A dynamic mosaicking method for finding an optimal seamline with Canny edge detector

Abstract

Image mosaicking is a process of assembling multiple images to create an image with a larger field of view. Therefore, to avoid failures and to solve errors that could occur during this process, a new method for dynamic mosaicking is presented. This method finds regions that are similar between the images and regions that are not alike, then an optimal seamline is computed by going through the similar regions as much as possible and by avoiding the not common regions. In fact, this is done by combining Canny edge detector and the outliers and inliers from the RANSAC method to identify these areas. In addition, by highlighting these regions in an intensity difference, a map that reveals which areas to avoid and to frequent was created. Thus, an optimal seamline was found. The experimental results show that the proposed approach is capable of generating robust mosaics against ghosting and parallax effects.