Detection of Rice Leaf Diseases Using Convolutional Neural Network

Abstract

Rice is one of the most important crops grown in India, and it is afflicted by a number of diseases which pose a significant menace to global food security. Furthermore, farmers find it challenging to manually recognize these infections with their naked eyes. The proposed work uses rice leaf images for disease classification since early diagnosis of rice leaf disease is critical. Therefore, a collection of various rice leaf disease images is acquired and trained on the best model Convolutional Neural Network (CNN), and then new rice leaf images are tested using the weights obtained from the CNN. Transfer learning is performed using the Inception v3 model to obtain the ideal possible weights for testing and validation in the proposed approach. The model's accuracy is 94.48 percent, and the results indicate that our developed model outperforms existing models in terms of detection performance.