A Comparative Analysis of Deep Learning Models Applied for Disease Classification in Bell Pepper

Abstract

Crop diseases are a major cause of degrading the quality and reducing the number of agricultural products. Hence, there is a strong need for the early diagnosis of the disease. The effectiveness of deep learning techniques in pattern matching and image processing motivated the authors to design an automatic tool for the detection of diseases in bell pepper plants. In this manuscript, the authors present the comparative analysis of different deep learning models applied for plant disease classification. They applied the deep learning models namely VGG16, VGG19, ResNet50, ResNet101, ResNet152, InceptionResNetV2, DenseNet121 on the publicly available dataset of the bell pepper plant. The experimental results prove that the model `DenseNet' requires less training time and gives the highest validation accuracy among all the above-stated models. It achieves a training accuracy of 97.49 <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sub> and the testing accuracy of 96.87% in classifying the bell pepper plants into healthy and diseased categories.