Automated Rice Panicle Blast Disease Severity Classification Using Hybrid CNN-LSTM Model

The plant species known as Oryza sativa, commonly called rice, plays a crucial role as a staple crop, providing sustenance for a significant proportion of the world's population. Nevertheless, the productivity of rice crops is consistently jeopardized by many illnesses, among which rice panicle blast disease emerges as a significant apprehension. The fungal disease produced by Magnaporthe oryzae has the potential to cause substantial reductions in crop productivity, hence presenting a considerable obstacle to the world food supply. Accurately assessing the severity of rice panicle blast disease is essential for the timely adoption of control measures. To address this challenge, we offer a novel hybrid DL model that combines Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) networks. Spatial and temporal data are used in this model to reliably assess disease severity. The research used an image dataset of over 15,000. These photographs represented patients at varying stages of sickness. We have achieved a remarkable overall accuracy rate of 94.06% by using careful data preparation approaches, developing models, and engaging in rigorous training. Precision, recall, and F1-score measures were also analysed for each severity level to provide a full assessment of performance. The confusion matrix obtained provides a visual representation of the model's efficacy in accurately categorizing different levels of disease severity, hence providing valuable insights into its practical applicability. In addition, our thorough comparison with current leading methods underscores the distinct contributions made by our hybrid model in enhancing the classification of illness severity. This study showcases the capabilities of DL in the field of agricultural disease control and emphasizes the importance of automated assessment of disease severity. The potential of our approach to significantly transform crop management practices, improve food security, and stimulate further advancements in agricultural technology is noteworthy.