Optimized Feature-Based MSVM Approach for Paddy Leaf Disease Classification using PCA and GWO Method

Nowadays, plants are among the planet's most vital resources for halting global warming. However, the plants are afflicted with several illnesses. Current years have witnessed the start of studies to recognize plant diseases. The key aim of this article is to detect paddy disease. Diseases that can affect paddy at different stages of growth include brown spot disease, blast disease, and blight disease. In this research work, to achieve the best results for identifying paddy leave disease, an Optimized Feature-based MSVM model means a combination of PCA, GWO, and MSVM method used to reduce and optimize the features, and Multi-class Support Vector Machine (SVM) for classification were used. The proposed model has achieved a performance of 95.5%, a recall of 96.59%, a precision of 99.50%, and an F1-score of 98.02% compared with the existing ML methods. The conclusion of the research work has achieved the highest accuracy of 99.50% and F1-score of 98.02%, indicating an excellent balance between minimizing false positives and maximizing the identification of true positives.