Abstract :
In this paper, we propose an efficient image registration method based on global optimization strategy and use it in image stitching successfully. The image registration problem is modeled as an optimization problem, which in our case, is solved by a fitness function that is minimized using genetic algorithm (GA). In the proposed method, we estimate the registration initial position, given by the best chromosome of the GA, and combine the overlapped image regions for smoothing the boundary to realize stitching process by employing image fusion. Experimental results show our approach builds effective image registration and stitching in high operation speed.
