixel-by-pixel subtraction between the image and the background image.

13. The method of claim 11 , wherein the new background image is a previous one of the images from the video camera.

14. A non-transitory computer-readable medium having instructions stored thereon that when executed by a video management system having a video camera are configured to:
receive an image from the video camera;
store a background image from the video camera;
compare the image to the background image to identify one or more regions of the image that are different from the background image and one or more regions of the image that are the same as the background image;
associate the one or more regions of the image that are different from the background image as one or more foreground regions;
associate the one or more regions of the image that are the same as the background image as one or more background regions;
obscure the one or more foreground regions;
anonymize any human faces in the one or more background regions; and
repeatedly update the background image with a new background image so that when one or more regions that differ from the background image remain as the background image is updated with the new background image, those one or more regions become part of the background image.

15. The non-transitory computer-readable medium of claim 14 , wherein any human faces in the one or more background regions are anonymized by blurring.

16. The non-transitory computer-readable medium of claim 14 , further comprising:
identify human faces in the one or more background regions by identifying regions of color that fall within one or more defined color ranges that are associated with human skin.

Assignees

• Honeywell Int Inc

Inventors

• Eswara Lalitha M
• Jain Abhisekh
• Shaik Ismaiel

Classifications

• G06T11/40 Primary

  Filling planar surfaces by adding surface attributes, e.g. adding colours or textures · CPC title

• G06T2207/30196

  Human being; Person · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title

• G06T7/194

  involving foreground-background segmentation · CPC title

• G06T7/90

  Determination of colour characteristics · CPC title