Registration of High Resolution SAR and Optical Satellite Imagery in Urban Areas

With the launch of high resolution remote sensing satellites in different modalities like TerraSAR-X, WorldView-1 and Ikonos, the contribution of remote sensing for various applications has received a tremendous boost. Specifically, the combined analysis of high resolution SAR and optical imagery is of immense importance in monitoring and assessing catastrophes and natural disaster. Although, latest satellites provide georeferenced and orthorectified data products, still registration errors exist within images acquired from different sources. These need to be taken care off through quick automated techniques before the deployment of these data sources for remote sensing applications. Modern satellites like TerraSAR-X and Ikonos have further widened the existing gap of sensor geometry and radiometry between the two sensors. These satellites provide high resolution images generating enormous data volume along with very different image radiometric and geometric properties (especially in urban areas) leading to failure of multimodal similarity metrics like mutual information to detect the correct registration parameters. In this paper we present a processing chain to register high resolution SAR and optical images by combining feature based techniques namely, homogeneous regions extracted from high resolution images and intensity based similarity metrics namely normalized cross correlation and mutual information. Our test dataset consist of images from TerraSAR-X and Ikonos acquired over the city of Sichuan, China. First results from registration show good visual alignment of SAR and the optical image