Disease Detection in Tomato Leaves using Machine Learning and Statistical Feature Fusion

Tomato is a domestic plant that is predominantly used worldwide and less tolerant to diseases. Among the diseases that affect tomato plants, often the type and cause of the disease is unknown to the farmer. Accurate plant disease diagnosis is critical for increasing crop yield, monitoring plant health, and providing appropriate treatment at an earlier stage. Therefore, this work aims at developing a system to identify the type of disease that affects the tomato leaves at an early stage. In the proposed system, Gray Level Co-occurrence Matrix (GLCM) and Local Binary Pattern (LBP) texture-based features are used to identify the type of disease by means of a Support Vector Machine (SVM) classifier. To improve the diagnostic accuracy of disease detection system, fusion of two different texture features is performed using Principal Component Analysis (PCA) and Canonical Correlation Analysis (CCA). A publicly available tomato leaves dataset is used to evaluate the proposed system. A considerable improvement in relevant performance metrics is observed in CCA based feature fusion. Henceforth the proposed system could be used as an aid to farmers and plant breeders to identify the type of disease and provide the necessary treatment.