Skip to Main content Skip to Navigation
Theses

Segmentation par modèle déformable surfacique localement régularisé par spline lissante

Abstract : Image segmentation through deformable models is a method that localizes object boundaries. When difficult segmentation context are proposed because of noise or a lack of information, the use of prior knowledge in the deformation process increases segmentation accuracy. Medical imaging is often concerned by these context. Moreover, medical applications deal with large amounts of data. Then it is mandatory to use a robust and fast processing.
This question lead us to a local regularisation of the deformable model. Highly based on the active contour framework, also known as snake, we propose a new regularization scheme. This is done by filtering the displacements at each iteration. The filter is based on a smoothing spline kernel whose aim was to approximate a set of points rather than interpolating it.
We point out the consistency of the regularization parameter in such a method. It deals with a real, positive λ value that tunes the cut-off frequency of a low-pass filter. As it is possible to link analytically the cut-off frequency and the sampling frequency, one may give a metric meaning to this cut-off frequency. Moreover a different λ value should be set at each point through a variation of the filter's coefficients. It induces a local cut-off frequency and a local regularization.
This new regularization method should be applied to surfaces. The main difficulty concerns the connectivity of the surface mesh. It has to be 4-valenced everywhere due to the bidimensional fashion of the filter. Segmentation results are given for such meshes. A special process of singular points on spherical mesh is proposed.
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00262088
Contributor : Jérôme Velut <>
Submitted on : Monday, March 10, 2008 - 5:54:10 PM
Last modification on : Wednesday, July 8, 2020 - 12:42:07 PM
Long-term archiving on: : Friday, September 28, 2012 - 11:00:23 AM

Identifiers

  • HAL Id : tel-00262088, version 1

Citation

Jérôme Velut. Segmentation par modèle déformable surfacique localement régularisé par spline lissante. Interface homme-machine [cs.HC]. INSA de Lyon, 2007. Français. ⟨tel-00262088⟩

Share

Metrics

Record views

498

Files downloads

1938