Plant Disease Classification and Detection using CNN

There have been various studies and research done on detecting plant disease. This work is intended to identify the plant disease using Deep Learning which can be used as a defense mechanism against the disease. The datasets were gathered from the internet source and captured from the camera sensor and segregated into different plant diseases. This work mainly concentrates on tomato leaf disease. To predict the state of the leaf, Convolution Neural Network (CNN) is used which consists of different layers. The training is performed on the prepared dataset. Using training data, datasets are tested and the output will be predicted with optimal accuracy. In the classification of different plant diseases, CNN has achieved prominent success. User Interface (UI) has been created for the detection of different leaf diseases namely Bacterial Spot, Leaf Mold, Yellow Curl Leaf Defect, Spectoria Leaf Spots, and Healthy Leaf. Once the disease is detected the UI will help to find the remedy for a particular leaf disease. Accordingly, farmers can make use of the remedy to prevent the disease. If there is no disease found in the training data then the plant is healthy. Optimal accuracy of 99% is achieved using the proposed methodology. The farmer community will be benefited by using this work to take precautionary measures in the early stage to improve the yield.