Registration and mosaicing for images obtained from UAV

Image registration is a fundamental topic in image processing and has a wide variety of applications in computer vision. It is the process of matching two or more images taken at different times from different sensors or from different viewpoints, so that the matched coordinate points in the two images correspond to the same physical region of the scene being imaged. Typically Image registration is required in remote sensing (multi spectral classification, image fusion, environmental monitoring, change detection, Image Mosaicing, Weather forecasting) in medicine, in cartography and in computer vision (target localization, automatic quality control). In this paper, a feature based image registration technique is introduced for mosaicing the sequence of UAV images. First, features are extracted using corner detectors and transformation parameters are computed using a discrete randomized approach. Then blending of images is carried out in a tree-based approach. Finally, results of merging for some sequence of UAV images are shown.