Tracking The Eyes Using Interdependence Mean Shift Tracking Algorithm With Appropriate Information Provided

• Main Authors: Masrullizam, Mat Ibrahim, Syafeeza, Ahmad Radzi, Soraghan, John
• Format: Article
• Language: English
• Published: Asian Research Publishing Network (ARPN) 2016
• Subjects: T Technology (General); TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utem.edu.my/id/eprint/17693/1/Tracking%20The%20Eyes%20Using%20Interdependence%20Mean%20Shift%20Tracking%20Algorithm%20With%20Appropriate%20Information%20Provided.pdf; http://eprints.utem.edu.my/id/eprint/17693/; http://www.arpnjournals.org/jeas/research_papers/rp_2016/jeas_0616_4528.pdf

The human eye tracking algorithm is very important, especially in the facial analysis application. Eyes represent rich of information that able to indicate the reaction, perception, or physiological status of individual. Most of the developed eyes tracking algorithm are not considered the condition of the eyes that would provide the appropriate information to be used in the processes of the facial analysis algorithm. This paper proposes a new technique in how the eyes are tracked and provide the appropriate information for further process. This technique uses Mean Shift tracking algorithm and interdependence scheme to track the eye and stop the tracking when eyes are out of best position and condition to deliver the appropriate information. This proposed algorithm is tested and evaluated using Strathclyde Facial Fatigue (SFF) video footage database, a collection of video footage from the internet and our own facial recording. Based on the experiment carried out, the proposed technique shows the promising result.