Advancing Rice Blast Disease Classification Through CNN-LSTM Integration

Magnaporthe oryzae, a fungus, causes rice blast disease, which threatens world rice production and, by extension, the food security of a large number of people. In light of the critical need for rapid and precise illness identification, this study presents a new method for multi-classification based on the five stages of rice blast disease namely multiclass (MC) 1 to MC 5 using a hybrid model of Convolutional Neural Networks (CNN) and Long Short-Term Memories (LSTM). An outstanding overall accuracy of 95.92% in differentiating between various illness phases was attained by the model, which was developed to harness both spatial and temporal information. After being tested extensively on a wide range of datasets, the model proved to be quite good at generalizing to new environments, which bodes well for its practical use in farming. It appears that the model can better reflect the complex course of rice blast disease when CNN and LSTMs are combined in a hybrid architecture. This research adds to the field of plant pathology by developing a reliable automated method for disease categorization, even if it can be difficult to differentiate between intermediate phases. In order to ensure the long-term viability of rice production around the world, the results highlight the significance of using cuttingedge technology for proactive disease management.