Image registration is the process of geometrically aligning two images of the same scene that have been acquired under varying conditions. It is usually employed as the first stage of other computer vision schemes and the applications of image registration include medical imaging, remote sensing, robot vision, motion detection and guidance systems, stereo vision and pattern recognition. In the remote sensing field it is frequently being used for change detection, topographic mapping and urban planning to name a few applications. For the case of change detection, the objective is to find the differences between two images of the same scene that have been taken from variable viewpoints, at different tim es, using different cameras and sensors.
In this project we develop automated aerial image registration methods based on alignment of stochastic edge maps computed using nonparametic density estimation.
Fig. 1: Image registration test-bed.
- Makrogiannis S, Bourbakis NG. 2010. Efficient registration of multitemporal and multisensor aerial images based on alignment of nonparametric edge features. J. Electronic Imaging. 19:013002.