Claim Missing Document
Check
Articles

Found 2 Documents
Search

Pemantauan Kepadatan Lalu Lintas di Kota Banda Aceh Secara Non-Invasif Menggunakan Bluetooth Low Energy Muhammad Ichsan; Dalila Husna Yunardi; Maya Fitria; Kurnia Saputra; Diky Wahyudi; Yahdina Ahsya; Rahmad Dawood
Prosiding SISFOTEK Vol 9 No 1 (2025): SISFOTEK IX 2025
Publisher : Ikatan Ahli Informatika Indonesia

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

Abstract

The rapid population growth in Banda Aceh City, reaching 259,538 inhabitants in 2024, has significantly increased traffic congestion and posed critical infrastructure challenges. Traditional video-based traffic monitoring systems raise privacy concerns due to the collection of personal data such as license plates and facial images. This research proposes a non-invasive traffic density monitoring system utilizing Bluetooth Low Energy (BLE) technology as a privacy-preserving alternative. The system architecture consists of three main layers: an Android-based mobile gateway for BLE signal detection, a cloud backend infrastructure using AWS services (API Gateway, Lambda, DynamoDB, and EventBridge) for data processing, and a web application developed with Next.js for visualization. Field testing conducted at Mesjid Raya Baiturrahman Bus Stop demonstrated that the system operates stably and effectively. The mobile gateway successfully detected BLE devices continuously, transmitted data to the backend without packet loss, and the web dashboard displayed traffic information through interactive maps and tables. The results indicate that BLE-based traffic monitoring provides an effective, and privacy-preserving solution for urban traffic management, particularly suitable for developing regions with limited technical infrastructure.
Pengembangan Aplikasi Pembelajaran Bahasa Isyarat Indonesia (BISINDO) yang Interaktif Berbasis Deep Learning Husaini Husaini; Ardiansyah Ardiansyah; Mahyus Ihsan; Kurnia Saputra
VOCATECH: Vocational Education and Technology Journal Vol 8, No 1 (2026): April
Publisher : Akademi Komunitas Negeri Aceh Barat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38038/vocatech.v8i1.290

Abstract

AbstractCommunication involving persons with hearing impairments and the wider Indonesian society continues to face various challenges, particularly due to the limited availability of accessible and interactive learning media for Indonesian Sign Language (BISINDO). This study focused on the development of an innovative and user-friendly Android-based BISINDO learning application designed as a supportive medium for beginner-level sign language learning. The developed application integrates real-time hand gesture recognition technology using a deep learning approach, employing a combination of MobileNetV2, MediaPipe Hands, and Convolutional Neural Networks (CNN). The application features three main components: sign language alphabet detection, a gesture dictionary presented in the form of visual animations, and an interactive quiz for learning evaluation. The application was developed following the Waterfall model, encompassing requirements analysis, design, implementation, testing, and deployment. Functional testing using Black-Box Testing method was conducted to ensure that all features operated according to specifications, while application usability was evaluated using the System Usability Scale (SUS) involving 32 respondents. The functional testing results indicate that all application features functioned properly and met the specified requirements. Meanwhile, the SUS evaluation yielded an average score of 74.8, which falls within the Good category, indicating that the application is well accepted by users. Thus, this application is intended to serve as an effective BISINDO learning medium and contribute to fostering more inclusive communication involving persons with hearing impairments and the wider community.AbstrakKomunikasi antara penyandang tunarungu dan masyarakat umum di Indonesia masih menghadapi beragam kendala, terutama berkaitan dengan terbatasnya media pembelajaran Bahasa Isyarat Indonesia (BISINDO) yang mudah diakses dan interaktif. Penelitian ini difokuskan pada pengembangan aplikasi pembelajaran BISINDO berbasis Android yang inovatif dan mudah digunakan sebagai media pendukung pembelajaran bahasa isyarat bagi pemula. Aplikasi yang dikembangkan mengintegrasikan teknologi pengenalan gerakan tangan secara real-time menggunakan pendekatan deep learning, dengan memanfaatkan kombinasi MobileNetV2, MediaPipe Hands, dan Convolutional Neural Network (CNN). Aplikasi ini dilengkapi dengan tiga fitur utama, yaitu deteksi abjad bahasa isyarat, kamus gerak dalam bentuk animasi visual, serta kuis interaktif sebagai sarana evaluasi pembelajaran. Pengembangan aplikasi dilakukan menggunakan model Waterfall yang meliputi analisis kebutuhan, desain, implementasi, pengujian, dan penerapan. Pengujian fungsionalitas sistem menggunakan metode Black-Box dilakukan untuk memastikan seluruh fitur berjalan sesuai spesifikasi, sementara tingkat kegunaan aplikasi dievaluasi menggunakan System Usability Scale (SUS) terhadap 32 responden. Hasil pengujian fungsionalitas mengindikasikan bahwa seluruh fitur aplikasi telah berfungsi secara baik dan dinyatakan memenuhi spesifikasi yang ditetapkan. Sementara itu, hasil evaluasi SUS memperoleh skor rata-rata sebesar 74,8 yang termasuk dalam kategori Good, menandakan bahwa aplikasi dapat diterima dengan baik oleh pengguna. Dengan demikian, aplikasi ini diharapkan dapat menjadi media pembelajaran BISINDO yang efektif serta mendukung terciptanya komunikasi yang lebih inklusif antara penyandang tunarungu dan masyarakat umum