Disease Identification in paddy leaves using CNN based Deep Learning

Rice is the staple food for most of the tropical and subtropical countries of the world. This entails large fields of paddy spanning hectares, whose maintenance and care becomes a tedious task for the farmers. The caretakers aren’t able to identify certain types of diseases and aren’t able complete the tedious task of crop care in such a short span. Thus, motivated by this arduous exercise, this paper suggests a solution for quick classification of paddy into diseased or healthy plants.If the plant is diseased, the area affected is identified. The image dataset used for this module is obtained from public platforms and consists of 3500 images of healthy and diseased paddy leaves. The classification module is created using convolutional neural network layers and provides accuracy of upto nearly 70%. This paper reviews existing works on image classification using CNN, introduces a new module for paddy disease classification and gives a reference for future work on the subject.