Leaf Disease Classification in Bell Pepper Plant using VGGNet

Abstract

In the era of artificial intelligence, deep learning, and computer vision play a vital role in leaf-based disease identification and categorization. Leaf diseases are the most dangerous calamity that has direct detrimental effects on farmers’ lives, and consequently on gross yield production and the world economy. Nutritious food for all is a great challenge faced by the farmer and agricultural research community. Bell peppers can be categorized as fruit or vegetable that is universally available and full of various nutrients like carbs, vitamins, and fat. Leaves of bell pepper plants infected by bacterial spot diseases affect their yield significantly. The aim of this study is to classify bacterial spots and healthy images of bell peppers’ leaf images taken from the PlantVillage dataset using CNN-based pre-trained architecture. Two CNN architectures, i.e., VGG16 and VGG19 are applied through transfer learning in the binary classification of leaf-based disease. A total of 2475 images are used for training, validation, and testing purposes, with 1478 healthy images and 997 images with bacterial disease spots. Although both VGG16 and VGG19 achieved good performances, VGG16 architecture performs slightly better than VGG19.