<![CDATA[The Grey Level Co-occurrence Matrix (GLCM) is a widely used texture extraction method for Synthetic Aperture Radar (SAR) imagery, although its visual representation is highly dependent on parameter configuration. This study evaluates the sensitivity of GLCM configurations for representing oil palm plantation characteristics using ALOS-2 PALSAR-2 imagery in PTPN III Sei Dadap, Asahan Regency, North Sumatra, Indonesia. After radiometric calibration, speckle filtering, and terrain correction, four GLCM texture measures (contrast, entropy, homogeneity, and variance) were extracted using three moving window sizes (5 × 5, 7 × 7, and 9 × 9 pixels), three quantization levels (8, 32, and 128 gray levels), and two polarization channels (HH and HV), producing 72 texture configurations. Visual interpretation was performed by comparing the texture images with high-resolution optical imagery. Each texture measure produced distinct visual characteristics of oil palm plantations. Among the evaluated measures, variance most clearly delineated plantation blocks, road networks, and intra-canopy variation, while HH polarization consistently provided more informative texture patterns than HV. These findings demonstrate that GLCM parameter configuration significantly influences the visual characterization of oil palm plantations in SAR imagery.]]>
Copyrights © 2026
