Artificial Intelligence Applications and Reconfigurable Architectures

7A Smart System for Obstacle Detection to Assist Visually Impaired in Navigating Autonomously Using Machine Learning Approach

Vijay Dabhade*, Dnyaneshwar Dhawalshankh^†, Anuradha Thakare^‡, Maithili Kulkarni^§ and Priyanka Ambekar^¶

Department of Computer Engineering Pimpri Chinchwad College of Engineering, Pune, India

Abstract

Obstacle detection is a popular approach for detecting barriers in the region of a subject. Visual impairment affects a large number of people around the world. Visually impaired or blind people encounter numerous challenges in their daily lives; the white cane is still the most widely used instrument for obstacle detection; in an unfamiliar environment, they rely entirely on other people to reach their desired goal. This will allow blind people to navigate independently without the use of any aids by utilizing object detection systems that detect objects over a period of time.

The proposed system in this article uses machine learning algorithms to see objects through the camera and then uses audio output to teach blind people about the item and its location. Obstacle detection approaches are discussed that can create and develop a system which will assist visually impaired people in navigating autonomously.

Keywords: Image processing, machine learning, obstacle detection, IoT

7.1 Introduction

According to the World Health Organization’s tenth edition, at least 2.2 billion individuals suffer from near or distance visual impairment. Vision impairment may ...