Artificial Vision for the Blind Using Motion Vector Estimation Technique

摘要

Blind person can be given artificial vision to know the locations of the pedestrians or moving objects around them. This paper proposes a system for detection of obstacles moving in front of a person on a walk by a camera. The system senses if a moving object in front of the person is going to hit him/her in the near future by analyzing the motion vectors acquired on the camera images. Finally the system will notify the user as these sentences quot;object is coming towardsquot;,quot; object is going awayquot; and quot; chance of collisionquot;. In this way, the proposed system helps safe walking of a visually impaired person. The proposed method perform comparative study of different kind of corner detectors such as SUSAN, CSS, NOBLE and Harris corner detector and also compare different feature trackers such as Lucas- Kanade- Tomasi feature tracker. Thus suggest a better combination of feature detectors- feature tracker technique.