Please use this identifier to cite or link to this item:
Title: Facial image processing - eyes extraction and glasses matching
Authors: Tong, Wing Yin
Department: Department of Computer Science
Issue Date: 2008
Supervisor: Supervisor: Dr Wong, Helena Tsui Fong ; First Reader: Dr. Ngo, Chong Wah; Second Reader: Prof. Ip, Horace Ho Shing
Abstract: Eye detection is not new to computer industry. It has been widely used in applications, for example red eye detection, eye graze detection, eye recognition, human-computer interaction system etc. Besides, it is a critical task of face detection as well. This project provides an alternative approach to develop an eye detection algorithm that is built on feature based eye detection methods. The new algorithm uses the image‟s pixel color as the only information for eye tracking. It is straightforward and does not contain any heavy processing procedures such as template matching or the implementation of crummy training data. Moreover, the project brings the suggested eye detection algorithm to life application - Eyewear Matching System. This system is to show how the algorithm can help to locate the eyes and merge an eyewear photo to the eyes position. As merging two images together may cause aliasing, merging techniques – erosion and Gaussian smoothing are included in this project.
Appears in Collections:Computer Science - Undergraduate Final Year Projects

Files in This Item:
File SizeFormat 
fulltext.html146 BHTMLView/Open

Items in Digital CityU Collections are protected by copyright, with all rights reserved, unless otherwise indicated.