Rice Disease and Pest Detection Using Deep Learning

Abstract

In the field of agriculture information, automatic detection and diagnosis of plant disease and pest is highly desirable. Feature extraction technologies play a critical and crucial role in leaf disease detection and diagnostic system. Researches in leaf disease detection have used many different feature detection techniques like color, texture, shape etc. Recently very promising results are found using deep learning in different types of computer vision problems. Now a days deep learning is hot research topic in pattern recognition, machine learning as well as artificial intelligence. Deep neural network-based models can be an effective solution to vegetable pathology. In this research I have proposed a novel rice disease and pest detection model which is based on deep convolutional neural networks (CNN). This model gives a training accuracy of 80.11% with 77.68% training accuracy.