The development of face recognition based on lip using neural network / Halizawati Mohd Nor

Mohd Nor, Halizawati (2006) The development of face recognition based on lip using neural network / Halizawati Mohd Nor. [Student Project] (Unpublished)

Abstract

Nowadays a lot of applications have been developed especially in face recognition
system. Face recognition became an interesting application in surveillance,
identification system and security system. This project is focusing on the
development of face recognition based on lip using Neural Network Technique. The
objective is to identify the persons' ID based on individual lip from different
expression of facial image. This project is divided into two phases. The first phase is
image pre-processing using MATLAB. The pre-processing is conducted by applying
all fimctions in MATLAB image processing Toolbox. Tthis project used 'imadjust'
and 'histeq' functions for Grayscale Enhancement, 'imfilter' for Linear Filtering and
'im2bw' for Image Segmentation. The second phase is NN application design and
development. Training and testing has been conducted to measuring the performance
of learning by manipulating the parameters to achieve a better performance. It
demonstrates that 60 percent of accuracy which person's lip can be recognized in the
form of normal and smile expressions.

Metadata

Item Type: Student Project
Creators:
Creators
Email / ID Num.
Mohd Nor, Halizawati
UNSPECIFIED
Subjects: Q Science > QA Mathematics > Instruments and machines > Electronic Computers. Computer Science > Neural networks (Computer science)
T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Electronics > Pattern recognition systems
Divisions: Universiti Teknologi MARA, Shah Alam > Faculty of Computer and Mathematical Sciences
Keywords: Face recognition system, Neural networks, Lip
Date: 2006
URI: https://ir.uitm.edu.my/id/eprint/1451
Edit Item
Edit Item

Download

[thumbnail of PPb_HALIZAWATI MOHD. NOR CS 06_5 1.pdf] Text
PPb_HALIZAWATI MOHD. NOR CS 06_5 1.pdf

Download (66kB)

Fulltext

Fulltext is available at:

ID Number

1451

Indexing

|

Statistic

Statistic details