Red Chili Leaf Disease Identification via GLCM and SVM Analysis

Red chili plants are highly susceptible to a variety of diseases, posing significant challenges to horticulturists. Among these diseases, jaundice caused by the geminivirus is a common affliction, characterized by yellowing leaves and pronounced curling. Geminivirus infection in red chili plants impedes their growth, resulting in stunted plants that fail to yield fruit. To address this issue, this study focuses on identifying red chili leaf disease. A dataset of 80 red chili leaf images, categorized into two types, is utilized. The images undergo a series of preprocessing steps, including cropping and conversion to grayscale using the Gray Level Co-Occurrence Matrix (GLCM) method. Texture features are then extracted from the red chili leaves using GLCM analysis, employing angles of 0°, 45°, 90°, 135°, and neighboring distances of
$\mathrm{d}=1,2,3,4$
, and 5. Key GLCM feature values such as contrast, homogeneity, energy, correlation, and dissimilarity are derived from this analysis. The subsequent identification stage employs the Support Vector Machine (SVM) algorithm to accurately classify and segregate the data, thereby evaluating the efficacy of the preceding method. The experimental results demonstrated that this approach achieved a remarkable accuracy rate of 98 percent, employing angles of 0°, 45°, 90°, and a neighboring distance of
$\mathrm{d}=5$
.