Method and system monitoring video image by updating template image

• US 6,108,033 A
• Filed: 05/23/1997
• Issued: 08/22/2000
• Est. Priority Date: 05/31/1996
• Status: Expired due to Term

• Current Assignee: Hitachi Denshi Kabushiki Kaisha
• Original Assignee: Hitachi Denshi Kabushiki Kaisha
• Inventors: Ito, Wataru, Ueda, Hirotada, Okada, Toshimichi
• Primary Examiner(s): LEE, Y YOUNG
• Application Number:
  US08/862,918
• Time in Patent Office: 1,187 Days
• Field of Search: 348/135, 348/169, 348/172, 348/214, 348/152, 382/103, 382/209, 382/216
• US Class Current: 348/152
• CPC Class Codes: G08B 13/19602 Image analysis to detect motion of the intruder, e.g. by frame subtraction
• CPC Class Codes: G08B 13/19604 involving reference image or background adaptation with time to compensate for changing conditions, e.g. reference image update on detection of light level change
• CPC Class Codes: G08B 13/19608 Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
• CPC Class Codes: G08B 13/19691 Signalling events for better perception by user, e.g. indicating alarms by making display brighter, adding text, creating a sound
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

A difference between an input image including images of a background image picked up by an image pick-up unit and an object to be monitored is produced as an image of the object, the image of the object is divided into a plurality of parts and the plurality of divided image of the object are produced as a plurality of templates. Each of the plurality of templates is matched with a new input image and images of a plurality of parts in the new input image which have highest degrees of matching with the plurality of templates are used to update the plurality of templates to output a plurality of new templates. Each of the plurality of new templates is matched with a further new input image and images of a plurality of parts in the further new input image which have highest degrees of matching with the plurality of new templates are used to update the plurality of new templates to output a plurality of further new templates. In this manner, the plurality of templates are sequentially updated for the new input images to trace the object.

Claims

1. A video image monitoring method for monitoring an object to be picked up by image pick-up means, comprising:

a first step for producing a difference between an input image and a background image, in order to obtain the image of said object, said input image including said background image and the image of said object to be monitored from said image pick-up means;

a second step for dividing the image of said object into a plurality of parts to produce a plurality of divided images of said object, each of which is a template;

a third step for matching each of said templates with a new input image and detecting a plurality of parts of said new input image having highest degrees of matching with said templates; and

a fourth step for updating said templates using said parts of said new image having said highest degrees of matching as a plurality of new templates, wherein said fourth step is sequentially executed for new input images to sequentially update said templates to trace said object.

2. A video image monitoring method according to claim 1 wherein said fourth step comprises:

a step for determining a difference of coordinates on a display between said templates and said new templates derived in said fourth step in order to obtain a distance of movement of the object to be monitored; and

a step for adding the distance of said movement to the coordinates of said templates to produce said new templates.

3. A video image monitoring method according to claim 1 wherein said fourth step comprises:

a step for determining a difference of coordinates on a display between said templates and said new templates derived in said fourth step, in order to obtain a distance of movement of said object to be monitored;

a step for determining a ratio of the distances between further new templates in a further new input image and said new templates derived in said fourth step, relative to the distance between said further new templates; and

a step for adding the distance of said movement to the coordinates of said templates to determine the positions of said new templates and updating the sizes of said templates based on said ratio to produce said new templates.

4. A video image monitoring method according to claim 3 further comprising a fifth step for determining the distance between said object to be monitored and said image pick-up means in accordance with said ratio derived in said fourth step.

5. A video image monitoring method according to claim 4 further comprising:

a sixth step for controlling a zoom ratio of said image pick-up means in accordance with the distance between said object to be monitored and said image pick-up means derived in said fifth step.

6. A video image monitoring method according to claim 1 further comprising:

a seventh step for displaying at least one of frames of said new templates updated in said fourth step and the images of said parts of said new image derived in said fourth step on a display of a display device.

7. A video image monitoring system comprising:

a TV camera for picking up an image of at least monitor area;

an image input interface for receiving an image signal from said TV camera;

processing means including a microprocessor and a memory for processing the image signal from said image input interface; and

a display device including a display screen,

wherein said processing means comprises;

first means for producing a difference between an input image from said TV camera and a background image, said input image including said background image from said TV camera and an image of the object to be monitored, in order to obtain the image of said object,

second means for dividing the image of said object into a plurality of parts to produce a plurality of divided images of said object which are a plurality of templates,

third means for matching each of said templates with a new input image from said TV camera and detecting a plurality of parts of said new input image having highest degrees of matching with said templates, and

fourth means for updating said templates using said parts of said new input image having said highest degrees of matching as a plurality of new templates,

wherein said fourth means causes sequential updating of said templates with respect to new input images to conduct tracing of said object.

8. A video image monitoring system according to claim 7 wherein said fourth means comprises:

means for determining a difference of coordinates on said display between said templates and said new templates in order to obtain a distance of movement of said object to be monitored; and

means for adding the distance of said movement to the coordinates of said templates to produce said new templates.

9. A video image monitoring system according to claim 7 wherein said fourth means comprises:

means for determining a difference of coordinates on a display between said templates and said new templates by said fourth means to obtain a distance of movement of said object to be monitored;

means for determining a ratio of the distances between further new templates in a further new input image and said new templates derived by said fourth means, relative to the distance between said further new templates; and

means for adding the distance of said movement to the coordinates of said templates to determine the positions of said new templates and updating the sizes of said templates based on said ratio to produce said new templates.

10. A video image monitoring system according to claim 9 further comprising fifth means for determining the distance between said object to be monitored and said image pick-up means in accordance with said ratio derived by said fourth means.

11. A video image monitoring system according to claim 10 further comprising:

sixth means for controlling a zoom ratio of said image pick-up means in accordance with the distance between said object to be monitored and said image pick-up means derived by said fifth means.

12. A video image monitoring method according to claim 7 further comprising:

seventh means for displaying at least one of frames of said new templates updated by said fourth means and the images of said parts of said new image derived by said fourth means on said display.

13. A video image monitoring method for monitoring an object to be picked up by image pick-up means, comprising:

a step for producing a difference between a background image and an image of said object to be monitored from an image from said image pick-up means;

a step for dividing the image of said object into a plurality of parts to produce a plurality of templates of said object; and

a step for matching each of said templates with a new input image and updating said templates into a plurality of new templates which match with said new input image,

wherein said steps are sequentially executed for new input images to sequentially update said templates to trace said object.