Virtual Eye- Assisting Visually Impaired
Abstract- For blind people who have no visual information, there are a lot of dangers in everyday-life environments.
Independent travel is well known to present significant challenges for such individuals, thereby reducing quality of life and
compromising safety. This project aims to develop an intelligent, high performance, affordable and easy to use computer
vision based visual information assistance system for blind. Visual information will be captured via two miniature cameras
mounted on head via a cap or sunglasses, while image processing and speech synthesis will be done by a portable computer.
Our detection results are from a large data set of indoor images including doors, staircases, cupboards, walls, etc. so that the
proposed method is robust and generic to the changes of scales, view points, and occlusions in different buildings. An area
based stereo matching is performed over the transformed images to calculate dense disparity image. Low texture filter and
left/right consistency check are carried out to remove the noises and to highlight the obstacles. The audio is conveyed to the
blind user through stereo headphones. The audio includes navigation assistance commands such as presence of obstacle, clear
path, object of interest, doors nearby, other humans surrounding etc.