Shielding Lemongrass: An Integrated Multimodal Strategy for Disease Detection with CNNs and Random Forests

Diseases that affect lemongrass can have a major influence on crop output, thus prompt and correct diagnosis is required. In the article, the Scientist introduces a novel method using Convolutional Neural Networks and Random Forest algorithms to forecast six prevalent illnesses in lemongrass, which is including Yellow Mosaic Disease, Rust, Leaf Blight, Sooty Mold, Red Root, & Damping-Off. A thorough evaluation of their model produced incredibly positive findings. The model successfully predicted all diseases with an accuracy of 99% for each class, with precision values ranging from 94.68% to 97.67%, recall values from 95.96% to 97.09%, with F1-scores from 95.87% to 97.32%. With respective values of about 96.69%, the macro, micro, and weighted averages confirm the model's consistency even further. The study demonstrates the considerable potential of combining CNN and Random Forest for effective forecasting of diseases in lemongrass, which is paving the path for more sophisticated, tech-driven approaches to agricultural disease control.