Identifying two of tomatoes leaf viruses using support vector machine

Abstract

One of the most harmful viruses is Tomato yellow leaf curl virus (TYLCV), which is widespread over the world with tomato yellow leaf curl disease (TYLCD). It causes some symptoms to tomato leaf such as upward curling and yellowing. This paper introduces an efficient approach to detect and identify infected tomato leaves with these two viruses. The proposed approach consists of four main phases, namely pre-processing, image segmentation, feature extraction, and classification phases. Each input image is segmented and descriptor created for each segment. Some geometric measurements are employed to identify an optimal feature subset. Support vector machine (SVM) algorithm with different kernel functions is used for classification. The datasets of a total of 200 infected tomato leaf images with TSWV and TYLCV were used for both training and testing phase. N-fold cross-validation technique is used to evaluate the performance of the presented approach. Experimental results showed that the proposed classification approach obtained accuracy of 90% in average and 92% based on the quadratic kernel function.