PRECISE IMAGE REGISTRATION BASED ON SIFT WITH RANSAC ALGORITHM

Abstract

–This paper discussesThe automatic image registration of oblique images taken at different viewpoints
remains a challenge until today. Image registration is the process of geometrically aligning two images taken at different
times, at different orientations, or by different sensors. It is vital step in image analysis. It has wide range of applications
in the field of satellite imaging, and several medical as well as computer vision fields. Image registration is the
fundamental task used to match two or more images of same scene taken at different environments. The steps in image
registration consist of feature point detection, feature point descriptor extraction and feature point matching. The main
aim of the matching is to find the correspondence between the two images. For the process of image matching extraction
of stable features (key points) is the major issue. Generally corner detectors used as stable features, apart from the
corners some region based detectors are there. One of the region based detector is Scale Invariant Feature Transform
(SIFT) algorithm. SIFT is one of the feature point detectors that can able to provide a set of features of an image that are
not affected by many of the complications experienced in other methods. Random Sample Consensus (RANSAC)
algorithm is applied to eliminate variety of mismatches and acquire transformation matrix between the images. The input
image transformed with the right mapping model for image mosaic. Image mosaic is combining or stitching several
images of a same scene or object taken from different angles into a single image with large field of view. In this work, an
algorithm for mosaicing of two images efficiently using SIFT feature detection method. The false matching pair is
removed by using RANSAC algorithm. The transformation model estimated from the features and the image warped
correspondingly