Leveraging deep learning algorithms for classification of tomato leaf diseases

Abstract

Tomato leaf disease has a significant impact on tomato yield. Identifying agricultural diseases is critical for the agricultural sector. Detecting and classifying different tomato diseases is often labor intensive and cumbersome. It is difficult to discover subtle distinguishing traits in different tomatoes disease images, which makes the task of visual categorization of the tomato leaf, based images both challenging and ambiguous. This causes huge scope destruction of harvests, diminishes development and ultimately prompts monetary loss of farmers. Septoria leaf spot, late blight, the bacterial spot, and yellow curved leaf diseases affect the crop quality of tomatoes are a few examples. The implementation's most important difficulty was deciding on a deep learning architecture.Therefore, to address this problem, our workexploits few of the widely used deep learning-based modelslike ResNet50 and InceptionV3Through qualitative analysis, the significance of the deep learning models has been highlighted.