Sistem Pendeteksi Bahasa Isyarat Indonesia (BISINDO) Berbasis Android

Aryananda, I Gusti Agung Oka (2025) Sistem Pendeteksi Bahasa Isyarat Indonesia (BISINDO) Berbasis Android. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6026221004-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (9MB) | Request a copy

Abstract

Komunikasi yang terjalin antara penyandang tunarungu dengan orang di sekitarnya kerap kali menjadi kendala dikarenakan kebanyakan orang belum memahami bahasa isyarat yang kerap dijadikan media komunikasi oleh penyandang tunarungu. Untuk mengatasi permasalahan tersebut, penelitian ini bertujuan mengembangkan sistem pendeteksi Bahasa Isyarat Indonesia (BISINDO) berbasis Android menggunakan metode Design Science Research dan framework Mediapipe untuk mendeteksi dan menerjemahkan gerakan tangan menjadi teks. Sistem yang dirancang dengan fitur deteksi real-time gerakan tangan, penerjemahan huruf BISINDO (A-Z), serta tombol spasi dan penghapusan karakter, telah diuji coba terhadap 15 responden ahli dan menunjukkan kemampuan mengenali gerakan tangan dengan akurasi di atas 70% untuk mayoritas huruf, terutama pada huruf C, E, F, G, I, J, L, O, Q, S, dan V. Penelitian menemukan bahwa faktor pencahayaan, jarak kamera (optimal 15-35 cm), posisi tangan, dan konsistensi gerakan signifikan memengaruhi akurasi sistem, meskipun masih mengalami kendala mengenali huruf T dan X serta gerakan kompleks. Berdasarkan penilaian System Usability Scale (SUS), sistem mendapatkan persentase rata-rata 75% dan dikategorikan baik, sehingga memberikan kontribusi penting dalam mengembangkan solusi teknologi untuk mendukung komunikasi penyandang tunarungu, sambil mengidentifikasi area pengembangan di masa mendatang.
====================================================================================================================================
Communication between deaf individuals and those around them often becomes a challenge because most people do not understand sign language, which is frequently used as a communication medium by the deaf community. To address this issue, this research aims to develop an Indonesian Sign Language (BISINDO) detection system based on Android using the Design Science Research method and Mediapipe framework to detect and translate hand movements into text. The system, designed with features for real-time hand movement detection, BISINDO letter translation (A-Z), as well as space and character deletion buttons, was tested on 15 expert respondents and demonstrated the ability to recognize hand movements with an accuracy above 70% for most letters, particularly for letters C, E, F, G, I, J, L, O, Q, S, and V. The research found that factors such as lighting, camera distance (optimal 15-35 cm), hand position, and movement consistency significantly influence system accuracy, although it still faces challenges in recognizing letters T and X and complex movements. Based on the System Usability Scale (SUS) assessment, the system received an average percentage of 75% and is categorized as good, thus providing a significant contribution in developing technological solutions to support communication for deaf individuals while identifying areas for future development.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Bahasa Isyarat, Bahasa Isyarat Indonesia (BISINDO), Machine Learning, Mediapipe, Android, Design Science Reasearch, sistem pendeteksi Bahasa Isyarat Indonesia (BISINDO), Image Processing, Sign Language, Bahasa Isyarat Indonesia (BISINDO), Machine Learning, Mediapipe, Android, Design Science Reasearch, Indonesian Sign Language detection system (BISINDO)
Subjects:	Q Science Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.76.A63 Application program interfaces Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > T Technology (General)
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 59101-(S2) Master Thesis
Depositing User:	I Gusti Agung Oka Aryananda
Date Deposited:	21 Jan 2025 00:45
Last Modified:	21 Jan 2025 00:55
URI:	http://repository.its.ac.id/id/eprint/116457

Actions (login required)

View Item