GGrantIndex
← Search

On-Screen Deconvolution to Facilitate Computer Access for Users with Visual Impairments Involving Higher-Order Wavefront Aberrations

$745,563FY2003CSENSF

Florida International University, Miami FL

Investigators

Abstract

Successful interaction with a computer through a graphical user interface implicitly requires the undistorted perception of the images displayed on the computer screen. This is a problem for individuals with severe and irregular visual limitations that cannot be corrected through traditional mechanisms such as eyeglasses or contact lenses. In this project the PI will explore the possibility of having the computer generate modified displays to correct conditions such as keratoconus, pellucid marginal degeneration, Terrien's marginal degeneration, and related refraction disorders that distort the images perceived by the eye in asymmetrical and irregular ways. The irregularity of the distortion introduced by these conditions is captured by the need to include higher-order (i.e., >2) terms in the Zernike polynomial approximation of the eye's wavefront aberration (WA) function, which represents the distortion of images through the optical components of the eye (cornea, lens). Recent advances in image processing and optometry, as well as the continued promise of increasing computing power for real-time applications, make it possible to envision a computer system that, given the proper characterization of the visual capabilities of its user, will generate a display that matches his/her vision limitations in a complementary way, thus overcoming the user's vision problem. Furthermore, instruments capable of accurately characterizing a person's WA are now commercially available. The PI and his team will use the point-spread function (PSF), obtained from a person's WA, to pre-compensate (by deconvolution) the images shown on the computer screen, so that they are perceived undistorted by the individual whose PSF is considered. Broader Impacts: This project will lead to the design of novel assistive technologies, which in turn may well lead to tools that benefit the population at large. A substantial impact on education is also to be expected, due to the large number of students who will participate in the project and in light of the fact that the PI's institution is one of the few minority institutions in the nation with comprehensive programs through the PhD level. The deconvolution work has the potential for broad impact in both technical and societal dimensions, in that it constitutes a novel approach to compensation of images for visual correction purposes via digital rather than optical means, and in consideration of estimates that higher-order visual impairment neglected in traditional solutions may affect as many as 7,000,000 Americans who have excessively irregular refractive errors.

View original record on NSF Award Search →