UUM ETD | Universiti Utara Malaysian Electronic Theses and Dissertation
FAQs | Feedback | Search Tips | Sitemap

Diagnosing Affected Organs Using Automated Iridology System

Albusaidi, Hilal Nasser (2009) Diagnosing Affected Organs Using Automated Iridology System. Masters thesis, Universiti Utara Malaysia.

[img] PDF
Hilal_Nasser_Albusaidi.pdf
Restricted to Registered users only

Download (1MB)
[img]
Preview
PDF
1.Hilal_Nasser_Albusaidi.pdf

Download (205kB) | Preview

Abstract

Iridology is the study of the iris of the eye for medical purposes. It is a preventive medicine since it can warn a person's tendency towards an apparent disease. Cleansing and healing of the body can be verified from changes in the iris. This study aims to design and develop an iris recognition system for automating iridology. The project involves 3 main steps: applying image processing techniques on eye image for data acquisition, collecting the database which is necessary for the iridology analysis and recognizing the affected organ in the body through iris analysis by using neural networks techniques. The image processing techniques are utilized for extracting eye images. A chart of right and left eyes has been acquired through the Internet and approved by an iridologist: Then, the extracted iris image is compared to the chart to determine the affected organ. Neural network with Back propagation is used to match the iris images with affected organ. A total of 159 images retrieved from internet was preprocessed and fed into NN engine. The Backprobagation network succeeded and getting best results because it attained to 96.2 % correction percentage.

Item Type: Thesis (Masters)
Uncontrolled Keywords: System Design, System Development, Iris Recognition System, Automating Iridology
Subjects: Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions: College of Arts and Sciences (CAS)
Depositing User: Mr Husni Ismail
Date Deposited: 31 Mar 2010 02:09
Last Modified: 28 Apr 2016 01:44
URI: http://etd.uum.edu.my/id/eprint/1643

Actions (login required)

View Item View Item