Klasifikasi Penyakit pada Daun Kelapa Sawit Menggunakan EfficientNetV2 dan VGG19 Berbasis Website

Abstract

Oil palm is a versatile cash crop that is widely used in the production of edible oils, industrial oils, and biofuels. Indonesia is the world's largest producer of oil palm, making it an important participant in the palm oil business. Nonetheless, there are still errors in identifying oil palm leaf diseases, especially for people who do not have knowledge about the plant. Research was conducted by utilizing the deep learning method with the EfficientNetV2 model to identify leaf diseases of oil palm plants. This research was conducted by performing several stages including dataset collection, data augmentation, model building, data training, and testing. The results showed that the model used achieved a training accuracy rate of 99.74% and a validation accuracy rate of 98.75% at the 30th epoch. Thus, the web-based palm leaf disease classification system developed in this study is expected to be a useful tool in supporting the understanding of palm leaf disease.