PDE-Based Parallel Deformable Registration on a Dual Core Cluster

Abstract

Most numerical approaches for image registration with PDE method are based on the computing over the pixel matrices. As the image size increases, more time consuming is needed. Then, one of the most efficient solving methods is parallel computing. In this paper, we discuss the parallel deformable registration computing carried on the cluster. First, we present the diffusion PDE model to describe the registration problem, detail the sequential adaptive parameter computing method for searching the weight parameter correspond to the local degrees of variability in the match, then perform the parallel implementation on a 4 nodes cluster disposed by WCCS. Some experimental results show that this method can produce the large size parallel image registration and reduce the computation complexity.