Automated video person tracking system

• US 6,437,819 B1
• Filed: 06/25/1999
• Issued: 08/20/2002
• Est. Priority Date: 06/25/1999
• Status: Expired due to Fees

• Current Assignee: Loveland, Rohan Christopher
• Original Assignee: Loveland, Rohan Christopher
• Inventors: Loveland, Rohan Christopher
• Primary Examiner(s): BRITTON, HOWARD W
• Application Number:
  US09/344,988
• Time in Patent Office: 1,152 Days
• Field of Search: H04/N.718, 348/143, 348/159, 348/169, 348/211, 348/212, 348/214, 348/170
• US Class Current: 348/143
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources
• CPC Class Codes: H04N 7/185 from a mobile camera, e.g. for remote control

Abstract

An automated system for controlling multiple pan/tilt/zoom video cameras in such a way as to allow a person to be initially designated and tracked thereafter as he/she moves through the various camera fields of view. Tracking is initiated either by manual selection of the designated person on the system monitor (82) through the usage of a pointing device (83), or by automated selection of the designated person using software. The computation of the motion control signal is performed on a computer (81) through software using information derived from the cameras connected to the system, and is configured in such a way as to allow the system to pass tracking control from one camera to the next, as the designated person moves from one region to another. The system self-configuration is accomplished by the user'"'"'s performance of a specific procedure involving the movement and tracking of a marker throughout the facility.

Claims

1. A system for automated tracking of persons with pan/tilt/zoom cameras, comprising:

(a) a computer system,

(b) a pointing device connected to said computer system, providing a means of designating a person to be tracked,

(c) a means of digitizing an input camera video signal connected to said computer system,

(d) an interface means allowing a camera motion control signal to be input to said computer system,

(e) a camera multiplexing means allowing a particular set of camera video and motion control signals to be selected from a multiplicity of said signals and passed through to appropriate input and output connections,

(f) an interface means allowing a selector signal and camera motion control signals to be output from said computer system to said camera multiplexing means; and

(g) a software program running on said computer implementing an algorithm providing a means of issuing camera motion control signals to said interface means and multiplexing means in such a manner as to keep the designated person to be tracked centered in the field of view of a pan/tilt/zoom camera.

2. The system of claim 1 wherein said pointing device is a mouse.

3. The system of claim 1 wherein said camera multiplexing means and interface means are combined in a single unit.

4. The system of claim 1 wherein the designation of the person to be tracked is performed automatically by a software program.

5. The system of claim 4 wherein the software program'"'"'s designation of the person to be tracked is based on an automated face recognition of the person to be tracked.

6. The system of claim 1 wherein said software program determines the necessary information about camera positions and fields of view through an automatic self-configuration procedure.

7. The system of claim 6 wherein said automatic self-configuration procedure is facilitated by the movement through the facility of a special, visibly distinct marker.

8. A method for producing camera motion control signals from a computer system in such a manner as to keep a designated person tracked in the view of a camera as that designated person moves about, comprising:

(a) designating the person to be tracked in an image,

(a) segmenting out the designated person being tracked from the rest of the image using motion-based techniques,

(b) determining various characteristics of the segmented out region of the image, including color histogram information, and height, width, and overall area in pixels,

(c) acquiring another image a short time later from the camera,

(d) matching the various characteristics of the designated person with those of all the moving segmented out objects in the new image, in order to find a best match; and

(f) computing the appropriate direction to move the camera, in order to keep the best match object approximately centered in the camera view.

9. The method of claim 8 wherein the step of determining said characteristics of the segmented out region of the image includes height, width, and overall area in pixels.

10. The method of claim 8 wherein the step of determining said characteristics of the segmented out region includes the last position of the region in the overall image, and estimating the next likely position of the region based on the last computed velocity.

11. The method of claim 8 further comprising the step of providing a self-configuration procedure whereby the necessary information about camera positions and fields of view is determined.

12. The method of claim 11 wherein said self-configuration procedure further includes the step of the movement through the facility of a special, visibly distinct marker.

13. The method of claim 8 wherein said step of designation of a person to be tracked further includes the step of using face recognition software.