Human Spontaneous Emotion Detection System

Radin Monawir, Radin Puteri Hazimah (2018) Human Spontaneous Emotion Detection System. Masters thesis, UTeM.

	Text (24 Pages) Human Spontaneous Emotion Detection System.pdf - Submitted Version Download (260kB)
	Text (Full Text) Human Spontaneous Emotion Detection System.pdf - Submitted Version Restricted to Registered users only Download (1MB)

Abstract

Having smart computerized system which can understand and instantly gives appropriate response to human is the utmost motive in human and computer interaction (HCI) field.It is argued either HCI is considered advance if human could not have natural and comfortable interaction like human to human interaction.Besides,despite of several studies regarding emotion detection system, current system mostly tested in laboratory environment and using mimic emotion.Realizing the current system research lack of real life or genuine emotion input,this research work comes up with the idea of developing a system that able to recognize human emotion through facial expression.Therefore,the aims of this study are threefold which are to enhance the algorithm to detect spontaneous emotion,to develop spontaneous facial expression database and to verify the algorithm performance.This project used Matlab programming language,specifically Viola Jones method for features tracking and extraction,then pattern matching for emotion classification purpose.Mouth feature is used as main features to identify the emotion of the expression.For verification purpose,the mimic and spontaneous database which are obtained from internet,open source database or novel (own) developed databases are used.Basically,the performance of the system is indicated by emotion detection rate and average execution time.At the end of this study,it is found that this system is suitable for recognizing spontaneous facial expression (63.28%) compared to posed facial expression (51.46%).The verification even better for positive emotion with 71.02% detection rate compared to 48.09% for negative emotion detection rate.Finally,overall detection rate of 61.20% is considered good since this system can execute result within 3s and use spontaneous input data which known as highly susceptible to noise.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Computer vision,Human-machine systems, Image processing, Digital techniques, Human Spontaneous, Emotion Detection System
Subjects:	T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Library > Tesis > FKP
Depositing User:	Mohd. Nazir Taib
Date Deposited:	27 Aug 2019 03:59
Last Modified:	21 Feb 2022 11:36
URI:	http://eprints.utem.edu.my/id/eprint/23332
Statistic Details:	View Download Statistic

Actions (login required)

View Item