Abstract
Three surface-based registration algorithms using different point projection-based matching criteria and using different distance minimization strategies are analyzed: Pelizzari’s head-hat algorithm, a variant of Borgefors’s hierarchical chamfer matching, and Besl and McKay’s iterative closest point algorithm. This work is part of COVIRA (Computer Vision in Radiology), project A2003 of the AIM (Advanced Informatics in Medicine) programme of the European Commission. COVIRA-specific object oriented aspects of the algorithms’ implementations are briefly discussed. Finally they are compared with respect to speed, accuracy and flexibility. This comparison is based on the registration results of the COVIRA reference data sets (MRI/MRA and MR/CT), and a database of MR/PET data sets.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have