Skip to Main content Skip to Navigation

Recalage non rigide d'images médicales volumiques : contributions aux approches iconiques et géométriques

Pascal Cachier 1
1 EPIDAURE - Medical imaging and robotics
CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : Non-rigid image registration is a classical problem in computer vision that consists in deforming one image so that it follows the geometry of another image. Registration techniques are very numerous, and are generally classified according to the kind of features they use in the images to deform them. On one hand, intensity-based algorithms use the intensity of the images. On the other hand, geometric algorithms use geometric features segmented from the images, such as object boundaries. In this thesis, we first show that this classification is not fine enough to explain some fundamental differences between some registration algorithms. We propose to part the intensity-based algorithms in two classes : we distinguish between the standard intensity based (SIB) algorithms, and the intensity feature based (IFB) algorithms. We introduce a general registration energy for iconic feature registration; then, we develop particular instances of this energy with special properties according to the application : additional geometric constraints obtained from a segmentatio- n, non uniform bias invariance, vectorial regularization with cross-effects between coordinates, invariance by exchange of the images to be registered. We give applications of our algorithm in brain tracking in volumetric ultrasound image sequences, in inter-subject brain registration using magnetic resonance imaging, and in shape-and-intensity interpolation.
Document type :
Complete list of metadatas
Contributor : Service Ist Inria Sophia Antipolis-Méditerranée / I3s <>
Submitted on : Monday, October 10, 2011 - 2:48:01 PM
Last modification on : Saturday, January 27, 2018 - 1:30:47 AM


  • HAL Id : tel-00630493, version 1



Pascal Cachier. Recalage non rigide d'images médicales volumiques : contributions aux approches iconiques et géométriques. Interface homme-machine [cs.HC]. Ecole Centrale Paris, 2002. Français. ⟨tel-00630493⟩



Record views


Files downloads