Display options
Cite
Knopp S, Bones P, Weddell S, et al. A wearable device for measuring eye dynamics in real-world conditions. Annu Int Conf IEEE Eng Med Biol Soc. 2013;2013:6615-8doi: 10.1109/EMBC.2013.6611072.
Knopp, S., Bones, P., Weddell, S., Innes, C., & Jones, R. (2013). A wearable device for measuring eye dynamics in real-world conditions. Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference, 20136615-8. https://doi.org/10.1109/EMBC.2013.6611072
Knopp, Simon, et al. "A wearable device for measuring eye dynamics in real-world conditions." Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference vol. 2013 (2013): 6615-8. doi: https://doi.org/10.1109/EMBC.2013.6611072
Knopp S, Bones P, Weddell S, Innes C, Jones R. A wearable device for measuring eye dynamics in real-world conditions. Annu Int Conf IEEE Eng Med Biol Soc. 2013;2013:6615-8. doi: 10.1109/EMBC.2013.6611072. PMID: 24111259.
Copy Download .nbib Format: NLM
Share it on

Annu Int Conf IEEE Eng Med Biol Soc. 2013;2013:6615-8. doi: 10.1109/EMBC.2013.6611072.

A wearable device for measuring eye dynamics in real-world conditions.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference

Simon Knopp, Philip Bones, Stephen Weddell, Carrie Innes, Richard Jones

PMID: 24111259 DOI: 10.1109/EMBC.2013.6611072

Abstract

Drowsiness and lapses of responsiveness have the potential to cause fatalities in many occupations. One subsystem of a prototype device which aims to detect these lapses as they occur is described. A head-mounted camera measures several features of the eye that are known to correlate with drowsiness. The system was tested with eight combinations of eye colour, ambient lighting, and eye glasses to simulate typical real-world input conditions. A task was completed for each set of conditions to simulate a range of eye movement-saccades, tracking, and eye closure. Our image processing software correctly classified 99.3% of video frames as open/closed/partly closed, and the error rate was not affected by the combinations of input conditions. Most errors occurred during eyelid movement. The accuracy of the pupil localisation was also not influenced by input conditions, with the possible exception of one subject's glasses.

MeSH terms

Publication Types