Spectral Information Analysis from Multisensor Image Fusion for Land Use/Land Cover Classification in a Tropical Area: A Case Study in Bogor, Indonesia (original) (raw)

the arab world geographer, 2011

Abstract

This paper evaluates the analysis of spectral information using RADARSAT, SPOT Panchromatic, and Landsat-TM images for land use/land cover classification in a tropical area. The case study is of Bogor, located in West Java, Indonesia. In order to increase spectral contrast of the Landsat-TM image, two procedures for information compression, colour composite and PCA (Principal Component Analysis) were tested. The best five RGB colour composites were produced, which were visually and statistically investigated, including the RGB colour composites of PC-1, PC-2, and PC-3, which were created from standard and selective principal component analysis (PCA), respectively. To merge images, two spectral-based image fusion techniques were applied: IHS (Intensity, Hue, and Saturation) and Brovey transforms. A land use/land cover classification using IHS transform from the RGB colour composites of PC-1, PC-2, and PC-3, where the intensity was replaced by SPOT Panchromatic, was the best result. It has an overall accura...

Hussein Harahsheh hasn't uploaded this paper.

Let Hussein know you want this paper to be uploaded.

Ask for this paper to be uploaded.