Improving computer interaction for users with visual acuity deficiencies through inverse point spread function processing

Human beings take advantage of their high visual acuity to perform many daily activities required of them. Specifically, normal visual acuity is a pre-requisite for proper usage of most contemporary graphic user interfaces. The most common forms of visual acuity loss are myopia, hyperopia, and astigmatism. Contact lenses or glasses can easily correct these simple visual aberrations. There exist, however, more complex aberrations that cannot be easily remedied by such means. These "high-order" aberrations are modeled through what is known as the point spread function (PSF) of the human eye. The PSF can be obtained indirectly through the wavefront aberration function of the human eye, currently accessible through wavefront analyzers. Thus, it is feasible that with the knowledge of the PSF, digital images could be altered according to a transformation opposite to the one they suffer in an aberrated eye. Consequently, when they are displayed to the user, he/she will perceive them undistorted. This paper presents an image processing approach, based on deconvolution of the PSF from the intended computer images that achieves this objective. The theoretical foundation of the approach is introduced along with simulation results using actual PSFs from subjects.