Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal RABIT: Jurnal Teknologi dan Sistem Informasi Univrab
Amelia Safrida
Universitas Dian Nuswantoro
Computer Science & IT Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
OPTIMASI KLASIFIKASI CITRA ALFABET SISTEM ISYARAT BAHASA INDONESIA (SIBI) MENGGUNAKAN AUGMENTASI DATA DAN FINE-TUNING MOBILENETV2 Amelia Safrida; Edy Mulyanto; Muhammad Naufal
Rabit : Jurnal Teknologi dan Sistem Informasi Univrab Vol 11 No 2 (2026): Juli
Publisher : LPPM Universitas Abdurrab

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36341/rabit.v11i2.8340

Abstract

The Indonesian Sign Language System (SIBI) is a communication medium used by the deaf community in Indonesia. However, classifying SIBI alphabet images presents challenges due to visual similarities between signs for different letters and limited training data variation. This study aims to optimize SIBI alphabet image classification by applying data augmentation and fine-tuning to the MobileNetV2 model. The dataset consists of 7,582 SIBI alphabet images sourced from Kaggle, covering 24 classes (letters A through Y, excluding J and Z). The research process involved preprocessing, data augmentation, dataset splitting into training, validation, and testing sets, model development using MobileNetV2 transfer learning, and performance evaluation based on accuracy, precision, recall, and F1-score. The results show that the model without augmentation achieved 96.75% accuracy, whereas the model with augmentation achieved 97.10% accuracy, accompanied by improvements in precision, recall, and F1-score. These results indicate that data augmentation enhances the model's generalization capabilities, resulting in more accurate and consistent classification. Thus, the combination of data augmentation and MobileNetV2 fine-tuning is effective for SIBI alphabet image classification. McNemar's test revealed a statistically significant difference in model performance following the application of data augmentation (p = 9.6517 × 10⁻¹²).
Co-Authors Edy Mulyanto Muhammad Naufal