A new EOG-based eyeblink detection algorithm

Accurate and efficient operator functional state classification and assessm ent based on physiological data have many important applications ranging fr om operator monitoring to interaction and control of human/machine systems. Eyeblink characteristics are frequently used as physiological indicators f or this purpose. In this paper, we describe an efficient and robust eyeblin k detection algorithm based on, nonlinear analysis of the electrooculogram (EOG) signal. The performance of the algorithm was evaluated via data analy sis results of several benchmark test sets in comparison with another eyebl ink detection algorithm.