Tomato Disease Classification using AlexNet
This work employs the AlexNet architecture for tomato diseases detection. The experimental setup is divided into two categories: In the first case a 10-class classifier is trained with healthy (1 class) and diseased (9 classes) leaf images and in the second case 9 different 2-class classifiers are trained with healthy (1 class) and diseased (1 class) leaf image. All cases are tested for three training scenarios. Finally, we point a particular feature of the PlantVillage dataset.
Keywords - Classification, plant, disease, leaves, AlexNet, PlantVillage.