论文信息 - On-line Separation Of Smooth Pursuit And Saccadic Eye Movements

On-line Separation Of Smooth Pursuit And Saccadic Eye Movements

A simple, but efficient, algorithm has been worked out for online analysis of visual tracking signal. This program enables to separate smooth pursuit from saccadic eye movements after estimation of the beginning and the end of the saccades. The used procedure comprises four steps : after having differenciated the eye position signal, a Kalman filter is used, the innovation of which is tested by means of Hinkley stopping rule. The event detected in this way is then recognized as a saccade using a few simple rules.

[1] Jean Merckle,et al. Filtering and Estimation with Antagonist Hypothesis for Failure Detection , 1987 .

[2] Jean-Francols Baland,et al. Algorthms for the Analysis of the Nystagmic Eye Movements Induced by Sinusoidal Head Rotations , 1987, IEEE Transactions on Biomedical Engineering.

[3] Michèle Basseville,et al. Detecting changes in signals and systems - A survey , 1988, Autom..

[4] Kiyotoshi Matsuoka,et al. Detection of rapid phases of eye movements using third-order derivatives , 1983 .

[5] D. Bouis,et al. An accurate and linear infrared oculometer , 1983, Journal of Neuroscience Methods.

[6] Alan S. Willsky,et al. A survey of design methods for failure detection in dynamic systems , 1976, Autom..

[7] A. Terry Bahill,et al. Frequency Limitations and Optimal Step Size for the Two-Point Central Difference Derivative Algorithm with Applications to Human Eye Movement Data , 1983, IEEE Transactions on Biomedical Engineering.