Abstract
Biosignals will play an important role in building communication between machines and humans. One of the types of biosignals that is widely used in neuroscience are electrooculography (EOG) signals. An EOG has a linear relationship with eye movement displacement. Experiments were performed to construct a gaze motion tracking method indicated by robot manipulator movements. Three operators looked at 24 target points displayed on a monitor that was 40 cm in front of them. Two channels (Ch1 and Ch2) produced EOG signals for every single eye movement. These signals were converted to pixel units by using the linear relationship between EOG signals and gaze motion distances. The conversion outcomes were actual pixel locations. An affine transform method is proposed to determine the shift of actual pixels to target pixels. This method consisted of sequences of five geometry processes, which are translation-1, rotation, translation-2, shear and dilatation. The accuracy was approximately 0.86° ± 0.67° in the horizontal direction and 0.54° ± 0.34° in the vertical. This system successfully tracked the gaze motions not only in direction, but also in distance. Using this system, three operators could operate a robot manipulator to point at some targets. This result shows that the method is reliable in building communication between humans and machines using EOGs.
Highlights
The electrooculography (EOG) signal is a bio-signal that measures eye activities
Many methods can learn the phenomena of eye movement, the potential difference [1] is the most broadly used by neuroscientists to investigate eye movements [2]
This paper proposes an affine transform method to build a gaze motion tracking system
Summary
The electrooculography (EOG) signal is a bio-signal that measures eye activities. Those activities generate a potential difference between the cornea and the retina. Barea et al [6] proposed a system to identify horizontal gaze motions based on a neural network. A low-cost computer interface based on the EOG was proposed by [10] This interface used the polarity of signals to differentiate the gaze motion direction on each channel. The peak amplitude and the slope of the signals were used to categorize the blink, eye movement and noise This method successfully recognized right, left, up and down gaze motion and blinks. A homogeneous affine transform was constructed by sequences of five geometry processes, which are translation-1, rotation, translation-2, shear and dilation This method was designed to detect the direction and the distance of gaze motions. A robot manipulator was used as the indicator of gaze motions to some target points
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
