Dynamic target addressing system

• US 6,236,737 B1
• Filed: 03/24/1998
• Issued: 05/22/2001
• Est. Priority Date: 03/26/1997
• Status: Expired due to Fees

• Current Assignee: Dalhousie University
• Original Assignee: Dalhousie University
• Inventors: Franklin, Steve M., Gregson, Peter H.
• Primary Examiner(s): Mancuso, Joseph
• Assistant Examiner(s): Tabatabai, A
• Application Number:
  US09/046,702
• Time in Patent Office: 1,155 Days
• Field of Search: 382/100, 382/103, 382/106, 382/107, 382/118, 382/153, 382/154, 382/285, 382/291, 382/113, 345/203, 345/433, 345/473, 345/474, 395/518
• US Class Current: 382/103
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments

Abstract

The present invention relates to an improved system and method for target motion tracking, that reduces the computational resources required, by providing for the dynamic control of the activation of targets. The present invention is comprised of a system for use in dynamically tracking the movement of a subject. A plurality of active targets are located on the subject at positions of movement of the subject. At least one camera is arranged for viewing the subject, wherein each of the at least one camera outputs a video signal. A computer receives the video signal of each and processes each video signal by means of an algorithm program to produce a control signal. A control unit is connected to the computer and to the plurality of active targets and receives the control signal and processes it to produce activation signals for selectively activating ones or groups of active targets so that the computer means can efficiently determine the movement of the active targets and thereby determine the movement of the subject.

Claims

1. A system for use in dynamically tracking the movement of a deformable subject comprising:

a plurality of active targets for location on the subject at positions of movement of the subject;

two or more camera means for viewing the subject, each said two or more camera means having as an output a video signal;

a computer means for receiving the video signal of each of said two or more camera means, said computer means processing each said video signal by means of an algorithm program to produce a control signal; and

control means connected to said computer means and to said plurality of active targets, said control means receiving said control signal and processing said control signal to produce activation signals for selectively activating ones or groups of said active targets so that said computer means can determine the movement of said active targets and thereby determine the movement of the subject;

wherein the algorithm program recognizes and selects targets or groups of targets that are moving either relatively quickly with respect to other targets or are relatively close to one another and processes the control signal to illuminate these selected targets or groups of targets more often that those targets not selected; and

wherein each of said two or more camera means is synchronized so that each camera means begins the scan of a frame at the same time.

2. The system according to claim 1, wherein the active targets are near infrared light emitting diodes.

3. The system according to claim 1, wherein the computer means produces said control signal as a serial control signal and a serial I/O interface is located between said computer means and said control means.

4. The system according to claim 1, wherein each of said two or more camera means is a CCD device and each of said two or more camera means has a lens, said lens being covered by a filter means which inhibits the passage of visual light and allows the passage of near infrared energy, so that the video signal output from each of said two or more camera means includes information concerning either the presence of a target or the absence of a target on a frame by frame basis.

5. The system according to claim 1, wherein the algorithm program additionally dynamically selects targets or groups of targets to be illuminated in each frame of the video signal, decided by rules-based criteria based on at least one factor selected from a group of factors including previous occlusion of targets, elapsed time since target last observed, subject kinematic constraints and available computing resources.

6. The system according to claim 1, wherein the algorithm program additionally dynamically selects targets or groups of targets to be illuminated in each frame of the video signal, decided by rules-based criteria based on at least one factor selected from a group of factors including previous occlusions of a target, kinematic constraints of said subject, predicted motion of a target, anticipated correspondence of targets, and available computing power.

7. The system according to claim 1, wherein the illumination of said plurality of active targets is synchronized with the synchronization of said two or more camera means.

8. The system according to claim 7, wherein said targets or groups of targets that are selected are illuminated for every frame of said two or more camera means and said targets or groups of targets not selected are illuminated at a rate that is less but synchronized with the frame rate of said two or more camera means.

9. A method of dynamically tracking the movement of a deformable subject comprising the steps of:

placing an array of active targets on the subject, said targets being located at positions of movement of said subject;

illuminating each of the array of active targets;

observing the illumination of said array of active targets with two or more camera means, each camera means of said two or more camera means providing a video signal;

processing each said video signal by a computer means using an algorithm program to provide a control signal;

processing said control signal with a target illumination control device connected to the computer means and the array of active targets to control the illumination of targets or groups of targets in the array of active targets; and

synchronizing said two or more camera means that each camera means begins the scan of a frame at the same time;

wherein the algorithm program recognizes and selects targets or groups of targets within the array of active targets that are moving either relatively quickly with respect to other targets or are relatively close to one another and processes the control signal to illuminate these selected targets or groups of targets more often that those targets within the array of active targets not selected.

10. The method according to claim 9, wherein the illumination of said plurality of active targets is synchronized with the synchronization of said two or more camera means.

11. The method according to claim 10, wherein said targets or groups of targets that are selected are illuminated for every frame of said two or more camera means and said targets or groups of targets not selected are illuminated at a rate that is less than but synchronized with the frame rate of two or more camera means.

12. The method according to claim 9, wherein the algorithm program additionally dynamically selects targets or groups of targets to be illuminated in each frame of the video signal, decided by rules-based criteria based on at least one factor selected from a group of factors including previous occlusion of targets, elapsed time since target last observed, subject kinematic constrains and available computing resources.

13. The system according to claim 9, wherein the algorithm program additionally dynamically selects targets or groups of targets to be illuminated in each frame of the video signal, decided by rules-based criteria based on at least one factor selected from a group of factors including previous occlusions of a target, kinematic constraints of said subject, predicted motion of a target, anticipated correspondence of targets, and available computing power.