Detection of Leaf Diseases for Multiple Plants using Convolutional Neural Network

Diseases can be harmful to plant health and affecting their growth. Several leaf diseases control the growth of plants. So, early diagnosis of plant diseases is important in agricultural field to produce a good crop yield. However, current research does not provide an accurate and fast diagnostic tool to ensure the healthy growth of agricultural sector. This paper suggests an in-depth study approach which is based upon a Convolutional Neural Network (CNN) algorithm for detecting the leaf infections in the real-time. In this paper, we have focused on detecting the diseases of multiple plants. Although many existing models are limited to large-scale disease detection, the current model deals with accurate diagnostic detection, which is multidisciplinary detection. The proposed model is designed to address both acquisition speed and accuracy and will be used to detect plant diseases of 10 plants in the real world. The system automatically diagnoses the disease and suggests the pesticide name and contact detail of pesticide supplier to the farmer through SMS/website. In addition to this, this proposed system provides the weather forecasting information to the farmers which is important for them for making the appropriate decisions related to the plant health. The CNN is used to find plant diseases. The proposed system can detect the 21 leaf diseases. The name of diseases which the proposed system can detect are Esca, Early Blight, Late Blight, Bacterial Spot, Red Rot, Black Rot, Leaf Mold etc.