May 2004
Volume 45, Issue 13
Free
ARVO Annual Meeting Abstract  |   May 2004
Effecting improved prosthetic visual acuity via image processing in a fixation–pursuit task
Author Affiliations & Notes
  • L.E. Hallum
    Graduate School of Biomedical Engineering, University of New South Wales, Sydney, Australia
  • G.J. Suaning
    School of Engineering, University of Newcastle, Newcastle, Australia
  • N.H. Lovell
    Graduate School of Biomedical Engineering, University of New South Wales, Sydney, Australia
  • Footnotes
    Commercial Relationships  L.E. Hallum, None; G.J. Suaning, None; N.H. Lovell, None.
  • Footnotes
    Support  none
Investigative Ophthalmology & Visual Science May 2004, Vol.45, 4175. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      L.E. Hallum, G.J. Suaning, N.H. Lovell; Effecting improved prosthetic visual acuity via image processing in a fixation–pursuit task . Invest. Ophthalmol. Vis. Sci. 2004;45(13):4175.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Abstract: : Purpose: To show that subjects undertaking a simulated prosthetic visually mediated task, wherein phosphene sizes are modulated, can capably fixate upon and pursue a target; and, that non–trivial image processing affords improved fixation and pursuit. Methods: A computer monitor was viewed by 20 subjects each trained for 3 hours. Viewing distance was fixed; phosphene pitch (PP) excited retinal foci separated by approx. 500 microns. Phosphenes were size–modulated, and of fixed intensity (white). The phosphene array was moveable via a joystick; subjects were required to track a small target (white square of width approx. 1/2 PP; black background). The target initially appeared in the centre of the monitor (for fixation), and after a short, random period it described a random, S–shaped course into the periphery (for pursuit) at an average velocity of 60 arcmin/s. Two image processing schemes were employed: regional averaging (RA), and Gaussian kernels (s.d. = 0.3 PP). For control, a normal (high–resolution) view of the target was afforded subjects, who tracked via the joystick cursor. Results: For fixation, subjects deviated, on average, 23.3 arcmin from the target for RA, and 14.9 arcmin for Gaussians (paired: 8.4 arcmin; P < 0.0001). For pursuit, subjects deviated 48.2 arcmin for RA, and 44.9 arcmin for Gaussians (paired: 3.3 arcmin; P < 0.0001). For control, average deviations for fixation and pursuit were 7.8 and 19.5 arcmin respectively. Conclusions: It is likely that prosthetic retinal implantees will be afforded capable fixation and smooth pursuit. Image processing, integral to a prosthetic device, that employs overlapping Gaussian kernels should improve these visual functions. Post hoc design of the Gaussian scheme should make for further improvement.

Keywords: image processing • low vision • shape, form, contour, object perception 
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×