الفهرس 
Research ScopeOnly 14 pages are availabe for public view
 |  | from | 132 |  |  |
| | | from | 132 | | |
Abstract
Restoring vision to the blind has been a challenge for long time. But today, it is no longer impossible thanks to Visual Prostheses. A visual prosthesis is a new hope to provide cure for the blind who suffers from damage in the eye or in the early stages of the visual pathway. Its main idea is to substitute the damaged organs of the visual pathway with an artificial system that mimic its functionality. Then, integrate it with the first intact organ in the pathway. The development of this system faced different challenges on the past decades. The main challenges were the unclear functionality of some organs in the visual pathway, technology limitations on the stimulating electrodes, and the small number of stimulating electrodes which limit the output quality of visual prostheses systems.
The main contribution in this research is the definition of an Image Processing Model to be followed in any visual prostheses system. This model defines the required image processing to highlight the main features in any visual input and remove the less important features. This will help to make use of the limited number of electrodes. Another role of this model is that it defines the functionality of the visual organs deep till the Lateral Geniculate Nuclease (LGN), which is the target of visual prostheses in this research. Another contribution is the development of a Features Extraction technique for dynamic scenes. Results of the new developed technique showed better perception and performance in comparison with another four state-of-the-art techniques. Finally, an Image Processing Toolbox was developed to handle the entire image processing functionality needed in visual prostheses. It was used to develop our new technique. This developed toolbox could be the starting point for future researches in visual prosthesis.