Abstract

Image registration is the process of finding the best spatial alignment between two images. In this talk I will discuss some recent projects which have used improved image registration techniques to achieve better performance. The first application is in improving motion compensation in video compression where we would like to find the best alignment between successive video frames in order to form a good prediction of one frame from the other. I will then give an overview of using image registration in a medical imaging application where we would like to register a 2D X-ray video frame with a 3D CT scan. Lastly I will discuss registering remotely sensed images which have been captured using different sensors to provide a richer set of features with which to classify objects and regions in the images.

Mark Pickering is an Associate Professor in the School of Engineering and Information Technology, The University of New South Wales, at the Australian Defence Force Academy. Mark holds Bachelor, Masters and Doctor of Philosophy degrees in Electrical Engineering and has co-authored over 50 journal and international conference papers on image processing, compression and communications. His main area of research is digital image processing. Within this broad discipline, he has supervised research projects on more specific topics including: digital video compression, hyperspectral data compression, image segmentation, streaming video over computer networks, medical image registration, digital video watermarking and automatic target detection.