A novel approach of assisting the visually impaired to navigate path and avoiding obstacle-collisions

There is no alternative of visual percepts to brain for accomplishing any easy to complex problem solution. Sustaining in the everyday world environment demands a lot from visual information processing of brain. But a large number of the human being is devoid of the blessings of visual percepts due to biological or accidental causes. This paper presents a critique on vision aided systems and suggests an inexpensive alternative to develop an aided system for visually impaired persons. We propose a real-time solution which utilizes image processing methodology and low cost hardware to support the visually impaired for every day path navigation and obstacle avoidance. A head mounted camera processes the image to identify plausible path and obstacle. The identified path direction and obstacle are converted into pre-defined vocabularies with audio annotation output to the physically challenged people´s ears. This method has been tested over varying environmental conditions and is found to be effective for visually impaired navigation.