Classifying Date Palm Tree Diseases Using Machine Learning

Abstract

One of Jordan's most significant agricultural crops is the date palm tree. The high level of interest in date palm farming is a result of the crop's superior economic viability when compared to other agricultural crops; Jordan's annual investments in this sector are expected to be more than $500 million. Recently, the Jordanian ministry of agriculture reported that many trees are vulnerable to damage because of several diseases related to date palms. In this study, the convolutional neural network (CNN) and support vector machine (SVM) algorithms are used to detect and classify date palm diseases. Four common diseases are considered in this paper: bacterial blight, brown spots, leaf smut, and white scales. The palm farms in the northern Jordan Valley, Kaggle, the National Center for Agricultural Research, and other sources provided the dataset used in this study. The experimental results show that CNN is effective mechanism for detecting and classifying Date Palm disease especially when large dataset is used in training the algorithm.