Parallel image registration with OpenMP and MPI

As the image size increases, most numerical approaches for image registration based on the pixel matrices computation will cost a large executing time, then one of the most efficient solving methods is parallel computing. In this paper, we discuss two parallel implementation methods to perform the PDE-based image registration computing. One is the multithreading process with OpenMP on a dual core computer, another is the data-block and multithreads processing with MPI and OpenMP on a 4 nodes cluster disposed by WCCS. Some experimental results show that these two methods can produce the large size image registration and reduce the computation complexity. ©2009 IEEE.