Claim Missing Document
Check
Articles

Found 16 Documents
Search

An Analysis of Distance Extension Method in Visible Light Communication (VLC) Performance KRISTIANA, LISA; DARLIS, ARSYAD RAMADHAN; DEWI, IRMA AMELIA; LIDYAWATI, LITA; ARCHANDHIKA, HEGAR REFALDY
ELKOMIKA: Jurnal Teknik Energi Elektrik, Teknik Telekomunikasi, & Teknik Elektronika Vol 8, No 1: Published January 2020
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/elkomika.v8i1.218

Abstract

ABSTRAK A Visible Light Communication (VLC) adalah teknologi yang menawarkan konsep inovatif karena VLC menerapkan cahaya tampak untuk mentransmisikan informasi dari satu titik ke titik lain. Tantangan utama dalam penerapan VLC adalah pelemahan sinyal cahaya tampak karena faktor jarak dari titik sumber ke titik tujuan. Penelitian ini berfokus pada metode untuk merancang dan menerapkan pemancar dan penerima VLC pada media udara. Dengan membandingkan berbagai macam tipe LED, pengukuran yang didapatkan menunjukkan bahwa pemancar dan penerima VLC dapat ditingkatkan kemampuannya sehingga mencapat jarak maksimum 8.5 meter dengan menggunakan LED HPL. Kata kunci: Visible Light Communication, VLC Transceiver, Distance Extension Method, Light Emitting Diodes (LEDs).  ABSTRACT A Visible Light Communication (VLC) offers the innovative concept in telecommunication since it implements visible lights to transmit information from one point to other points. The main challenge in VLC is the attenuation due to the distance from source to destination. This research focuses on extension method to design and implement the VLC transceiver in an air medium. By comparing the real measurement of several types of LEDs, the distance of VLC transceiver can be extended up to 8.5 meters by applying HPL LED. Keywords: Visible Light Communication, VLC Transceiver, Distance Extension Method, Light Emitting Diodes (LEDs).
PENGEMBANGAN APLIKASI PENDUKUNG PEMETAAN GEOSPASIAL BERBASIS FOTOGRAMETRI UDARA DI DESA PASIRMULYA Miftahuddin, Yusup; Dewi, Irma Amelia; Tahttadu, Winter Bee; Pratama, Yanuarsyah; Haidar, Muhamad Rafi; Fadhlurrohman, Naufal Mufid; Almawahib, Zulfa Sulthany; Ibrahim, Farrel Adi
Jurnal Abdimas Ilmiah Citra Bakti Vol. 5 No. 4 (2024)
Publisher : STKIP Citra Bakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38048/jailcb.v5i4.3929

Abstract

Desa Pasirmulya, kecamatan Banjaran, kabupaten Bandung, provinsi Jawa Barat belum memiliki peta wilayah yang aktual dan terbaru. Terdapat perselisihan mengenai batas wilayah oleh warga. Berdasarkan hal tersebut diperlukan kegiatan dalam pembuatan peta wilayah dengan ortofoto. Tujuan utama kegiatan ini adalah menyediakan solusi teknologi yang mampu meningkatkan akurasi, efisiensi, dan aksesibilitas data geospasial dalam mendukung pengelolaan wilayah desa. Tahapan dalam pembuatan peta adalah insisialisasi, pembuatan aplikasi, survey, sosialisasi hasil dan evaluasi. Kegiatan dilaksanakan pada tanggal 16 September - 16 Januari 2024. Hasil dari kegiatan ini adalah peta ortofoto wilayah Desa Pasirmulya dengan tingkat kepuasan warga dalam pembuatan peta sebesar 80 %. Hal ini menunjukkan bahwa aplikasi yang dikembangkan mampu memproses data fotogrametri udara secara efisien, menghasilkan peta geospasial dengan akurasi tinggi yang relevan untuk kebutuhan perencanaan dan pengelolaan desa. Dengan demikian, aplikasi ini memberikan kontribusi penting dalam mendukung pembangunan berbasis data di tingkat desa. Ke depan, pengembangan lebih lanjut diperlukan untuk menambahkan fitur-fitur canggih, seperti integrasi dengan data demografi dan analisis spasial otomatis, guna meningkatkan manfaat aplikasi dalam pengelolaan wilayah secara menyeluruh.
The Detection of Objects and Distance for the Visually Impaired by Using Deep Learning ResNet-152 and the Triangulation Method Ichwan, Muhammad; Dewi, Irma Amelia; Salsabilla, Nadiati
Journal of Artificial Intelligence and Software Engineering Vol 5, No 1 (2025): March
Publisher : Politeknik Negeri Lhokseumawe

