HIGH SPEED AUTOMATIC OBJECT IDENTIFICATION BASED ON ROBUST FEATURE EXTRACTION FOR VISUALLY IMPAIRED PEOPLE
SHAIK FAHMEEN TAZ, MD.NOOREBANU
Under graduate student, E.C.E, Nellore, Andhra Pradesh, shaikfahmeentaz@gmail.com
Under graduate student, E.C.E, Nellore, Andhra Pradeshnoorebanu14@gmail.com
Abstract
The project presents the automatic object identification system based on Speeded up Robust Features Extraction for visually impaired people. Navigation assistance for visually impaired is a vision substitute system designed to assist blind people for autonomous navigation. It working concept is based on ‘image to sound’ conversion. Navigation systems have been used for quite a long to support human perception of surrounding environment beyond reach of eyes. This concept has applied in this project. This research aims at finding a reliable and easy to implement solution for visually challenged persons at a very small cost. The input image was captured by the camera and it is pre processed. The SURF (Speeded up Robust Features Extraction) based feature extraction is done. For output analysis, we have some reference templates in different orientation are stored. A matching technique measures the minimum distance and the input was identified. Then that object name is transferred to the hardware module using serial communication process and heard by the person through the speaker setup.
