Abstract

Tracking and surveillance methods and systems for monitoring objects passing in front of non-overlapping cameras. Invention finds corresponding tracks from different cameras and works out which object passing in front of the camera(s) made the tracks, in order to track the object from camera to camera. The invention uses an algorithm to learn inter-camera spatial temporal probability using Parzen windows, learns inter-camera appearance probabilities using distribution of Bhattacharyya distances between appearance models, establishes correspondences based on Maximum A Posteriori (MAP) framework combining both spatial temporal and appearance probabilities, and updates learned probabilities throughout the lifetime of the system.

Document Type

Patent

Patent Number

US 7,450,735

Application Serial Number

10/966,769

Issue Date

10-11-2008

Current Assignee

UCFRF

Assignee at Issuance

UCFRF

College

College of Engineering and Computer Science (CECS)

Department

Computer Science

Allowance Date

6-26-2008

Filing Date

10-16-2004

Assignee at Filing

UCFRF

Filing Type

Nonprovisional Application Record

Donated

no

Download Patent

Share

COinS