Rancang Bangun Detachable Multi-Sensor Glove Penerjemah Sistem Isyarat Bahasa Indonesia bagi Individu Tunarungu-Wicara

Darmawan, Yakina Athiyyah (2023) Rancang Bangun Detachable Multi-Sensor Glove Penerjemah Sistem Isyarat Bahasa Indonesia bagi Individu Tunarungu-Wicara. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 07311940000054-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 September 2025. Download (4MB) | Request a copy

Abstract

Hingga saat ini, prevalensi penyandang disabilitas sensorik terus meningkat secara pesat, salah satunya termasuk tunarungu dan tunawicara. Metode komunikasi yang umumnya digunakan tunarungu-wicara ialah bahasa isyarat, namun metode ini memiliki keterbatasan di mana tidak semua orang memahami dan mampu mempraktikkannya. Penelitian ini mengembangkan rancangan detachable multi-sensor glove yang mampu menginterpretasikan Sistem Isyarat Bahasa Indonesia (SIBI) ke dalam format teks dan audio, dengan mengakomodasi intersubject variability yang terkait dengan ukuran tangan dan pola gerakan. Digunakan beberapa jenis sensor meliputi flex sensor, inertial measurement unit (IMU) berupa modul GY-952, serta sensor tekanan berupa force sensitive resistor (FSR) untuk membaca karakteristik yang membedakan masing-masing gesture alfabet dalam pedoman SIBI. Data analog karakteristik yang diperoleh lalu diolah melalui analog-to-digital-converter (ADC), untuk kemudian melalui proses klasifikasi. Hasil pengolahan data tersebut kemudian dikirimkan ke smartphone pengguna hingga menghasilkan keluaran berupa sebuah kata yang komprehensif dalam percakapan, dengan format teks dan audio. Glove pada alat asistif ini didesain khusus agar komponen dapat dilepas dan dipasangkan ke glove dengan tiga variasi ukuran, sehingga dapat menyesuaikan variasi ukuran tangan pengguna tanpa mengganggu performa sensor. Desain detachable memungkinkan penyesuaian classification rule berdasarkan ukuran, sehingga dapat meminimalisir nilai error akibat perbedaan ukuran tangan pengguna. Pengujian pada penelitian ini mencapai akurasi sebesar 92,18% untuk protokol klasifikasi alfabet, dan akurasi sebesar 91% untuk protokol kata. Intersubject variability terkait ukuran tangan dan pola gerakan memberikan pengaruh yang signifikan terhadap performa sistem interpretasi, sehingga ukuran yang sesuai penting untuk mencapai akurasi yang lebih baik.
===================================================================================================================================
To date, the prevalence of sensory disability continues to increase rapidly, one of which includes hearing and speech impairment. Individuals with this condition are known as hearing and speech impaired. The commonly used communication method for hearing and speech impaired individuals is sign language, but this method has limitations in that not everyone understands and is able to practice it. This study develops a design for a detachable multi-sensor glove that is able to interpret the Indonesian Sign Language (SIBI) into text and audio format, accommodating intersubject variability related to hand size and movement patterns. Several types of sensors are used, including flex sensors, inertial measurement units (IMUs) in the form of accelerometers and gyroscopes, and pressure sensors in the form of force sensitive resistors (FSRs) to recognize the characteristics that distinguish each alphabet in the SIBI dictionary. The analog characteristic data obtained is then processed through an analog-to-digital-converter (ADC), then undergoes classification and segmentation process. The processed data is then sent to the user's smartphone, with a comprehensive string as the final output, in text and audio format. The glove in this assistive device is designed so that the components can be removed and attached to another glove in three different sizes, accomodating user's hand size variation without affecting sensors’ performance. Detachable design enables adjustments of classification rules according to size, which could minimize the error value caused by hand size differences. The trials in this study achieved an accuracy of 92.18% for the alphabet classification protocol and an accuracy of 91% for the string protocol. Intersubject variability related to hand size and movement patterns has a significant influence on the performance of the interpretation system, therefore appropriate sizing is crucial to achieve better accuracy.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Sensor glove, detachable, Sistem Isyarat Bahasa Indonesia (SIBI), tunarungu-wicara, intersubject variability; Sensor glove, detachable, Indonesian Sign Language (SIBI), hearing and speech impaired, intersubject variability
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK351 Electric measurements. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Yakina Athiyyah Darmawan
Date Deposited:	05 Oct 2023 01:32
Last Modified:	05 Oct 2023 01:32
URI:	http://repository.its.ac.id/id/eprint/103603

Actions (login required)

View Item