Manuscript Details


Title

Indoor Navigation System for Visual Impaired (INSVI)

Authored By

Prof. Sudipta Bhadra

Assistant Professor, Information Technology, Heritage Institute of Technology


Prof. Anindya Jyoti Pal

Associate Professor, Information Technology, Heritage Institute of Technology


Mr. Sk Irfan Tanveer

M.Tech Student, Information Technology, Heritage Institute of Technology

Category

INFORMATION TECHNOLOGY (IT)

Abstract

Visually impaired people face a common problem of independent mobility in a familiar/unfamiliar indoor environment. At present almost 8.90 Cr. Indian are visually impaired. Majority of those cannot travel independently. In this paper we have described a system that facilitates visually impaired person to move in any indoor environment. The system is named as INDOOR NAVIGATION SYSTEM FOR VISUAL IMPAIRED (INSVI). The system can also be used by users with cognitive disabilities. We have created a simple and less costly system by eliminate the installation expenditure and minimize the amount of sensors required for the system. The goal of our system is to detect and avoid obstacles. Our system also provides the necessary instructions to visually impaired persons for navigation within a building. Visual impaired people's real world navigation experience can be increased by our prototype by vocal navigation instructions. Our system can replace conventional techniques of avoid obstacles such as guide dog and long cane. Advances in computer vision like invention of Kinect depth sensor have made our goal possible. The system has mainly two major process data extraction and navigation process. Kinect camera facilitates our system's data extraction by capture and analysis of image. This range camera gives us the distance value of obstacle from camera for each pixel in a particular frame. The navigation process uses two key pieces of information first direction of user's movement and second position of objects in the environment. Our system can be connected with Internet for Internet of Things based implementation for remote assistance.