Study on video segmentation

Pao Yue-kong Library Electronic Theses Database

Study on video segmentation


Author: Man, Yiu-keung
Title: Study on video segmentation
Degree: M.Sc.
Year: 2000
Subject: Videodiscs
Image processing
Hong Kong Polytechnic University -- Dissertations
Department: Multi-disciplinary Studies
Dept. of Electronic and Information Engineering
Pages: [56] leaves : ill. ; 31 cm
Language: English
Abstract: It is well known that object-based coding techniques substantially increase the data compression rate. This is very important; especially in very low bit rate applications like video conferencing. It is a big challenge for transmitting the real time video data under 64k bit/s in a channel. The key issue of an object-based coding is segmentation in which the moving objects are extracted or segmented out from a picture sequence. Segmentation is a classical topic in image processing. It is a process to extract a specific group of pixels of interest and identify them from others. In can be done either in two dimensional case or three dimensional case. Traditionally, block based motion estimation is used in video pictures coding. The scope of this dissertation is to investigate the effect of motion estimation and to apply it to segmentation and background removal. If we divide the pictures of video sequences into a number of blocks, we can locate the moving segments with motion estimation. Once the moving blocks are detected, we split them into a number of small blocks and carry out motion estimation again. The moving segments are more accurately located as the block size decreases. Moreover, for video sequences with moving backgrounds, they usually move in the same direction. By examining the direction of background segments movement, we can locate the moving block with the help of motion estimation. We have been able to remove the background based on this block motion estimation technique. For a comparison purpose, some existing video segmentation algorithms have also been studied.

Files in this item

Files Size Format
b15176861.pdf 3.015Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.


Quick Search


More Information