Muhammad Nazrenda Ramadhan
Fakultas Ilmu Komputer, Universitas Brawijaya

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Deteksi Tangga Naik dan Turunan untuk Notifikasi Keamanan pada Tunanetra menggunakan YOLO Versi 4 berbasis Jetson Nano B01 Muhammad Nazrenda Ramadhan; Fitri Utaminingrum; Dahnial Syauqy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 1 (2022): Januari 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Blind people currently use The White Cane to help with their daily activities. However, The White Cane has a drawback where its detection range is limited to the length of the cane. In addition, The White Cane cannot distinguish objects that are in front of blind people. This study aims to develop a Jetson Nano B01-based system that can detect floors and stairs, both going up and down to assist the activities of blind people. With the help of artificial intelligence, this system is expected to be able to notify blind people that there is a stair in front of them by activate a buzzer. Then, to be able to produce the right stair detection, pattern recognition is needed with the You Only Look Once (YOLO) method which has a fast detection speed. When the stairs are identified, the system will give a notification in the form of a buzzer sound to notify that there is a stair ahead. The tests carried out, obtained the results of the classification accuracy of object detection (Floor, Upstairs, and Downstairs) of 90%, the average computation time of 0.177s, and the integration accuracy of YOLOv4 detection with a buzzer of 100%.