By Topic

Detecting eye movement direction from stimulated Electro-oculogram by intelligent algorithms

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Anwesha Banerjee ; Sch. of Biosci. & Eng., Jadavpur Univ., Kolkata, India ; Amit Konar ; D. N. Tibarewala ; R. Janarthanan

To improve the quality of life of many physically challenged people Human computer interfacing is an emerging alternative. Human computer interface such as intelligent rehabilitation aids can be controlled by eye movements. It can be helpful for severely paralyzed people. Electro-oculography is a simple method to track eye movements. Electro-oculogram (EOG) is the biopotential produced in the surrounding region of eye due to eye ball movements. The signal is easy to acquire using surface electrodes placed around the eye. This paper presents a comparative study of different methods for Electro-oculogram classification to utilize it to control rehabilitation aids. In this experiment, Electro-oculogram is acquired with a designed data acquisition system and the wavelet transform coefficients and statistical parameters are extracted as signal features. Those features are used to classify the movements of the eyeball in left and right direction. Classification is done by linear & quadratic discriminant analysis, K-nearest neighbor method, linear support vector machines and artificial neural network with backpropagation algorithm. In comparative study good accuracy (above 75%) has been observed in all cases but KNN showed better performance. Based on these classified signals control commands can be generated for human computer interface.

Published in:

Computing Communication & Networking Technologies (ICCCNT), 2012 Third International Conference on

Date of Conference:

26-28 July 2012