Skip to Main content Skip to Navigation

Contribution à l'étude du recalage
de données 3D /couleur

Abstract : Creating a 3D object model requires generally the acquisition of multiple partially overlapping range
images captured with different sensor locations. The 3D registration of these images that are assumed to
be pair-wisely overlapping consists in estimating the rigid transformation that places them in a common
coordinate system. This thesis presents solutions for automatically registering two unstructured 3D-color
point sets provided by a high-resolution scanner. Our first aim was to improve performance of classical
ICP (Iterative Closest Point) approaches by taking into account colour information for point matching.
When colour noise is not too important, using photometric information allows us to improve the
convergence and to decrease the residual error of the registration process. Moreover, it is possible to
reduce the drawbacks of intensity variations by choosing a perceptual colour space where intensity and
chromatic components are separated. The second aim of this work was to develop a fully automatic
registration method, which does not require any pose measuring hardware or manual intervention. In
fact, ICP requires an a-priori knowledge of an approximate estimate of the rigid transformation. This has
been solved using a set of interest point pairs extracted from the two colour images with the Harris
accurate detector. The initial 3D transform is then estimated by applying the RANSAC algorithm to the
set of 3D point pairs issued from the previous colour matching. This randomised estimator is robust to
outliers and provides a rough estimation that is enough accurate to initialise the iterative registration
process. Experimental results show the capabilities of the proposed solutions with colour range images
of real-world objects.
Document type :
Complete list of metadata

Cited literature [75 references]  Display  Hide  Download
Contributor : Lounis Douadi <>
Submitted on : Monday, April 16, 2007 - 6:23:58 PM
Last modification on : Friday, May 17, 2019 - 11:38:38 AM
Long-term archiving on: : Tuesday, April 6, 2010 - 11:11:06 PM


  • HAL Id : tel-00141930, version 1



Lounis Douadi. Contribution à l'étude du recalage
de données 3D /couleur. Automatique / Robotique. Université Montpellier II - Sciences et Techniques du Languedoc, 2006. Français. ⟨tel-00141930⟩



Record views


Files downloads