Recording a sequence of images using two recording procedures

• US 8,953,674 B2
• Filed: 12/02/2004
• Issued: 02/10/2015
• Est. Priority Date: 12/03/2003
• Status: Active Grant

• Current Assignee: Envysion Incorporated (Motorola Solutions, Inc.)
• Original Assignee: LightHaus Logic, Inc.
• Inventors: Henson, Michael Anthony, Sherrah, Jamie Roy, Quartermain, Peter Richard
• Primary Examiner(s): LEE, Y YOUNG
• Application Number:
  US11/002,444
• Publication Number:
  US 20050122397A1
• Time in Patent Office: 3,722 Days
• Field of Search: 375/240.01, 375/240.08, 375/240.26, 348/152, 348/243, 382/205
• US Class Current: 375/240.09
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G08B 13/19613 Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
• CPC Class Codes: G08B 13/19656 Network used to communicate with a camera, e.g. WAN, LAN, Internet
• CPC Class Codes: G08B 13/19669 Event triggers storage or change of storage policy
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources
• CPC Class Codes: H04N 7/188 Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position signal generation from motion picture films H04N5/253

Abstract

A sequence of images received from a camera is recorded according to a first procedure. The images are analyzed to identify events of interest, and upon identifying an event of interest, an area of interest of a plurality of said images is identified. The identified area is recorded according to a second procedure.

Claims

1. A method comprising recording a sequence of images received from a camera, including the steps of:

receiving a sequence of images from a camera, wherein each one of said sequence of images comprises a plurality of pixels;

establishing a background model from said sequence of images that defines expected pixel values for each one of said sequence of images;

recording said sequence of images according to a first procedure;

analyzing each one of said sequence of images with reference to said background model in which pixels having expected values are classified as belonging to the background and pixels having unexpected values are classified as belonging to the foreground, to identify a plurality of connected pixels classified as belonging to the foreground that represent an event of interest by measuring a number of pixels in each one of the images that have been classified as belonging to the foreground as a proportion to the total number of pixels in the image to obtain a level of activity for each one of the images, wherein said event of interest occurs in a plurality of said sequence of images;

upon identifying said event of interest, defining the portion of each one of said plurality of said sequence of images that covers the plurality of connected pixels classified as belonging to the foreground that represent said event of interest as an area of interest;

for each of said plurality of said sequence of images, recording only said area of interest according to a second procedure that is different than the first procedure; and

storing images recorded by said first and second procedures in a data store.

2. A method according to claim 1, wherein the detection of an event of interest is brought to the attention of an observer to allow action to be taken in response to an observation made by said observer.

3. A method according to claim 1, wherein said first procedure records images at a first definition and said second procedure records images at a second definition, and said second definition is higher than said first definition.

4. A method according to claim 1, wherein said first procedure compresses image data.

5. A method according to claim 4, wherein said second procedure compresses image data but said first procedure applies a greater degree of compression to images than said second procedure.

6. A method according to claim 5, wherein said first procedure causes data loss.

7. A method according to claim 1, wherein output image data is produced by compositing said areas of interest against respective frames recorded according to said first procedure.

8. A method according to claim 1, wherein said events of interest are identified by detecting new objects that have entered a captured scene, and said areas of interest include image data representing said objects.

9. The method according to claim 1, wherein said second procedure includes placing a bounding box around an object corresponding to said event of interest in said image and creating a sub-image for a first plurality of pixels located within said bounding box having a compression rate that is lower than a compression rate of said first procedure.

10. The method according to claim 9, wherein a second plurality of pixels located outside of said bounding box are recorded according to said first procedure.

11. Image processing apparatus, including a data store, a processor, and an input, wherein said processor is configured to:

receive, via said input, a sequence of images from a camera, wherein each one of said sequence of images comprises a plurality of pixels;

establish a background model from said sequence of images that defines expected pixel values for each one of said sequence of images;

process said sequence of images according to a first procedure and store said images in said data store;

analyze each one of said sequence of images with reference to said background model in which pixels having expected values are classified as belonging to the background and pixels having unexpected values are classified as belonging to the foreground, to identify a plurality of connected pixels classified as belonging to the foreground that represent an event of interest by measuring a number of pixels in each one of the images that have been classified as belonging to the foreground as a proportion to the total number of pixels in the image to obtain a level of activity for each one of the images, wherein said event of interest occurs in a plurality of said sequence of images;

upon identifying an event of interest, define the portion of each one of said plurality of said sequence of images that covers the plurality of connected pixels classified as belonging to the foreground that represent said event of interest as an area of interest; and

for each of said plurality of said sequence of images, process only said area of interest according to a second procedure and store said area of interest in said data store.

12. Image processing apparatus according to claim 11, wherein said first procedure records images at a first definition and said second procedure records images at a second definition, and said second definition is higher than said first definition.

13. A non-transitory computer-readable medium having computer-readable instructions executable by a computer or by a network of computers such that when executing said instructions said computer(s) will perform the steps of:

receiving a sequence of images from a camera, wherein each one of said sequence of images comprises a plurality of pixels;

establishing a background model from each one of said sequence of images that defines expected pixel values for said sequence of images;

recording said a sequence of images according to a first procedure;

analyzing each one of said sequence of images with reference to said background model in which pixels having expected values are classified as belonging to the background and pixels having unexpected values are classified as belonging to the foreground, to identify a plurality of connected pixels classified as belonging to the foreground that represent an event of interest by measuring a number of pixels in each one of the images that have been classified as belonging to the foreground as a proportion to the total number of pixels in the image to obtain a level of activity for each one of the images, wherein said event of interest occurs in a plurality of said sequence of images;

upon identifying said event of interest, defining the portion of each one of said plurality of said sequence of images that covers the plurality of connected pixels classified as belonging to the foreground that represent said event of interest as an area of interest;

for each of said plurality of said sequence of images, recording only said area of interest according to a second procedure; and

storing images recorded by said first and second procedures in a data store.

14. A non-transitory computer-readable medium having computer-readable instructions according to claim 13, such that when executing said instructions a computer will also perform the step of bringing to the attention of an observer the detection of an event of interest to allow action to be taken in response to an observation made by said observer.

15. A non-transitory computer-readable medium having computer-readable instructions according to claim 13, wherein said first procedure records images at a first definition and said second procedure records images at a second definition, and said second definition is higher than said first definition.

16. A non-transitory computer-readable medium having computer-readable instructions according to claim 13, wherein said first procedure compresses image data.

17. A non-transitory computer-readable medium having computer-readable instructions according to claim 16, wherein said second procedure compresses image data but said first procedure applies a greater degree of compression to images than said second procedure.

18. A non-transitory computer-readable medium having computer-readable instructions according to claim 17, wherein said first procedure causes data loss.

19. A non-transitory computer-readable medium having computer-readable instructions according to claim 13, wherein output image data is produced by compositing said areas of interest against respective frames recorded according to said first procedure.

20. A non-transitory computer-readable medium having computer-readable instructions according to claim 13, wherein said events of interest are identified by detecting new objects that have entered a captured scene, and said areas of interest include image data representing said objects.