Real Time Detection and Recognition of Hand Held Objects to Assist Blind People

Abstract

There are approximately 315 million people with visual imparity worldwide. The number of these
people is also increasing as this generation ages. Recent development in computer vision, digital cameras, and
portable computers make it possible to assist these people by developing camera-based products that incorporate
computer vision technology. We propose a camera-based object recognition application to help visually impaired
people to recognize labeled and unlabelled hand held objects in their daily lives. To differentiate the object from
complex backgrounds or other surrounding objects in the camera view, we first propose a motion based method to
define a region of interest (ROI) in the image by asking the user to hold the object which is efficient and effective.
Reference objects and test images are accompanied by local features (regions, interest points, etc.). If the test image
get match with reference image then the output signal created in the form of audio. Output audio is generated and
sent towards head-phone through which the blind user could hear the information about the product which is held
on hand. In this proposed system we use speeded up robust features (SURF) for tasks such as object recognition,
classification or 3D reconstruction, image registration.