Rice Plant Diseases Classification using a New Convolutional Neural Network based Approach

In India, agriculture is a significant source of income and essential for livelihood. The majority of India's land is used to grow rice which is a fundamental component of diet. Plant diseases have a significant impact on rice harvesting which results in significant losses for the agricultural. Plant pathologists are trying to find a precise and trustworthy method to diagnose rice plant diseases. In past few years, machine learning techniques have been proven to be useful in the automated categorization of crop diseases. Presently, deep learning is widely used for disease detection in agriculture. In this paper, a new convolutional neural network (CNN)-based technique is developed to identify four different types of rice plant diseases: brown spot, bacterial blight, blast, and tungro. The proposed CNN architecture used in this study distinguishes diseases of rice plants based on deep features extracted from the five consecutive convolution layers with the activation function RELU. For experimentation of the proposed technique, a rice plant dataset of 5932 images are used. For the training of the CNN model, 4152 rice plant images are used while testing is performed using 890 images of four different classes. The classification technique used in this paper achieves 98.10% accuracy, 96.10% sensitivity, 98.72% specificity, 96.36% precision and 96.21% F1-score. The proposed technique offers better performance than many state of the art techniques.