HomeUE Research Bulletinvol. 21 no. 1 (2019)

Application of Support Vector Machine in Corn Disease Detection

Sherlyn B. Avendaño



This study focuses on the application of Support Vector Machine (SVM) in the detection of two of the most common corn leaf diseases in the Philippines. SVM is primarily a method that performs classification tasks by constructing hyperplanes in a multidimensional space that separates cases of different class labels. The properties of SVM make it a suitable algorithm that is very easy to apply when there are only two class labels or groups to be classified as in the case of this study which deals with the detection of the two most common corn leaf diseases, namely, northern corn leaf blight and corn leaf rust. In this study, the corn leaf is evaluated first as “other classes” or “unhealthy.” “Other classes” pertains to samples which are either healthy or with no indications of diseases as well as samples with indications of diseases not covered by this study, while the “unhealthy” category pertains to those samples exhibiting indications of leaf blight and leaf rust, before they are fed to the disease detection stage. Results of testing show that 92.73% of the total corn leaf samples were accurately classified by the system.