Automatic Image Registration of Multiview Satellite Images by using Surf and Ransac

Image registration is a process of overlaying two or more images of the same scene taken from different viewpoints, at different times, and/or by different sensors. Image registration is a key process in various fields like remote sensing, medical imaging, computer vision and cartography. Image registration process consists of four basic steps: feature detection, feature matching, transformation model estimation and image transformation and resampling. Automatic image registration is still a challenging task, particularly for remote sensing images. Auto-transformation of satellite images is a crucial step in Automatic image registration. SURF is the fastest algorithm for feature detection and feature matching. RANSAC algorithm is used for autotransformation of satellite images and finally generates the registered image. This paper shows the results of feature detection and feature matching using SURF and final registered image obtained by RANSAC.