Abstract

The Video Exploitation and Novelty Understanding in Streams (VENUS) system is a complete software solution for video surveillance, which consists of motion detection, motion tracking, novel event detection, missed expected event detection, object recognition, scene description, and database mining. This research focuses on the assignment of degrees of interest for events in motion and at rest for the VENUS system. The Interesting Video Event Extraction (IVEE) system is the second module in the processing pipeline of the VENUS system. The novel features of the IVEE sysem include the ability to assign a degree of interest to an event, to develop a representation of the weekly activities from the input video stream, and to detect when an expected event did not occur. The IVEE system maintains independence through self-learning and without the aid of human intervention to understand the difference between normal and abnormal behaviors.

Library of Congress Subject Headings

Computer vision; Motion perception (Vision)--Computer simulation; Video surveillance; Video recordings--Data processing; Optical pattern recognition; Machine learning

Publication Date

2007

Document Type

Thesis

Student Type

Graduate

Degree Name

Computer Science (MS)

Department, Program, or Center

Computer Science (GCCIS)

Advisor

Gaborski, Roger

Advisor/Committee Member

Reynolds, Carl

Advisor/Committee Member

Butler, Zack

Comments

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works in December 2013.

Campus

RIT – Main Campus

Download

Share

COinS