Fusion of Thepade SBTC and GLCM Features for Improved Rice Leaf Disease Detection

The agricultural sector is negatively impacted by plant diseases. This illness reduces yield productivity and causes farmers to lose a lot of money. It is critical to identify illnesses in plants in order to conserve agricultural crop output and reduce pesticide use. The study presents image classification using the n-ary Thepade's Sorted Block Truncation Coding (TSBTC) with the Gray Level Cooccurrence Matrix(GLCM) for rice leaf disease detection. The database used here is “Rice Leaf Disease Identification,” downloaded from Kaggle, containing 3334 images with four rice leaf disease categories, namely hispa, leafblast, brownspot, and healthy. Various degrees of feature vectors are generated by adopting a fusion of TSBTC and GLCM. The classification algorithms belonging to different families are used for image classification from Functions, Trees, and Bayes families. The performance of the proposed method is examined using classification accuracy. The ensemble of random forest, random tree, and J48 classifier using a fusion of Thepade's SBTC and GLCM features has performed better in comparison to other methods studied.