Image registration

Abstract

Image Registration is considered one of the main image processing tasks and has been researched extensively. Its use spans a wide range of applications such as remote sensing, medical imaging, security and surveillance and photography to name a few. All the developed approaches, however, considered unconstrained platforms regardless of processing and memory capabilities. This chapter reviews the basics of image registration as well as state of the art algorithms found in literature. Two developed approaches are then discussed: OESR and AMIR. While OESR applies an optimized Exhaustive search to register two images in a multi-resolution pyramidal scheme, AMIR offers an automatic multimodal image registration based on gradient descent optimization. Both algorithms exhibit comparable performance to state of the art approaches while decreasing the processing burden. © 2014 Springer Science+Business Media, LLC.