Toward Preparing the Spectral Signature Library for the Ground Features Using Hyper-Spectral Remote Sensing Techniques

Nawal Kh. Ghazal

Abstract: The research represents the starting point in the preparation of the so-called spectral signature library of the ground features based on hyper-spectral satellite imaging techniques. This research represents a new approach (at the level of Iraq) for the study of the spectral signatures of different land cover types. From an experimental point of view, a satellite scene belongs to the Hyperion sensor ( mounted on the EO-1 satellite) has been download via the official web site for the (USGS), this scene covers part of the city of Baghdad-Iraq with 242 spectral bands ranging from the ultraviolet region to the short infrared region of the electromagnetic spectrum. The specialized ERDAS Imagine 2014 software package was used to generate a unified stack image file of all of these 242 bands. This will be more easy in handling and processing of this huge bundle of bands, after then, the bad bands of the scene have been determined by indicating their spectral regions in the spectral signature plot. Finally, five samples of the some mainland covers have been taken with plotting the spectral signature for each one of them at different positions within the region study, these land covers are ( vegetation, two types of bare soils, built-up and water). The general shape of the spectral signature for each sample represents the spectral behavior of the main land cover that clearly indicating and distinguishing all the other related secondary land cover classes

Keywords: Hyperspectral, Imagery Pre-processing, Hyperion, EO-1

Edition: Volume 7 Issue 1, January 2018,

Pages: 1401 - 1404