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30811/jaise.v5i1.6419

Abstract

This research aims to detect objects and determine the distance from the mobile camera to facilitate and assist visually impaired users in recognizing the surrounding environment using several models made with the RetinaNet Method and Residual Network-152 architecture. Three object detection models were generated by ADAM and SGD parameter optimizers. Object recognition was performed using the TensorFlow framework with a dataset of 2,444 images. The first model with training parameters used is ADAM optimizer, epoch 50, batch size 16, and lr 1e-5. The second model has training parameters, such as ADAM optimizer, epoch 100, batch size 16, and lr 1e-5. The third model uses SGD optimizer training parameters, epoch 50, batch size 16, and lr 1e-5. Based on 250 tests on each model, the results show that the best model is the first model, which shows a precision value of 82%, a recall value of 98%, an f1 score value of 89%, and an accuracy value of 86%. The distance from the mobile camera is tested in multiples of 10 at a distance of 100-300 cm with a camera height of 100-130 cm and a camera angle of 80⁰-90⁰ getting reasonable distance detection results at a camera height of 130 cm because it gets the smallest total difference value of 14.3 cm.
Multiscale Facial Detection using RetinaFace Architecture with Loss Function Dewi, Irma Amelia; Maryadi, Nadhiva Adzra Tsania
Journal of Computer Networks, Architecture and High Performance Computing Vol. 7 No. 3 (2025): Articles Research July 2025
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v7i3.6161

Abstract

Facial recognition technology has become increasingly prevalent in modern applications, from security systems to social media platforms. However, one of the most significant challenges in this field remains the accurate detection of faces across varying scales, orientations, and image qualities. Traditional face detection methods often struggle when faces appear at different sizes within the same image or when dealing with low-resolution imagery, leading to inconsistent performance that can compromise system reliability. The RetinaFace architecture emerges as a promising solution to address these multiscale detection challenges. By incorporating a Feature Pyramid Network (FPN), the system creates a hierarchical representation of features that enables effective detection of faces regardless of their size in the image. The FPN combines low-resolution, semantically strong features with high-resolution, semantically weak features, creating a robust feature pyramid that simultaneously captures facial characteristics at multiple scales. Context modules within RetinaFace further enhance detection capabilities by providing additional contextual information that helps distinguish faces from background noise and other objects. This comprehensive approach allows the system to maintain high accuracy even in challenging scenarios where faces appear small, partially occluded, or at unusual angles. The comparative analysis between ArcFace and SphereFace loss functions reveals important insights into optimization strategies for facial recognition systems. The experimental results on the WIDERFACE dataset demonstrate exceptional performance, with the RetinaFace-ResNet152-SphereFace combination achieving 94% accuracy. These findings highlight the importance of architectural choices and loss function selection in developing robust facial recognition systems capable of handling real-world deployment challenges
Identifikasi Nada antara Suling Sunda dan Suling Rekorder dengan Menggunakan Metode Frequency Cpstral Coefficients (MFCC) dan Dynamic Time Warping (DTW) Suryadikarsa, Fawwaz Muhammad; Nurhasanah, Youllia Indrawaty; Dewi, Irma Amelia
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 7 No 1: Februari 2020
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

