Description
The human eye is a rich and multifaceted organ; it has proven to be an enigma that has not quite completely understood. The development of the Electrooculography opened a new line of study within engineering discipline due to its potential use as a contributing mechanism for the software and computer applications. The EOG (Electrooculography) method was to establish better to understand the eye and underlying pattern of the eye. In the EOG technique, electrical signals originating from the extra-ocular muscles are measure. In this thesis, a set of six electrodes placed around the eye region and commercially available INTAN 512ch recording controller system used to record the eye movement signal, to build a real-time classification system capable of detecting and classifying eye movements (vertical or horizontal) and blink. The purpose of this study is to develop a new method to detect eye movements from the EOG signals. It was based on the probability map (signature matrix) and the Procrustes method. The Procrustes method is a rigid shape examination that uses isomorphic translation, scaling, rotation, and error to find the “best” fitting among two or more landmarked shapes. The signature matrix provides the mathematical calculation, and a Procrustes method provides the graphical representation of the eye movement signals. The developed technology can find a use for people suffering from motor neuron disorder and have difficulties in mobility and communication during daily life activities.
