Algorithmes d’imagerie SAR polarimétrique basés sur des modèles à sous-espace. : Application à la détection de cible sous couvert forestier.

Abstract : This thesis deals with the development of new SAR processors for detection of targets under foliage. For this application, classical SAR images are very noisy; the target has a low response and there are a lot of false alarms due to the interferences which are mainly the trunks of the forest. To detect target and reduce interferences, we reconsider the way processing a SAR image by using models which take into account the scattering properties of the target and the interferences. These models are developed for signals in sigle polarisation and in double polarisation (HH and VV) and are defined with low-rank subspaces generated from the responses of canonical elements. From this modeling, we develop new SAR processors. First, we focus on the detection of the target and on the reduction of false alarms. The performances of these processors in terms of detection and false alarms reduction are evaluated with realistic simulated data and we emphasize on the interest of using a polimetric information. We also apply these algorithms to real data that allows us to analyze the images in real-life context.
Complete list of metadatas

Cited literature [57 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-01286553
Contributor : Laetitia Thirion-Lefevre <>
Submitted on : Friday, March 11, 2016 - 9:36:40 AM
Last modification on : Tuesday, May 14, 2019 - 9:36:20 AM
Long-term archiving on : Monday, June 13, 2016 - 8:47:18 AM

Identifiers

  • HAL Id : tel-01286553, version 1

Collections

Citation

Frédéric Brigui. Algorithmes d’imagerie SAR polarimétrique basés sur des modèles à sous-espace. : Application à la détection de cible sous couvert forestier.. Electromagnétisme. Université Paris Ouest-Nanterre La Défense, 2010. Français. ⟨tel-01286553⟩

Share

Metrics

Record views

244

Files downloads

223