Claim Missing Document
Check
Articles

Found 3 Documents
Search

Analisis Algoritma Convolutional Neural Network (CNN) untuk Pengenalan Pola Tangan Berdasarkan Citra Garis Telapak Tangan Fazila Nazifa Edilia; Lailan Sofinah Harahap; Andita Utami; Zianah Nafisah Simbolon
Cosmic Jurnal Teknik Vol 3 No 1 (2026): Februari
Publisher : Ali Institute or Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosmic.v3i1.1495

Abstract

Individual identification is an important aspect of security and biometric authentication systems. One biometric characteristic with high potential is palm print patterns, as each individual has a unique pattern that is difficult to forge. Therefore, it is necessary to develop an individual identification system based on palm print images using a pattern approach with the Convolutional Neural Network (CNN) method. Test results show that this method is capable of identifying individuals with a fairly high level of accuracy, depending on the image quality and training parameters as well as the network architecture used.
Perancangan dan Implementasi Sistem Informasi Ekstrakurikuler Sekolah di SMP Harapan 3 Medan Menggunakan Model Incremental Tengku Tanzil Azhari Risky; Fahrialdy Febriansyah Putra; Jelita Rahmah Zebua; Qisti Azraladiba Batubara; Puan Syaharani Sitorus; Zianah Nafisah Simbolon; Fazila Nazifa Edilia; Rio Rinaldi; Dhafa Hibrizi Sitorus; Ilka Zufria
DEVICE : JOURNAL OF INFORMATION SYSTEM, COMPUTER SCIENCE AND INFORMATION TECHNOLOGY Vol. 6 No. 2: DESEMBER 2025
Publisher : Universitas Dharmawangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46576/device.v6i2.7121

Abstract

Penelitian ini bertujuan untuk merancang dan mengimplementasikan sistem informasi ekstrakurikuler berbasis web di SMP Harapan 3 Medan. Sistem ini dikembangkan untuk mengatasi kendala pengelolaan ekstrakurikuler secara manual yang sering menyebabkan data hilang, jadwal tumpang tindih, dan informasi tidak tersampaikan dengan baik. Metode incremental digunakan agar sistem dapat dibangun secara bertahap sesuai kebutuhan pengguna. Fitur utama meliputi pendaftaran online, manajemen data, serta pemberitahuan ekstrakurikuler. Hasil implementasi menunjukkan peningkatan efisiensi pengelolaan dan akses informasi real-time bagi siswa dan pembina secara terstruktur dan terintegrasi.
Web-Based Binary Image Compression Using Thresholding and Run-Length Encoding Adli Rahman Harun Harahap; Fazila Nazifa Edilia; Andita Utami; Zianah Nafisah Simbolon; Faiz Akhtar Aqila
Journal of Information Technology and Computer System Vol. 2 No. 1 (2026): June
Publisher : CV. Multimedia Teknologi Kreatif

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.65230/jitcos.v2i1.116

Abstract

This study implements the Run-Length Encoding (RLE) algorithm for binary image compression in a web-based environment using thresholding as a preprocessing stage. The system was developed using JavaScript and the HTML5 Canvas API to perform grayscale conversion, thresholding, and image compression directly within a web browser. Thresholding was applied to transform RGB images into binary images with simpler and more repetitive pixel patterns suitable for RLE compression. The compression process encodes consecutive pixels with identical values into tuples consisting of run lengths and pixel values. Experiments were conducted using a test image and five threshold values (50,75,100,127,150) to evaluate their influence on compression performance. The results show that threshold selection affects pixel redundancy, the number of generated tuples, and compression efficiency. The highest performance was achieved at a threshold value of 127, resulting in storage savings of 63.6% and a compression ratio of 2.75:1. The findings indicate that RLE is effective for compressing binary images with repetitive pixel patterns. Furthermore, the RLE compression stage is lossless with respect to the binary image generated after thresholding, whereas the preprocessing stage from RGB to binary is lossy because color and intensity information are discarded.