PENGARUH JUMLAH SALURAN SPEKTRAL, KORELASI ANTAR SALURAN SPEKTRAL DAN JUMLAH KELAS OBJEK TERHADAP AKURASI KLASIFIKASI PENUTUP LAHAN (Effect of Number of Spectral Bands, Correlation among Spectral Bands and Number of Object Class to Land Cover Classification Accuracy)

Land cover is one of the most important information that can be obtained from remote sensing data. It were needed as a basis data for government to determine the direction of development policy, regional development planning, and management of natural resources. Therefore, inventory and mapping of land need to be implemented in a sustainable, rapid, precise, and also accurate. The purposes of this study is to determine changes of the accuracy level from land cover classification of remote sensing image as the increased number of spectral bands that are involved, the higher the correlation between spectral bands involved, and as the addition of the number of class objects. The results of this study showed that the increasing number of spectral bands that are involved can improve accuracy, the higher correlation between spectral bands make the accuracy obtained decreased, classification using more number of object classes the accuracy obtained decreased