Assistive System for the Visually Impaired using Multiple Cameras and Sensors

On a global scale, at least 2.2 billion people are affected by a near or distance vision impairment. Vision is an essential aspect of human life, and vision loss is a problem many unfortunate people experience. Many blind people rely on white cane sticks and guide dogs for navigation, but they come with their own set of drawbacks. Assistive technologies are those utilities that aid visually impaired people in carrying out some specific tasks. This paper presents a novel assistive device that overcomes the issue of low visual coverage of the surroundings by using multiple cameras. The blind user can wear the cameras in the front, right, back, and left directions surrounding the neck, getting wider visual coverage. A raspberry pi microcomputer captures images from the cameras and uses them to find objects and faces without using the internet. The microcomputer provides voice feedback on the detections and the camera direction through earphones. An Arduino microcontroller connected to an ultrasonic sensor senses the obstacles and alerts the user with a vibration motor. A smartphone can connect to these Arduino and Raspberry pi modules via Bluetooth, and the user can send commands with the mobile application by touching or by voice. The system could get a broad picture of the surroundings and provide output in a reasonable amount of time.