Band-to-band registration of remotely sensed imagery using mutual information and its application to Suomi NPP VIIRS data

Retrieval of band-to-band registration between spectrally disparate bands is difficult due to the fact that intensity imagery from different spectral regions often appear quite different. This is due to the fact that most materials interact with electromagnetic waves differently at different wavelengths. In this paper, we will present a new approach to the problem base on statistical image analysis and how to overcome the undersampling problem in VIIRS data. We will show the registration of intensity imagery from approximately 640 nm to 11450 nm, covering VisNir, S/WMIR and LWIR focal planes on the VIIRS instrument.