Registration is a fundamental operation in image processing to align images taken at different times, from different sensors or from different viewing angles. Automatic image registration procedures are gaining importance to efficiently register the large volumes of remote sensing data available these days. In this Letter, we investigate an automated mutual information-based registration technique for remote sensing data. The performance of a number of interpolation algorithms to compute mutual information for registration of multi-sensor and multi-resolution Landsat TM, Radarsat SAR and IRS PAN images is evaluated.
ASJC Scopus subject areas
- Earth and Planetary Sciences(all)