Microcomputer-based artificial vision support system for real-time image processing for camera-driven visual prostheses

摘要

It is difficult to predict exactly what blind subjects with\udcamera-driven visual prostheses (e.g., retinal implants) can perceive.\udThus, it is prudent to offer them a wide variety of image processing\udfilters and the capability to engage these filters repeatedly in any userdefined\udorder to enhance their visual perception. To attain true portability,\udwe employ a commercial off-the-shelf battery-powered general\udpurpose Linux microprocessor platform to create the\udmicrocomputer-based artificial vision support system (µAVS^2) for\udreal-time image processing. Truly standalone, µAVS^2 is smaller than a\uddeck of playing cards, lightweight, fast, and equipped with USB, RS-\ud232 and Ethernet interfaces. Image processing filters on µAVS^2 operate\udin a user-defined linear sequential-loop fashion, resulting in vastly\udreduced memory and CPU requirements during execution. µAVS^2\udimports raw video frames from a USB or IP camera, performs image\udprocessing, and issues the processed data over an outbound Internet\udTCP/IP or RS-232 connection to the visual prosthesis system. Hence,\udµAVS^2 affords users of current and future visual prostheses independent\udmobility and the capability to customize the visual perception\udgenerated. Additionally, µAVS^2 can easily be reconfigured for other\udprosthetic systems. Testing of µAVS^2 with actual retinal implant carriers\udis envisioned in the near future.