Premium
Improvement of Determination Algorithm for Eye Glance Input Interface
Author(s) -
ITO YUTA,
ITAKURA NAOAKI,
MITO KAZUYUKI,
MIZUNO TOTA
Publication year - 2016
Publication title -
electronics and communications in japan
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.131
H-Index - 13
eISSN - 1942-9541
pISSN - 1942-9533
DOI - 10.1002/ecj.11790
Subject(s) - eye movement , computer science , interface (matter) , computer vision , eye tracking , eye tracking on the iss , gesture , artificial intelligence , signal (programming language) , algorithm , contrast (vision) , gaze , gesture recognition , bubble , maximum bubble pressure method , parallel computing , programming language
SUMMARY In our previous studies, an eye control input interface using an electrooculograph (EOG) amplified by ac coupling has been developed. We proposed an eye gesture input interface using a combination of several eye movements. It is unnecessary for eye gesture input to restrict head movement, in contrast to conventional eye gaze input. However, it is necessary to use a start operation before eye gesture input. Therefore, eye glance input has been proposed. Eye glance input uses only a combination of eye movements in opposite directions, so that it does not need start operations. However, eye glance input has not always been successful because the algorithm for detection of eye movement from EOG signals was not adapted to slight shifts of the signal baseline. In this study, we propose a new determination algorithm for the eye glance input interface. In experiments using our improved algorithm, we achieved direction determination success rates of over 80% for six subjects.