Detection of Plant Leaf Disease Using Deep Learning and Convolutional Neural Networks

When plants and crops are attacked by pests, it affects the country agricultural production. Farmers and professionals usually look at plants with the naked eye to detect and identify diseases. However, this method can be time consuming, expensive and error prone. Automatic detection using image processing technology provides fast and accurate results. This paper describes a novel approach to develop a model for plant disease detection based on leaf image classification using convolutional network. Advancement in computer vision offer opportunities to improve precision crop protection practices and expand the market for computer vision products in the precision agriculture sector. The new training methods and methodology used make it quick and easy to implement the system in practise. All the mandatory steps required to implement this disease detection model are fully explained throughout the document. First, collect images to create an agronomist-rated database and create a deep learning framework to perform deep CNN training. This method paper is a novel approach to plant disease detection using a deep convolutional neural network trained and fine-tuned to accurately fit a database of individually collected plant leaves for various plant diseases