Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Register: Jurnal Ilmiah Teknologi Sistem Informasi
Tri Wahyuningrum, Dr. Rima
Unknown Affiliation
Computer Science & IT

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

Found 1 Documents
Search

 1 
Classification of Betel Leaf Diseases Based on Convolutional Neural Network to Increase Production Herbal Spice Materials Tri Wahyuningrum, Dr. Rima; Hamed Ayani, Irham; Bauravindah, Achmad; Siradjuddin, Indah Agustien; Faradisa, Irmalia Suryani
Register: Jurnal Ilmiah Teknologi Sistem Informasi Vol 11 No 1 (2025): January
Publisher : Information Systems - Universitas Pesantren Tinggi Darul Ulum

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26594/register.v11i1.4653

Abstract

Traditional medicine is the practice of utilizing medicinal plants to treat various illnesses, passed down from generation to generation. In Indonesia, there are various traditional medicines, one of which is using green betel leaves. One part of the green betel plant that is commonly attacked by pests is the leaf. The Convolutional Neural Network (CNN) method is a very common method used for image classification because this method produces the highest accuracy in classification and pattern recognition. This research uses data totaling 4000 images which are divided into four classes: healthy green betel leaves, anthracnose green betel leaves, bacterial spot betel leaves, and healthy red betel leaves. Detecting the disease type facilitates farmers in acknowledging the necessary measures required to provide treatment. Therefore, this study utilizes the benefits of the CNN approach, specifically its capability to conduct precise object detection and classification in image data, to minimize the widespread of disease. The CNN architectures implemented are DenseNet201, EfficientNetB3V2, InceptionResNetV2, MobileNetV2 and XceptionResnet50V2. Based on our research, the InceptionResNetV2 model achieved the highest performance with an accuracy of 86.0%, loss of 0.3880, and ROC of 98.0%. In the other hand, the MobileNetV2 and EfficientNetV2B3 models suffered from overfitting and underfitting and the models failed to classify betel leaf diseases.
Co-Authors Bauravindah, Achmad Hamed Ayani, Irham Irmalia Suryani Faradisa Siradjuddin, Indah Agustien