Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Systemic: Information System and Informatics Journal
Muhammad Fahmi Herlambang
Institut Teknologi Nasional Bandung
Computer Science & IT

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

Found 1 Documents
Search

 1 
Pengenalan Karakter Huruf Braille dengan Metode Convolutional Neural Network Muhammad Fahmi Herlambang; Asep Nana Hermana; Kurnia Ramadhan Putra
Systemic: Information System and Informatics Journal Vol. 6 No. 2 (2020): Desember
Publisher : Program Studi Sistem Informasi Fakultas Sains dan Teknologi, UIN Sunan Ampel Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29080/systemic.v6i2.969

Abstract

Braille characters consists of 6 dots that are designed in such way to be a writing system to help blind people. However, learning or reading Braille characters isn’t an easy thing to do, because fingers sensitivity and understanding the writing system are needed to be able to read Braille. Therefore, there are some researches on Braille characters recognition with different methods and technologies, such as deep learning. The Convolutional Neural Network (CNN) is used. CNN method has been used in various recognition researches, such as face recognition, document analysis, image classification, etc. In this research, the CNN method is used to perform Braille characters recognition. The system performs the Braille character recognition process per character based on a model that has been trained using a dataset with the 26 Braille characters. The result of 81.54% accuracy is achieved for Braille character image acquisition with a smartphone with 0 to 4 degrees tilting and 30cm distance with training model using learning rate of 0.0001 and Adam optimizer.
Co-Authors Asep Nana Hermana Kurnia Ramadhan Putra