DETEKSI PENYAKIT RUMPUT LAUT DENGAN RESIDUAL NEURAL NETWORK

Abstract

This research aims to detect seaweed diseases using the Residual Neural Network (ResNet) deep learning model. Seaweed, or Thallus, is a crucial fishery commodity in Indonesia, but it is often threatened by diseases such as Ice-ice and Bulu Kucing, which are challenging to distinguish visually. The dataset used in this study consists of images of healthy and diseased seaweed, which undergo preprocessing steps like resizing, augmentation, and data splitting. The ResNet model is trained on this processed data and evaluated using a Confusion Matrix, achieving an accuracy of 96.78% and a validation accuracy of 99.68%. These results demonstrate that ResNet has significant potential in detecting seaweed diseases, which can contribute to increasing productivity and improving the welfare of seaweed farmers.