Article Per Year (5 Year)
p-Index From 2020 - 2025
0.444
P-Index
This Author published in this journals
All Journal Jurnal Media Elektro : Journal of Electrical Power, Informatics, Telecommunication, Electronics, Computer and Control System
Dwiandiyanta, Bernadectus Yudi
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Energy Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
TRANSFORMING WOVEN IKAT FABRIC: ADVANCED CLASSIFICATION VIA TRANSFER LEARNING AND CONVOLUTIONAL NEURAL NETWORKS Tena, Silvester; Dwiandiyanta, Bernadectus Yudi
Jurnal Media Elektro Vol 12 No 2 (2023): Oktober 2023
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jme.v12i2.12579

Abstract

The woven ikat fabric from Nusa Tenggara Timur is a local wisdom that must be preserved. Due to its vast array of motifs, users often encounter challenges in its recognition. For this study, the TenunIkatNet dataset was employed. One prominent recognition method involves classification based on the motif type and geographical origin. The efficacy of the classification is heavily contingent upon the method of extraction employed. The Convolutional Neural Network (CNN) method is used for feature extraction and classification processes. This research compares the classification performance of the VGG16 baseline model and the proposed model. The proposed model modifies the baseline at the fully connected layer and the training process from the first convolution layer. Incorporating elements such as Global Average Pooling (GAP), Batch Bormalization (BN), and Dropout has proven instrumental in mitigating overfitting. The transfer learning strategy is used for feature extraction and classification because the model has been intelligently trained on a large dataset. The research findings unequivocally indicate that the performance of the modified model supersedes that of the baseline model. Based on the evaluation metrics, the proposed model is superior to the baseline model with precision, recall, accuracy, and F1-score, respectively 98.73%, 98.54%, 98.54%, and 98.53%
FAST IMAGE RETRIEVAL BERBASIS LOCALITY SENSITIVE HASHING DAN CONVOLUTIONAL NEURAL NETWORK Tena, Silvester; Dwiandiyanta, Bernadectus Yudi; Ina, Wenefrida Tulit
Jurnal Media Elektro Vol 13 No 1 (2024): April 2024
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jme.v13i1.15137

Abstract

Image retrieval systems with a fast search process are still challenging for researchers. Fast search methods are one of the most important parts of image retrieval. One of the techniques used is reducing feature dimensions using the Locality Sensitivity Hashing (LSH) method. Apart from that, feature types and image extraction methods are selected. Image feature extraction uses the Convolutional Neural Network (CNN) method in this research. Measuring similarity using the Hamming Distance (HD) and Euclidean Distance (ED) methods. The datasets used are TenunIkatNet and Batik300. The LSH method forms a hash table as a bucket to group similar images based on probability and in the form of binary code. The research results show that the LSH+HD+ED method provides faster search results than ED. The image retrieval time for the LSH+HD+ED and ED methods is 0.252 seconds and 4.5 seconds, respectively, for the TenunIkatNet dataset. Meanwhile, the Batik300 dataset is 0.03 seconds and 0.9 seconds. Using the LSH method is very effective for large datasets. Retrieval accuracy using the LSH+HD+ED method was 99.705% and 84% for the TenunIkatNet and Batik300 datasets, respectively. Meanwhile, the ED method produces 94.17% and 82% retrieval accuracy, respectively.
Co-Authors Silvester Tena Wenefrida T Ina