Detection of Oil Palm Seedling Disease Based on Leaf Images Using the MobileNetV2-CNN Architecture

Detection of Oil Palm Seedling Disease Based on Leaf Images Using the MobileNetV2-CNN Architecture

Plant Disease Detection, MobileNetV2, Convolutional Neural Network, Deep Learning

This study aims to develop and implement a plant disease detection system for oil palm seedlings based on leaf images using the MobileNetV2 architecture, which is based on Convolutional Neural Networks (CNN). The model was trained using a dataset of oil palm leaf images to detect several types of plant diseases. In the experiments, the applied model showed excellent results, with training accuracy increasing from 79% in the first epoch to 96% in the 15 epoch, and validation accuracy also increasing from 89%to 97%. These results demonstrate that the model can effectively detect plant diseases with good generalization ability on unseen data. With stable loss reduction and continuously improving accuracy, this study proves that the MobileNetV2 architecture canbe efficiently used for plant disease detection. The research also highlights the potential integration of the model into anapplication to provide a practical solution in oil palm plantation managementand to support decision-making and improve agricultural outcomes

Ego Oktafanda1, Adyanata Lubis2, Elyandri Prasiwiningrum

https://ijicom.respati.ac.id/index.php/ijicom/article/view/71/73

2025

Fajar Bagus W

PDF

English

Text