Simulated Prosthetic Vision: Improving text accessibility with retinal prostheses

Image processing can improve significantly the every-day life of blind people wearing current and upcoming retinal prostheses relying on an external camera. We propose to use a real-time text localization algorithm to improve text accessibility. An augmented text-specific rendering based on automatic text localization has been developed. It has been evaluated in comparison to the classical rendering through a Simulated Prosthetic Vision (SPV) experiment with 16 subjects. Subjects were able to detect text in natural scenes much faster and further with the augmented rendering compared to the control rendering. Our results show that current and next generation of low resolution retinal prostheses may benefit from real-time text detection algorithms.

I. Introduction

Several research groups are designing neuroprotheses which aim at partly restoring vision for blind people. Among these systems, retinal implants are the most advanced (see [1] for a review). To date, two radically different approaches have been proposed: (1) the video images from an external camera are transformed into electrical pulses delivered wirelessly to an internal electrode array, (2) light entering the eye is converted into a pattern of electrical stimulations via micro-photodiodes integrated within the implant. In both devices, electrical stimulations elicit white/yellow spots of light called phosphenes [2]. The resulting crude vision is a set of phosphenes displayed in a restricted portion of the visual field (field of view of 15° to 20°) with little dynamic range (4 to 10 luminance levels per phosphene).