Coffee Leaf Diseases Classification and the Effect of Fine-tuning on Deep Convolutional Neural Networks

This research proposes a method for the automatic diagnosis and classification of leaf diseases in Kenyan Arabica coffee leaves. We trained Deep Convolution Neural Learning models on JMuBEN2 obtained from Mendeley data public access to determine whether particular image contains Phoma, Cercospora, or Rust. The proposed this work were ResNet50, Densenet-121, VGG19 architectures, all are well-known models. They using transfer learning fine-tuning their respective outputs compared based these methods training. After training dataset aforementioned models, Densenet-121 model was superior others gaining an accuracy 95.44% after 99.36% model.