Vision-based hand gesture recognition

Pao Yue-kong Library Electronic Theses Database

Vision-based hand gesture recognition

 

Author: Zhang, Yang
Title: Vision-based hand gesture recognition
Degree: M.Sc.
Year: 2011
Subject: Hong Kong Polytechnic University -- Dissertations
Human-computer interaction
Gesture
Department: Dept. of Computing
Pages: ix, 87 leaves : ill. ; 30 cm.
InnoPac Record: http://library.polyu.edu.hk/record=b2426872
URI: http://theses.lib.polyu.edu.hk/handle/200/6046
Abstract: As computer systems have been used more and more widely in our daily life, the ability to interact with them without the need for special equipment is very attractive. Vision based Human Computer Interaction (HCI) has the potential of making this possible in a form that is both easy and natural for people to use. However, there are great technical challenges both in the creation of hand image detector and feature extraction. One strategy to overcome these challenges is to create vision algorithms that are more specific to hand gesture recognition, so this thesis develops a series of systematic techniques for hand gesture recognition. This thesis divides the vision-based hand gesture recognition into three steps: dynamic gesture extraction, modeling and identification. This thesis is mainly focused on the first two steps and the development of novel and effective techniques, and the performance of these techniques can be shown through the identification step. Based on the skin-color dataset, a kind of skin-color classifier is developed to enable the detection of hand in video sequences; a lot of tests are carried out to compare this classifier with other popular ones which are the HSV, YCbCr and KL. The combination of temporal difference and this classifier can extract dynamic gesture and then the hand is modeled using a novel method referred to as the boundary statistical sampling. In the identification part, the BP network is used to recognize these models and output the result. According to the recognition rate of almost 90%, these techniques have more potential research value.

Files in this item

Files Size Format
b24268720.pdf 2.914Mb 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

Browse

More Information