Effect on prosthetic vision visual acuity by filtering schemes, filter cut-off frequency and phosphene matrix: a virtual reality simulation

Visual acuity of prosthetic vision was examined under virtual reality simulation. Prosthetic vision was simulated by first filtering an image using circular mean filters or Gaussian smoothing filters of different cut-off frequencies. Pixel values at 100 fixed sites of the filtered image were taken, sampling either with a regular rectangular or hexagonal matrix. Each pixel value was transformed into a Gaussian intensity profile centered at the corresponding position at which the sample was taken to simulate the evoked visual effect of an electric stimulation. Visual acuity scores of three subjects, each completing two sets of results, were recorded across different filtering schemes, cut-off frequencies and sampling matrices. The best mean score recorded was 1.55 logMAR, with the worst being 1.70 logMAR. The difference was mostly attributed to filter cut-off frequency. Differences between filtering schemes were insignificant. Results also showed emerging trends demonstrating differences between rectangular and hexagonal sampling matrices.