Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Journal of Data Insights
Indah Manfaati Nur
Statistics Study Program, Universitas Muhammadiyah Semarang
Computer Science & IT Mathematics

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

Found 1 Documents
Search

 1 
EfficientNetV2-B0 Hyperparameter Tuning for Rice Leaf Disease Classification in Malang Regency: Algoritma EfficientNetV2-B0 dengan Hiperparameter Tuning untuk Klasifikasi Penyakit Daun Padi di Kabupaten Malang. Mohammad Idhom; Trimono Trimono; Indah Manfaati Nur
Journal of Data Insights Vol 4 No 1 (2026): Journal of Data Insights
Publisher : Department of Sains Data UNIMUS Universitas Muhammadiyah Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26714/jodi.v4i1.1349

Abstract

Rice is a strategic food commodity that plays a crucial role in maintaining food security; however, leaf diseases can reduce productivity and are difficult to identify manually. This study aims to develop a rice leaf disease classification model using EfficientNetV2-B0 through transfer learning and hyperparameter optimization. The research dataset consists of Kaggle images and images collected directly from rice fields in Malang Regency, East Java, especially in Lawang region. The model was developed to classify six categories: blast, brown spot, leaf blight, tungro, healthy leaves, and non-rice images. The research process included image preprocessing, data augmentation, model development, comparison of five optimizers, and classification performance evaluation. Experimental results show that Adam achieved the highest validation accuracy (94.24%), outperforming Adamax, Nadam, SGD, and RMSprop. In the final evaluation, the model achieved 95% accuracy. We also used Grad-CAM visualization to identify the specific image regions that contributed to the model's predictions. The findings suggest that EfficientNetV2-B0 can be applied for automated classification of rice leaf diseases. Nevertheless, the model's ability to recognize non-rice images requires further improvement to support more reliable implementation in smart agriculture.
Co-Authors Mohammad Idhom Trimono, Trimono