Return to search

Visual guidance for the disabled using intelligent tele-agents

M.Sc. / We live in a modern world in which visual perception has become an absolute necessity. Navigating and walking around in a city without getting lost is a difficult enough task when using all senses, but if you are visually disabled, this becomes a near impossible task. All around us are signs, billboards, motorcars, buildings, computers, and other similar signs of modern times, which are most effectively observed visually. The next logical step in assisting the visually disabled to experience the world around them more freely, is to make more effective use of the technology that has created the shift to the visual world in the first place. It now appears possible to design a framework to incorporate not only current, but also future hardware and software into a solution to the above-mentioned problems. Such a framework has to be flexible to allow it to keep up not only with hardware, but software advances as well. Furthermore, it needs to take into account the needs of a typical blind user. One way of implementing this framework is to make use of a form of sensereplacement. Where the visual sense is impaired, technology can be used to analyze and interpret the visual world, obtain meaningful information from the scene, and then re-route this information to another sense. This dissertation is divided into three sections. The first section will provide an overview of the rest of the dissertation. It will also investigate similar research that is currently being undertaken, and provide a model for a possible solution to the above-mentioned problems. The second section will provide the background study needed to make informed decisions when implementing a prototype system. The third section will investigate the implementation of a prototype model, as well as the construction of a pilot project.

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uj/uj:2064
Date27 February 2012
CreatorsBasson, Steyn Nel
Source SetsSouth African National ETD Portal
Detected LanguageEnglish
TypeThesis

Page generated in 0.002 seconds