Object tracking system

• US 7,720,257 B2
• Filed: 06/16/2005
• Issued: 05/18/2010
• Est. Priority Date: 06/16/2005
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Johnson, Andrew, Bazakos, Michael E., Ma, Yunqian, Morellas, Vassilios
• Primary Examiner(s): Seth; Manav
• Application Number:
  US11/154,839
• Publication Number:
  US 20060285723A1
• Time in Patent Office: 1,797 Days
• Field of Search: 382/100, 382/103, 382115-118, 382/154, 348/42, 348/94, 348/143, 348/149, 348152-156
• US Class Current: 382/103
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06T 7/254 involving subtraction of images
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• CPC Class Codes: G06V 10/24 Aligning, centring, orientation detection or correction of the image
• 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: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition

Abstract

A system for tracking objects across an area having a network of cameras with overlapping and non-overlapping fields of view. The system may use a combination of color, shape, texture and/or multi-resolution histograms for object representation or target modeling for the tacking of an object from one camera to another. The system may include user and output interfacing.

Claims

1. A method for tracking an object using sequential Monte Carlo (SMS) methods, the method comprising:

initializing tracking;

selecting a camera have a field of view;

selecting an object to be tracked from the camera field of view;

obtaining a target model of the tracked object;

generating a first particle of the target model;

predicting a next time'"'"'s particle by a dynamic model; and

calculating features of the first particle.

2. The method of claim 1 further comprising:

forming candidates from the features;

matching the target model with the candidates;

updating the particle; and

updating the target model.

3. The method of claim 2, wherein the updating the target model comprises reporting tracking based on a location of the particles.

4. The method of claim 2, further comprising:

changing to a second camera from a first camera;

transferring the target model of the tracked object from the first camera to the second camera;

generating a particle;

predicting a next particle by the target model; and

calculating features of the particle.

5. The method of claim 4, wherein the field of view does not overlap with the field of view of the first camera.

6. The method of claim 4, further comprising:

forming candidates from the features;

matching the target model with the candidates;

updating the particle;

reporting tracking based on a location of the particle; and

updating the target model.

7. The method of claim 6, wherein the field of view does not overlap with the field of view of the first camera.

8. The method of claim 1, wherein a feature of the particle is determined by:

computing histograms of various level resolution images of the target model;

normalizing the histograms;

forming a cumulative histogram from the histograms of the various level resolution images;

computing difference histograms between the histograms of consecutive various level resolution images; and

concatenating the difference histograms to determine the feature of the particle.

9. A tracking system comprising:

a plurality of cameras;

a camera selector connected to the plurality of cameras; and

an image processor connected to the plurality of cameras and the camera selector;

wherein the image processor comprises;

a target model generator; and

a particle generator connected to the target model generator.

10. The system of claim 9, wherein:

each camera of the plurality of cameras is situated at a location of an area; and

the location of each camera is different from the location of another camera of the plurality of cameras.

11. The system of claim 10, wherein some of the cameras of the plurality of cameras have non-overlapping fields of view.

12. The system of claim 9, wherein the particle generator comprises a particle feature generator.

13. The system of claim 12, wherein the image processor further comprises:

a candidate generator connected to the particle generator; and

a matcher connected to the candidate generator and the target model generator.

14. The system of claim 13, wherein the particle feature generator comprises a histogram generator.