Numerical evaluation of poisson image blending problem using AOR iteration via five-point Laplacian operator

Abstract

Poisson equation has been widely applied in mathematics and engineering fields. It is a famous equation which is extensively used in image processing problem. The purpose of this study is to examine the potency of the Accelerated Over Relaxation (AOR) iterative method for Poisson image blending problem. In order to do the comparative analysis, the implementation of Gauss-Seidel (GS), Successive Over Relaxation (SOR) and AOR iterative methods are carried out to get the numerical solution of the equation that is generated by applying the five-point Laplacian operator. Numerical results are analyzed on the basis of two key criteria, the number of iterations and computational time. Lastly, it can be shown that AOR iterative method has superiority in solving Poisson image blending problem as compared with the other two methods because the iterative method eventually required lower number of iterations and lesser computational time.