| A number of prevalent eye diseases exist which may lead to partial or total blindness, especially for which caused by Retinitis Pigmentosa (RP) or Age-related Macular Degeneration (AMD) that there are no means or treatments by which to cure or restore lost vision. With the development of micro technology, visual prostheses become an available approach for patients blinded to restore partial or complete sight. Despite different functional electrical stimulation (FES) locations including the visual cortex, retina and optic nerves have been considered for visual prostheses to elicit the perception of light. All visual prostheses share other same parts which are image capture, image processor, RF transfer, and micro-stimulator to provide the simple and available information for electrode array.In this manuscript we studied image processing methods and developed devices of image capture and processing for visual prostheses in order to obtain most available image features for recognition and low resolution image information for electrode array stimulating and evoking visual perceptions. Due to excellent processing capability, high-speed data transmission and mass storage memory, the TMS320DM642 DSP (Digital Signal Processor) for video image processing is used as our processing chip which could support most of the image processing algorithms. And we chose ARM (LPC2214) to output digital signals. Beyond developing the hardware system, image processing methods are indispensable part of the software design in our system. According to their different complexity, the images will be processed in different ways based on different complexity. Finally, we introduced code optimizing methods and got good results according to performance evaluation which showed that our system is reliable enough to provide signals for microelectrodes. |
PDF Full Text Request