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

Abstract

Suling adalah sebuah instrumen musik yang biasa digunakan oleh para pemain musik ataupun masyarakat pada umumnya. Suling sunda merupakan alat musik tradisional asal Pasundan ini mampu menghipnotis yang mendengarkannya karena nada khasnya yang indah, suling rekorder adalah alat musik modern dengan bunyi seperti peluit. Tetapi tidak banyak orang tahu bahwa nada pada suling sunda bisa juga dimainkan pada suling rekorder, sehingga pelajar yang mengikuti dengung (seni sunda) harus membawa 2 buah suling ke sekolah apabila bertepatan dengan kelas musik. Identifikasi nada antara suling sunda dan rekorder ini adalah sebuah sistem yang digunakan untuk membandingkan dan mencocokan frekuensi nada yang sama antara suling sunda dan suling rekorder, agar musik yang dimainkan di suling sunda bisa juga dimainkan di suling rekorder. Penelitian ini dibuat dengan menggunakan algoritma Mel Frequency Cepstral Coefficient (MFCC) sebagai metode proses ekstraksi ciri dan algoritma Dynamic Time Warping (DTW) sebagai identifikasi nada dengan perbedaan waktu pada saat perekaman. Berdasarkan hasil penelitian yang dilakukan, sistem mengidentifikasi nada suling sunda ke suling rekorder dengan total tingkat akurasi sebesar 70% dengan data latih diambil dari seorang ahli, dan sistem gagal  mengidentifikasi nada suling sunda ke suling rekorder dengan total 30%. Ketidaksesuaian identifikasi nada diakibatkan jarak ekstrasi ciri antar nada yang berdekatan dan karena suling sunda bisa menggunakan nada rendah, standar, dan tinggi dan untuk penelitian ini hanya nada standar saja yang digunakan dan pada saat pengambilan data uji semua peniup adalah orang awam terhadap meniup suling sehingga kerap terjadi kesalahan pada saat proses pengambilan data uji. AbstractFlute is a musical instrument commonly used by music players or public people. Sundanese flute is a traditional musical instrument from Pasundan that is can hypnotize people who hear it because of the beautiful special tone, flute recorder is a modern musical instrument with sounds like whistle. But not many people know the tone  Sundanese flute can be played using recorder flutes, so that students who follow the dengung (Sundanese art) must bring 2 flutes to school when it coincides with the music class. The identification between tone of Sundanese flute and flute recorder is a system to compare and match frequency same tones between Sundanese flute and recorder flute so the music that is usually played on Sundanese flute can also be played on the flute recorder. This research was made using an algorithm Mel Frequency Cepstral Coefficient (MFCC) to perform feature extraction and algorithm processes Dynamic Time Warping (DTW) is used to identify the time difference of recording. Based on results of research, the system can identifies Sundanese flute tones to refine recorders with total accuracy rate of 70% with training data taken from an expert, and the system fails to identify the tone flute Sunda to flute recorder with a total of 30%. Incompatibility matching tones caused by the distance between adjacent tones and because Sundanese flutes can use low, standard, and high tones and for this study only standard tones are used and when taking test data all blowers are laymen to blow flutes so that errors often occur during the process of taking test data.
Application of VGG Architecture to Detect Korean Syllables Based on Image Text Dewi, Irma Amelia; Shaneva, Amelia
JOIN (Jurnal Online Informatika) Vol 6 No 2 (2021)
Publisher : Department of Informatics, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/join.v6i2.653

Abstract

Korean culture began to spread widely throughout the world, ranging from lifestyle, music, food, and drinks, and there are still many exciting things from this Korean culture. One of the interesting things to learn is to know Korean letters (Hangul), which are non-Latin characters. If the Hangul letters have been learned, the next thing that lay people must learn is the Korean syllables, which are different from the Indonesian syllables. Because of the difficulty of learning Korean syllables, understanding a sentence needed a system to recognize Korean syllables. Therefore, in this study designing a system to acknowledge Korean syllables, the method used is Convolutional Neural Network with VGG architecture. The system performs the process of detecting Korean syllables based on models that have been trained using 72 syllable classes. The tests on 72 Korean syllable classes obtain an average accuracy of 96%, an average precision value of 96%, an average recall value of 100%, and an average F1 score of 98%.