Rice Leaf Detect Net: Machine Learning Framework for Detection and Classification

The production of rice as a globally important crop faces various diseases which reduces both crop yield and economic potential. Traditional disease identification methods need specialized skills because they demand slow performance and lengthy labor alongside tedious work processes. Automatic identification of rice leaf diseases can now be achieved through the application of Deep Learning (DL) and Machine Learning (ML) and Artificial Intelligence (AI) technology. Disease diagnosis at an early stage proves more accurate with assistance from imaging technology which utilizes pattern recognition processes. The proposed work evaluates AI diagnosis systems based on Logistic Regression, SVM, Random Forest, XGBoost, along with other ML classifiers to identify Bacterial Leaf Blight, Brown Spot and Leaf Blast through examination of preprocessing techniques and segmentation methods together with feature extraction approaches for rice disease detection. The research examines their proposed techniques while evaluating their methods for determining F1 scores and accuracy rates and recall levels and precision rates of their results and assessing different approach effectiveness. Advanced models remains a challenge along with dataset restrictions and problems implementing real-time applications as well as methods for upgrading current AI-based rice disease detection systems.