Preprocessing of hyperspectral imagery with consideration of smile and keystone properties

Satellite hyperspectral imaging sensors suffer from ‘’smile’’ and ‘’keystone’’ properties, which appear as distortions of spectrum images. The smile property is a center wavelength shift and the keystone property is a band-to-band misregistration. These distortions degrade the spectrum information and reduce classification accuracies. Furthermore, these properties may change after the launch. Therefore, in the preprocessing of satellite hyperspectral images, the onboard correction of the smile and keystone properties is an important issue as well as the radiometric and geometric correction. The main objective of this work is to propose the prototype of the preprocessing of hyperspectral image with consideration of smile and keystone properties. Image registration based on phase correlation is used for detecting the optical properties. Cubic spline interpolation is adopted to modify the spectrum because of its good trade-off between the smoothness and shape preservation. Smile and keystone detection simulation using the EO-1 Hyperion imagery taken at various times in the past nine years proved that the optical properties have been changing due to the onboard secular distortion. Therefore, onboard optical properties should be updated periodically and built into the radiometric and geometric corrections for future satellite hyperspectral sensors. The proposed method may be the prototype of the preprocessing of future satellite hyperspectral sensors.