Hyperspectral satellite data (Hyperion) preprocessing—a case study on banded magnetite quartzite in Godumalai Hill, Salem, Tamil Nadu, India

Abstract

This paper deals with many calibration methods involved in preprocessing of hyperspectral remote sensing data, while radiation calibration with respect to magnetite minerals without altering image reflectance. Radiometric calibration is the basis and important for both hyperspectral data applications and quantitative remote sensing. While at present there are many methods available for radiometric calibration in the spaceborne hyperspectral data, here, we used the highest spectral resolution Hyperion EO-1 satellite data using a variety of calibration models, which include the calibration model based on atmospheric radiation theory, conversion model as scene dependent such as flat field model, internal average reflectance model, logs residual modified model, and linear model based on the experience of the ground calibration, and so on, and hyperspectral data calibration study was implemented with magnetite mineral. After the above three models contrast, we have arrived at the conclusion that calibration model based on atmospheric radiation theory is the finest radiation calibration method for Hyperion satellite data preprocessing.