Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Teknologi Dan Sistem Informasi Bisnis
Pranatha, Made Doddy
Unknown Affiliation
Computer Science & IT Electrical & Electronics Engineering

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

Found 1 Documents
Search

 1 
Implementasi Arsitektural Resnet-34 Dalam Klasifikasi Gambar Penyakit Pada Daun Kentang Pranatha, Made Doddy; Maricar, M Azman; Setiawan, Gede Herdian
Jurnal Teknologi Dan Sistem Informasi Bisnis Vol 6 No 3 (2024): Juli 2024
Publisher : Prodi Sistem Informasi Universitas Dharma Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47233/jteksis.v6i3.1431

Abstract

This research develops and implements an image classification method using the Residual Network (ResNet) architecture to identify potato leaf diseases, achieving an accuracy of around 97%. The dataset used consists of 2152 potato leaf images, categorized into three classes: early blight, late blight, and healthy. The selected model is ResNet-50, known for its ability to address the vanishing gradient problem, allowing for the training of very deep networks. The model training process involves data augmentation to enhance dataset diversity and prevent overfitting. Additionally, hyperparameter optimization was performed to maximize the model's performance. Evaluation of the model shows that ResNet-50 can achieve an accuracy of approximately 97% on the test data, indicating the model's high capability in accurately recognizing and classifying the condition of potato leaves. These results demonstrate the significant potential of using ResNet in plant disease image classification applications, which is crucial for decision-making in agricultural management. This research underscores the importance of deep network architectures and data augmentation techniques in improving the performance of deep learning models.
Co-Authors Maricar, M Azman Setiawan, Gede Herdian