Date of Award

2012

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Electrical Engineering

First Advisor

Graves, Corey Dr.

Abstract

An individual’s emotional, as well as physical state plays a valuable role in their overall productivity. If one has a balance of physical and emotional well being, the likelihood of their performance will increase towards their daily tasks and assignments. One’s nutritional diet plays a vital role in creating and producing balance into their lives. Studies have shown that the foods you eat have a significant effect on your physical, mental and emotional state.The objective of this research was to develop a system that will enable more research for emotion monitoring and feedback with the long term goal of developing a system that improve the emotional health and overall nutritional well being of an individual. The system estimates the emotion of an individual through the EigenFace Expression Recognition approach developed by Md. Iftekhar Tanveer, and then suggests a list of healthy snacks based on their tastes to an individual to improve their mood through the influence of Traditional Chinese Medicine. This research studied five facial expressions: neutral, anger, happy, disgust and sad, which in return expressed the emotional state of the person. The neutral expression represents the reference of the system. The intensity of a particular expression can be identified by its level of dissimilarity from the neutral expression. xiv The Traditional Chinese Medicine Theory uses the Five Element System to treat and diagnose their patients. This system includes earth, wood, fire, water, and metal to create balance in the body and its corresponding taste and emotion. Based on the particular emotion, which is the diagnosis portion, the taste that corresponds to the element represents the treatment for that emotion. A series of trials were run to test the accuracy of the Emotional Health Monitoring System. The highest performing emotion was the anger at a rate of 80% and the lowest performing emotion was sad at 50 %. This caused the system to perform at an overall accuracy rate of 64%.

Download

Share

COinS