Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Khazanah Informatika: Jurnal Ilmu Komputer dan Informatika ILKOM Jurnal Ilmiah
Dyah Aruming Tyas
Universitas Gadjah Mada
Computer Science & IT

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

Found 2 Documents
Search

 1 
CNN Ensemble Learning Method for Transfer learning: A Review Yudha Islami Sulistya; Elsi Titasari Br Bangun; Dyah Aruming Tyas
ILKOM Jurnal Ilmiah Vol 15, No 1 (2023)
Publisher : Prodi Teknik Informatika FIK Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v15i1.1541.45-63

Abstract

This  study provides a review of CNN's ensemble learning method for transfer learning by highlighting sections such as review studies, datasets, pre-trained models, transfer learning, ensemble learning, and performance. The results indicate that the trend of ensemble learning, transfer learning ensemble, and transfer learning is growing every year. In 2022, there will be 35 papers reviewed related to this topic in this study. Some datasets contain apparent information starting from the dataset name, total data points, dataset splitting, target dataset availability, and type classification. ResNet-50, VGG-16, InceptionV3, and VGG-19 are used in most papers as pre-trained models and transfer learning processes. 50 (90.1%) papers use ensemble learning, and 5 (9.1%) do without ensemble learning. The reviewed paper summarizes several performance measurements, including accuracy, precision, recall, f1-score, sensitivity, specificity, training accuracy, validation accuracy, test accuracy, training losses, validation losses, test losses, training time, and AUC, DSC. In the last section, 49 papers produce the best model performance using the proposed model, and 6 other papers use DenseNet, DeQueezeNet, Extended Yager Model, InceptionV3, and ResNet-152.
Contact Lens Detection Using Domain Specific BSIF and Discrete Wavelet Transform Muhamad Ilham Aji Vachroni; Raden Sumiharto; Dyah Aruming Tyas
Khazanah Informatika Vol. 9 No. 2 October 2023
Publisher : Department of Informatics, Universitas Muhammadiyah Surakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23917/khif.v9i2.20084

Abstract

Iris is one of the reliable biometrics because it has a texture that rich properties and the texture is not changeable lifetime. Iris recognition has drawbacks in the matching process when using contact lenses. Contact lens can changes in the texture of the iris, which can reduce the accuracy of recognition. Therefore, a system is needed to detect contact lenses while someone is detected using contact lens, the system can reject the registration or authentication process. Methods used to detect contact lenses are Domain Specific Binarized Statistical Image Feature (BSIF) and Discrete Wavelet Transform (DWT) for feature extraction. Both methods are fused and modeled using the Support Vector Machine (SVM). Based on the test results, the most optimal kernel is 5x5 12bit. Using the kernel, the accuracy and f1 score obtained 99.1%. In the experiments conducted, this research applies Principal Component Analysis (PCA) to reduce features. However, the role of PCA does not affect the performance of the model. The best model tested with real life data, the Pocophone f1 smartphone and CCTV were used to take pictures of the eyes. The Result 6 experiments wich are 4 without contact lenses and 2 wearing contact lenses, there are only 2 detected correctly. This is because the ability of the images taken from the Poco F1 and CCTV have low resolution.
Co-Authors Elsi Titasari Br Bangun Muhamad Ilham Aji Vachroni Raden Sumiharto Yudha Islami Sulistya