부산대학교 도서관

One of the greatest challenges to independence for the blind and low vision (BLV) individuals are difficulties in self-navigation. This paper presents a wearable assistive technology to augment the perception of the BLVs along their navigational pathway. The research addresses specifically the challenge of negotiating surface discontinuities typically found within urban areas. A lightweight, small and unobtrusive wearable prototype is developed, and it utilizes a stereo camera to collect depth data from the navigational pathway. A depth profile is proposed to reduce the dimension of depth map computed from the image pair, and a deep belief network is then assembled to classify the depth profiles such that appropriate haptic feedback can be communicated to the user.