Navigation and Object Identification Aid for the Visually Challenged

Shrey Painuli; Sanjay Gairola; Kirtika Arya; Divyani Tiwari; Balram

doi:10.18090/samriddhi.v13spli02.12

Shrey Painuli Deptt. of Electrical Engineering, Govind Ballabh Pant Institute of Engineering and Technology, Pauri Garhwal246194, Uttarakhand, India
Sanjay Gairola Deptt. of Electrical Engineering at Govind Ballabh Pant Institute of Engineering and Technology, Pauri Garhwal 246194,Uttarakhand, India.
Kirtika Arya Deptt. of Electrical Engineering at Govind Ballabh Pant Institute of Engineering and Technology, Pauri Garhwal 246194,Uttarakhand, India.
Divyani Tiwari Deptt. of Electrical Engineering at Govind Ballabh Pant Institute of Engineering and Technology, Pauri Garhwal 246194,Uttarakhand, India.
Balram Deptt. of Electrical Engineering at Govind Ballabh Pant Institute of Engineering and Technology, Pauri Garhwal 246194,Uttarakhand, India.

DOI: https://doi.org/10.18090/samriddhi.v13spli02.12

Keywords: Camera Module, ERM Vibration Motors, Haptic Feedback, Navigation Aid, Object Detection, Raspberry Pi,Tensorflow, Ultrasonic Sensor, Visual Impairment

Abstract

The aim of this paper is to develop an affordable consumer device to help the visually impaired people travel independently. The device is an electronic navigation system designed in the shape of a pair of eyeglasses, which utilizes 2 ultrasonic sensors to detect objects up to 400 cm in the front of the device and 2 vibration motors to produce vibro-tactile feedback to inform the wearer about the object's localization. The 2 vibration motors are so calibrated that the wearer can easily comprehend the distance of obstacles surrounding him. Simultaneously an object identification model captures live video from a camera module and recognize objects in the front, informing the wearer about the object detected through earphones, thus assisting the wearer to search any required object. The device uses a 'Raspberry Pi 3B+' microcomputer to process real time data, employing tensorflow for image processing.

Downloads

Download data is not yet available.