Enhanced CNN Model Integrated with Generative AI for Real-Time Pest and Disease Detection in Rice Farming

Rice is a major food to the global food supply, rice production aspects are an important problems from diseases and pests that makes a significant lose in profit. The crop health maintenance can improve farmers' livelihoods, if the primary and specific identification of these issues were addressed earlier. In this research an Enhanced Convolutional Neural Network (ECNN) algorithm wad proposed for rice plant disease diagnosis using advances in Artificial Intelligence and Artificial Neural Network methodologies. It is an integrated mobile technology; the outcome of the research enables farmers to upload photos in the mobile application for real-time diagnosis and to get insights about the pest attacks. The proposed algorithm was enhanced by adding a layer to address overfitting, environmental variability, and gradient issues, the proposed Enhanced CNN(ECNN) grows on traditional CNN designs. The model guarantees strength in a range of perspectives by simulating a range of real-world scenarios applying generative AI tools and strong data augmentation techniques. Existing datasets were used for training data, then enhanced with generated and field-collected images. Performance metrics helped significantly to improve the results for proposed methodology. This ECNN model, effectively diagnoses the crop disease and pests in challenging environments including dim lighting and complex backgrounds, with a validation accuracy of 92.5%. Integration of ANN components improve the model's capability to simplify through diverse pest and disease types.