Abstract: The agricultural sector remains a key contributor to the Indian economy, representing about 20% of gross domestic product (GDP). Plant diseases are the major cause of low agricultural productivity and cause heavy losses in the country’s economy. Mostly the farmers encounter difficulties in controlling and detecting the plant diseases. Therefore, early detection can mitigate the severity of diseases and protect crops. This paper focuses on supervised machine learning techniques such as linear discriminant analysis (LDA), Random Forest (RF), Extreme Gradient Boosting (XGB), Support Vector Machine (SVM) algorithms applied to tomato leaf disease detection was carried out with the help of the images of the tomato plant leafs. The aforementioned machine learning techniques are analyzed and compared in order to select the best suitable model with the highest accuracy for tomato leaf disease detection prediction. Comparative experiment results show that Support Vector Machine (SVM) algorithm results achieved with the highest accuracy of 81% as compared to the rest of the classification techniques. This work shows aforementioned model can be used by the farmers to encounter difficulties in controlling and detecting the tomato leaf disease as a preventive measure.

Machine learning · Tomato leaf disease detection · Random Forest (RF), linear discriminant analysis (LDA), Extreme Gradient Boosting (XGB), Support Vector Machine (SVM)