Automatic Control Point Generation for Satellite Image Registration

Main Article Content

Dhanya Raveendran
S. Priyadarsini


Image registration is the process of spatially aligning two or more images of the same scene taken at different times, from different viewpoints, and/or by different sensors. In this process, two images (the base image and sensed image) are geometrically aligned in order to compare the difference between them. The key operation regarding automatic registration of satellite images is to generate an accurate set of control points and then apply a suitable transformation function to the pair of images to be registered. This paper proposes a fundamentally new approach to automatically generate the control points and based on these points, registration of satellite images is performed. After removing the outliers from the initially generated random set control points, a mapping function is generated as a hypothesis using the proposed method. Parameters of the mapping function are used in the transformation step to align the base image and sensed image. The sampling technique used in this paper is more efficient than the existing methods like RANSAC and LMedS.

Keywords: Image registration; transformation; outliers; mapping function; RANSAC; LMedS


Download data is not yet available.

Article Details