Deep Learning Based Multi-Classification Model for Rice Disease Detection

To recognise and categorise images of rice diseases, a A convolutional neural network (CNN) model built on deep learning (DL) has been developed. The model is developed for the classification of rice hispa images collected from the Punjab province of India. The entire approach aids in the classification of photos of rice hispa based on the 5 different illness severity levels.. Hence, the successful execution of the model on the collected images dataset resulted in an exceptional binary classification accuracy of 98.86% and 98.6% best accuracy in the case of multi-classification of the collected rice hispa image dataset. Along with this the proposed work also discusses the proposed four-step methodology which helps to conduct the whole process of implementation from scratch.