Date of Award

Winter 2014

Project Type

Thesis

Program or Major

Electrical and Computer Engineering

Degree Name

Master of Science

First Advisor

Richard Messner

Second Advisor

Andrew Kun

Third Advisor

Wayne Smith

Abstract

Video Copy Detection is the process of comparing two videos to determine their similarity and determine if they are copies. This thesis enhances some of the common algorithms used in Video Copy Detection by utilizing the Log-Polar transformation as a pre-processing step. This pre-processing step is expected to increase speed of the overall Video Copy Detection process while maintaining the accuracy of the algorithms. The results of this research show that the addition of a Log-Polar pre-processing step reduces the computation time of the overall Video Copy Detection process. The additional time necessary to perform the Log-Polar pre-processing step is outweighed by the overall reduction in computation time. The accuracy and recall are slightly affected by the addition of this pre-processing step. The results also show that the video frame size can be significantly compressed with minimal effect to the algorithm's overall performance.

Download

Share

COinS