%0 Conference Proceedings
%T VISUAL SALIENT SIFT KEYPOINTS DESCRIPTORS FOR AUTOMATIC TARGETRECOGNITION
%+ Lab-STICC_ENSTAB_MOM_PIM
%+ Pôle STIC_REMS
%+ Laboratoire de Recherche Informatique et Télécommunications  (LRIT)
%+ Lab-STICC_ENSTAB_CID_TOMS
%+ Laboratoire de Recherche en Informatique et Télécommunications [Rabat] (GSCM-LRIT)
%A Karine, Ayoub
%A Toumi, Abdelmalek
%A Khenchaf, Ali
%A Hassouni, Mohammed El
%< avec comité de lecture
%B EUVIP
%C Marseille, France
%8 2016-10-25
%D 2016
%K classification
%K Automatic target recognition
%K inverse synthetic aperture radar
%K SIFT
%K visual attention model
%Z Engineering Sciences [physics]/Electromagnetism
%Z Engineering Sciences [physics]/Signal and Image processingConference papers
%X This paper addresses the problem of automatic target recognition(ATR) using inverse synthetic aperture radar (ISAR) images.In this context, we propose a novel approach for featureextraction to describe precisely an aircraft target from ISARimages. In our approach, a visual attention model is adoptedto separate the salient regions from the background. Afterthat, the scale invariant feature transform (SIFT) method isused to extract the keypoints and their descriptors. Then, alocal salient feature is built by considering only the keypointslocated in the salient region. For the classification step, thesupport vector machines (SVM) classifier is adopted. To validatethe proposed approach, ISAR images database whichwas collected from anechoic chamber is used.
%G English
%L hal-01406143
%U https://hal.archives-ouvertes.fr/hal-01406143
%~ CNRS
%~ UNIV-BREST
%~ UNIV-UBS
%~ INSTITUT-TELECOM
%~ LAB-STICC
%~ ENSTA-BRETAGNE
%~ ENSTA-BRETAGNE-STIC
%~ ENIB
%~ LAB-STICC_ENIB
%~ INSTITUTS-TELECOM