Integrating Remote Sensing and Deep Learning for Precision Agriculture in Cinnamon Farming

This study presents a comprehensive system for cinnamon farming that incorporates disease detection, yield prediction, and nutrition level prediction using machine learning models. The system utilizes convolutional neural networks (CNNs) and linear regression to achieve accurate results. For disease detection, the VGG16 CNN model demonstrates superior performance over ResNet-50, achieving an impressive accuracy of 86%. The ANN model achieves a satisfactory yield prediction accuracy of 86.8%, with potential for further enhancements through expanded and refined datasets. In nutrition level prediction, the CNN model achieves 91% accuracy in detecting nutrient deficiencies, while the regression model predicts nutrient levels with 88% accuracy. The combined predictions result in an overall accuracy of 80.1%. Further research and development, along with advancements in data collection and integration with emerging technologies, can enhance the system's accuracy and contribute to the growth and sustainability of the cinnamon farming industry.