Implementation of Pretrained VGG16 Model for Rice Leaf Disease Classification using Image Segmentation

Rice is an agricultural sector that produces rice which one of the staple foods for majority population in Indonesia. In cultivation plants there are also factors affect production and not realized by farmers causing they late handling diagnosing symptoms making decline. Therefore, it necessary to have early diagnosis identify them correctly, quickly accurately. Machine learning classification techniques detect various plant diseases such as plants. There several studies on machine using Convolutional Neural Network with VGG16 model classify leaf Image Segmentation datasets make image becomes a form too complicated analyze. The data used this research Leaf Disease consists 3 classes including Bacterial blight, Brown spot, smut. Then segmentation carried out two techniques, namely threshold k means. augmentation dataset has large varied number training hyperparameter tuning obtained 91.66% accuracy results scenarios k-means dataset.