Hybrid CNN-LSTM Techniques for Plant Leaf Disease Detection and Farmer-Based Agricultural Practice Analysis in Western Tamil Nadu

Plant diseases greatly affect crop productivity and agricultural sustainability, making early and accurate disease detection essential. This study investigates hybrid deep learning techniques for plant leaf disease detection and classification, focusing on combining Convolutional Neural Networks and Long Short-Term Memory networks. A review of existing studies revealed that the CNN-LSTM hybrid model achieved over 98% accuracy, outperforming several other hybrid approaches. In addition, a survey of 140 farmers from the western districts of Tamil Nadu analyzed farming practices, disease identification methods, and management techniques. The findings indicate that younger farmers are gradually adopting modern agricultural technologies, while older farmers continue to rely on traditional practices. The study highlights the effectiveness of CNN-LSTM models for smart agriculture and emphasizes the importance of farmer awareness and technology adoption for sustainable crop management.