Out of seven billion of the world’s population, two billion and two million that amounts to 31.43% have visual impairment or blindness according to the World Health Organization (WHO) statistics report. Hence, the need to develop a wearable device with reduced size, efficient power usage, and for more comfortability of the visually impaired or blind people. This work aims at designing an obstacle detection system using an ultrasonic sensor interfaced with an Arduino board to track location, alert patient, and send location messages of visually impaired patient to guardians as a feedback mechanism using a GPRS and GSM module. The C programming language was used as the instruction code to interface Arduino device to carry out given tasks. At the design level, the circuit was first tested on Proteus software for simulation purposes before its hardware implementation. The results obtained from the test show the variation of distance as the patient approaches the obstacle, and messages received when a fix was obtained. This design concept would help reduce danger across the way of those with sight defects and allow them to go to familiar places without any aid smoothly.
Proposed real-time obstacle detection system for visually impaired assistance based on deep learning
