Land cover fractions mapping using modis imagery over Nanggroe Aceh Darussalam (NAD) province, Indonesia

Land cover analysis plays an important role in many environmental applications nowadays. The objective of this study is to assess the capability of Moderate Resolution Imaging Spectroradiometer (MODIS) scene for land cover mapping over Nanggroe Aceh Darussalam (NAD) province, Indonesia. Standard supervised classification techniques were used in this study namely maximum likelihood, minimum distance-to-mean, parallelepiped and parallelepiped (with Maximum Likelihood Tie Resolution) by using PCI Geomatica 9.1 image processing software. The accuracy of each classification map was assessed using the reference data set consisted of a large number of samples collected per category. The result from this multispectral classification analysis of this study area indicated that urban features could be clearly identified and classified relative to the surrounding terrain and its associated desert features. In this study, Kappa statistic and overall accuracy were calculated and compared for the three supervised classification techniques. Geometric correction was performed to the digital image using the nearest neighborhood method with second order polynomial. This preliminary study has produced a promising result with a geocoded image. This indicates that land features can be carried out using remote sensing classification method of the MODIS imagery.