Optimizing Tomato Leaf Disease Identification Using a Hybrid Spatial-Temporal Model and Attention Mechanism

Tomato leaf diseases pose a significant threat to global tomato production, impacting both personal gardening and commercial farming. Early and accurate disease detection is crucial for preventing the spread of diseases and minimizing crop damage. This chapter introduces a novel hybrid spatial-temporal model enhanced with attention mechanisms to address the challenges in tomato leaf disease classification. The model leverages convolutional neural networks (CNNs) for spatial analysis and recurrent neural networks (RNNs) for temporal analysis, enabling it to capture both static and dynamic features of diseases. Attention mechanisms further refine the model's focus on critical regions within leaf images, improving detection accuracy. The proposed method achieves high performance metrics, including an accuracy of 98.63%, sensitivity of 97.37%, and specificity of 97.15%, making it a reliable tool for precision agriculture. The chapter also reviews existing deep learning approaches, highlighting the need for models that can handle real-world variability and complex backgrounds. By integrating spatial, temporal, and attention data, this research offers a comprehensive solution for enhancing tomato crop health and productivity, ultimately contributing to a more sustainable and efficient food supply chain.