Comparasing Landuse/Landcover Classification Using Pixel and Object Based Random Forest and SVM Algorithm on Sentinel-2 Imagery Over Enrekang Region

Nurfadila JS (Hasanuddin University)
Rismaneswati Rismaneswati (Hasanuddin University)
Syaeful Rahmat (Hasanuddin University)

Indonesian Journal of Agriculture and Environmental Analytics
Vol. 5 No. 2 (2026): July 2026

Land cover and Land use maps are base for monitoring, understanding and evaluating physical and socio-economic characteristics in a region. This study uses sentinel 2b level 2a as an material for analysis. Classification using Pixel and Object Based Random Forest and SVM Algorithm on Sentinel-2 Immagery over Enrekang Region. Confusion matrix were then performed using visually interpreted Pan-sharpened and orthorectified SPOT-6/7 imagery to calculate the accuracy. Overall accuracy shows that classification using objects based Random Forest or SVM Algorithm is greater than classification using pixels based Random Forest or SVM Algorithm. Overall accuracy of the random forest algorithm increases by 1-3% from the SVM algorithm on pixel or object classification.