Improving Outdoor Sound Distance Navigation for Visually Impaired People Using BLSTM and CRF

Abstract

Background and Aims: This study aims to develop a prototype that utilizes sound-based systems to localize individuals with visual impairments. In this study, the wireless ultrasound network is proposed and used to facilitate autonomous navigation and maneuvering for individuals with visual impairments. Methods: In the proposed method, we designed robust algorithms for ultrasonic systems with sound waves of high frequency to detect obstacles in the surrounding environment. The algorithms were developed using voice recognition method, bi-directional long short-term memory (BLSTM), and conditional random fields (CRFs). The proposed system was implemented with the help of assisting hardware instruments, such as an ultrasonic sensor network and a global positioning system (GPS). To analyze different outdoor settings, the Raspberry pi4 was used to locate and store the coordinates (points of interactive latitude and longitude) of four different outdoor places. These including home, mosque, laundry, and supermarket. Results: According to our results, the proposed approach was capable of accurately navigating the specific site with an accuracy of 98.5%. Conclusions: The proposed method was able to obtain a robust outdoor localization for VIP.