Image registration by iterative closest point (ICP) algorithm

Eda Egüz

Semester Project

Electrical and Electronics Engineering, EPFL

June 2009

Abstract

Nowadays, having a precise and clear image through a microscope carries great importance in the field of medical image processing. However, there are certain distorting factors as the incoherent distribution of local images, which are the part of the whole image.
One of the solutions to this problem is the use of the Iterative Closest Point Algorithm (ICP). This algorithm is a vision the technique which permits registration between two images. Thus, it can be efficiently employed in correcting the distribution of these local images.
This project especially focuses on registration of dendrite images.