Sugarcane Leaf Disease Classification Using Hybrid CNN and RNN

Deep learning methods have shown promise in automating agricultural disease diagnosis. In this paper, a comparison of a hybrid Convolutional Neural Network (CNN) model and a hybrid Recurrent Neural Network (RNN) model for identifying leaf diseases to determine which approach performs better in terms of accuracy and efficiency. The hybrid CNN, integrating EfficientNet and DenseNet, excels in image-based tasks, while the hybrid RNN, combining SimpleRNN and LSTM, is suited for sequential data processing. The objective is to assess these models based on performance metrics, computational complexity, generalization, and application suitability. Empirical testing shows the hybrid CNN achieved 94.8% accuracy and efficiency, outperforming the hybrid RNN's 36.4% accuracy and 4% efficiency. The results provide insights into each model's strengths and limitations, aiding decision-making for data analysis tasks. The choice between hybrid CNN and hybrid RNN models depends on data type and task requirements.