One of the most vital nutrients for humans on the planet is rice. India and China are two of the most rice-dependent nations on the planet. A variety of factors influence the production of this crop, including soil, water supply, pesticides employed, time period, and disease infection. One of the most important factors impacting rice yield and quality is rice plant disease (RPD). Farmers are always faced with the issue of recognising the type of rice plant disease and adopting appropriate corrective treatment. The most prevalent diseases that affect rice plants are Bacterial Leaf Blight (BLB), Brown Spot (BS), and Leaf Smut (LS). It is particularly difficult to diagnose this disease since the contaminated leaf must be processed by the human eye. In this chapter, we used machine learning techniques to describe and classify the RPD. To collect data on infected rice plants, we used the UCI Machine Learning repository. There are 120 photos of contaminated rice plants in the data set, containing 40 BLB images, 40 BS images, and 40 LS images. In the studies, decision tree-based machine learning algorithms RandomForest, REPTree, and J48 were used. To extract numerical characteristics from the infected photos, we used ColorLayoutFilter, which is provided by WEKA. 65 percent of the data is used for training and 35 percent is used for testing in the experimental analysis. According to the trials, the Random Forest algorithm works remarkably well in predicting RPD.

Author (S) Details

R. Sahith
CSE, CVR College of Engineering, Hyderabad, India.

P. Vijaya Pal Reddy
CSE, Matrusri Engineering College, Hyderabad, India.

Satyanarayana Nimmala
CSE, CVR College of Engineering, Hyderabad, India.

View Book :- https://stm.bookpi.org/AAER-V16/article/view/1990

Leave a Reply

Your email address will not be published. Required fields are marked *