Combining Spatial and Temporal Analysis: A CNN-LSTM Hybrid Model for Maize Disease Classification

In contrast to traditional meta-disease detection approaches, this study proposes a new hybrid model based on CNN and LSTM that is used for classifying maize leaf diseases. The combination of CNNs and LSTMs is probably the best model because it takes advantage of both spatial and temporal aspects to analyze leaf disease symptoms. The model was trained on an augmented dataset of images obtained from maize leaves and showed a high level of accuracy of 97.33%, precision, and recall much exceeding that rendered by the standard CNN models. These findings demonstrate the applicable power of the model in that maize disease identification stands to benefit as well, providing an effective tool for optimizing agricultural disease management and furthering AI research on plant pathology.