Alexa Enabled Assistive System For Visually Impaired Persons Using AWS, Raspberry PI, And C#

Abstract

Purpose: One of the most challenging tasks visually impaired persons struggle with is identifying people or objects at every step. The rise in image processing and the development of face detection and recognition algorithms motivates the development of devices that can assist the visually impaired. In this research, we use image processing to represent the design and implementation of a facial recognition system for the visually impaired. The device developed consists of programmed Raspberry Pi hardware. The data is fed into the machine in the form of images. The images are preprocessed, and then the input image captured is processed inside the Raspberry Pi module using the KNN algorithm; the face is recognized, and the name is fed into text in the text-to-speech conversion module. The visually impaired people will quickly identify the person who is standing in front of them. The device is Alexa enabled, and the user can ask to identify the person or object by voice command.Design/Methodology/Approach: Here, we design a comprehensive assistive technology tool designed and implemented to support the various challenges faced by Visually Impaired (VI) individuals in day-to-day life. The system comprises smart glasses, a processing unit, and a smartphone application, providing the individual with a convenient and cost-effective solution. The Alexa interface made the device more convenient. The user can ask Alexa to identify people or even objects. Findings/Result:The Experiment results show high face detection accuracy and promising face recognition accuracy in suitable conditions. The developed system was tested on people with visual impairments and was well-appreciated. The device is built in such a way as to improve the cognition, interaction, and communication of visually impaired students in schools and colleges.Originality/Value: There are several devices already available. These devices have yet to become popular due to their lack of consideration for real-life aspects and cost. But we consider several real-life facts to facilitate reading in English and Arabic, in-door navigation, face recognition, and object detection, thus empowering them to lead independent lives. These multiple features made this unique.