Edge directed automatic control point selection algorithm for image morphing

Abstract

Image morphing is a technique used for the metamorphosis of one image to another. In order to do morphing, new positions (control points) for the pixels in each of the images must be determined. The control point selection is a difficult process that is conducted manually in most cases, which is time consuming and lacks accuracy. In this research, we propose an Edge Directed Automatic Control Point Selection algorithm that uses Canny Operator to detect edges and a Central Axis-based Partition method to establish the correspondence between the source and the target images. Experimental results show that the proposed low complexity algorithm performance is better than that of a manually chosen point selection method in terms of both accuracy and time.