Fast Block Matching Registration with Entropy-based Similarity

Eduardo Suarez-Santana, Rafael Nebot, C.-F. Westin, Juan Ruiz-Alzola

November, 2006

Download full paper

Abstract

This paper describes the implementation of a multidimensional block-matching nonrigid registration algorithm. The main features of the algorithm are its simplicity, its free form nature, the modularity of the similarity measure, which makes it possible using local entropy-based similarity measures and the avoidance of the optimization module. The algorithm implementation described in this paper is based on the method by Suarez et al. [5, 3]. This paper is accompanied with the source code, input data, parameters and output data used for validating the algorithm described in this paper.

Kernels for convolution with point-wise similarity images.

Reference

Suarez-Santana E, Nebot R, Westin CF, Ruiz-Alzola J. Fast Block Matching Registration with Entropy-based Similarity, 2006.

Bibtex entry

@manual{suarez-santanainsightjournal06,
  author         = "Eduardo Suarez-Santana and Rafael Nebot and Carl-Fredrik   
                   Westin and Juan Ruiz-Alzola",                               
  title          = "Fast Block Matching Registration with Entropy-based        
                   Similarity",                                                
  year           = "2006",                                                     
  journal        = {},                                                         
  pages          = {1-8},                                                      
  month          = {November}, }

Grants

NIH U54-EB005149 (NAMIC), NIH P41-RR13218 (NAC)