Design and Implementation of a Reading Auxiliary Apparatus Based on Electrooculography

Eye movements have been proven the most frequent of all human activities; therefore, research on a relationship between different eye movement patterns become a hotspot in human-computer interface fields. The motivation of this paper is to develop a reading auxiliary apparatus by measuring and analyzing the electrooculography signals. We first describe the saccade detection algorithm based on the wavelet packet decomposition and the derivation blink detection algorithm. Furthermore, consecutive blinks were used to control the system's working state and a magnifier, whose position is adjusted according to the results of saccade detection. Experiential results on six participants show that recognition accuracy ratio (F1 score) is 90.096%, which reveal that the proposed system has a good recognition performance on reading activity detection and analysis.