Skip to Main content Skip to Navigation
Theses

Détection de filaments dans des images 2D et 3D : modélisation, étude mathématique et algorithmes

Alexis Baudour 1
1 ARIANA - Inverse problems in earth monitoring
CRISAM - Inria Sophia Antipolis - Méditerranée , Laboratoire I3S - SIS - Signal, Images et Systèmes
Abstract : This thesis is devoted to the filament modeling and to the detection problem in 3D images. We developed some variational methods for four specific applications: we use the new concept of total curvature for the road detection in order to preserve a smooth network with corners ; detection and completion of filaments with strong noise component and occultations. We use magnetostatic and Ginzburg-Landau's theory to represent filaments as the singularities set of a vector field ; filaments detection in biological images produced by confocal microscopy. We modelize filaments taking into account characteristics of the confocal microscopy. The filaments are obtained with a maximum a posteriori estimation ; targets detection in sequences of infrared images. In this application we try to find trajectories optimizing the average luminosity between the trajectory and its neighbourhood, taking into account the differences between sensors ; furthermore, we prove theoretical results relative to the total curvature and the convergence of Alouges' method associated to Ginzburg-Landau systems. This work gathers both modeling, theoretical results and development of effective numerical algorithms adapted to actual applications.
Document type :
Theses
Complete list of metadatas

Cited literature [1 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00507520
Contributor : Estelle Nivault <>
Submitted on : Friday, July 30, 2010 - 4:25:28 PM
Last modification on : Monday, October 12, 2020 - 10:30:12 AM
Long-term archiving on: : Tuesday, October 23, 2012 - 11:56:10 AM

Identifiers

  • HAL Id : tel-00507520, version 1

Collections

Citation

Alexis Baudour. Détection de filaments dans des images 2D et 3D : modélisation, étude mathématique et algorithmes. Interface homme-machine [cs.HC]. Université Nice Sophia Antipolis, 2009. Français. ⟨tel-00507520⟩

Share

Metrics

Record views

581

Files downloads

346