ANALYSIS OF OBJECT-ORIENTED CLASSIFICATION RESULTS DERIVED FROM PAN-SHARPENED LANDSAT 7 ETM+ AND ASTER IMAGES (original) (raw)

Abstract

In this study, performance of object-oriented classification approach has been tested using medium resolution satellite dataset of Zonguldak testfield. For this purpose, Landsat 7 ETM+ and ASTER images were used because of their nearly similar ground sampling distance (GSD). As a first step, pan-sharpened images were created based on the combination of panchromatic and color bands available in the dataset using a special methodology implemented in the PCI Geomatica v9.1.4 software package. Following this, resulted images were handled by the eCognition v4.0.6 software with the main steps of segmentation and classification. After determining the optimal segmentation parameters correctly, classification of main object classes were realized and verified by the auxiliary data e.g. maps, aerial photos and personal information.

Murat OruÇ hasn't uploaded this paper.

Let Murat know you want this paper to be uploaded.

Ask for this paper to be uploaded.