Deep Learning to Detect Plant Diseases

Economy of any nation shares a major part with the agriculture and crop production. Good yield is badly impacted by the diseases in plants and crops. Due to involvement of manual inspection on majority, poses a challenge to identify the plant diseases and in turn the crop yield is reduced, or quality is affected. Monitoring plants and crops spread over a large area is tedious task for the farmers or cultivators. Sometimes, the disease may not be known to the farmer. The present paper presents a system involving a standard smartphone to predict the plant diseases using machine learning approach. The proposed system collects data, as plant disease images, and that dataset is used to detect various diseases of plants and crop. It potentially benefits the cultivators as it is capable to detect the diseases without minimal human intervention with prompt results. Further, the proposed technique helps in detecting diseases during its early stage to safeguard the yield. Neural network-based model is trained to detect plant diseases and the crop types. During test results, the model achieves an accuracy of 96.78% in detecting diseases which is of significant use to the cultivators. Further, the system recommends the possible pesticides to use in every category of the disease.