Plant Leaves Disease Detection (Efficient Net)

Abstract

The Indian economy heavily depends on agricultural productivity. One of the reasons that disease detection in plants is crucial is because disease in plants is common in the agricultural sector. If sufficient care is not taken here, serious consequences result on plants, as a result of which the corresponding product quality and quantity the impact on productivity. Finding plant diseases using a useful automatic technique is one that minimizes a lot of work of monitoring in large agricultural farms, as well as at the very beginning itself it recognizes disease symptoms on plant leaves at a very early stage. This paper presents a solution using machine learning with help of EfficientNetB3 model. A public dataset of 87K rgb images of healthy and diseased crop leaves is used, which is categorized into 38 different classes. An EfficientNetB3 model is trained to identify healthy and disease plant leaves.