Centre de documentation
scientifique

Detection of buried targets via active selection of labeled data: Application to sensing subsurface UXO / Y. Zhang in IEEE Transactions on geoscience and remote sensing, vol 42 n° 11 (November 2004)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 42 n° 11 (November 2004) . - pp 2535 - 2543 Titre : Detection of buried targets via active selection of labeled data: Application to sensing subsurface UXO Type de document : Article/Communication Auteurs : Y. Zhang, Auteur ; X. Liao, Auteur ; L. Carin, Auteur Année de publication : 2004 Article en page(s) : pp 2535 - 2543 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] cible cachée [Termes IGN] classification [Termes IGN] détection de cible [Termes IGN] erreur moyenne quadratique [Termes IGN] matrice [Termes IGN] matrice d'information de Fischer [Termes IGN] mine antipersonnel [Termes IGN] signature spectrale Résumé : (Auteur) When sensing subsurface targets, such as land-mine and unexploded ordnance (UXO), the target signatures are typically a strong function of environmental and historical circumstances. Consequently, it is difficult to constitute a universal training set for design of detection or classification algorithms. In this paper, we develop an efficient procedure by which information-theoretic concepts are used to design the basis functions and training set, directly from the site-specific measured data specifically, assume that measured data (e.g., induction and/or magnetometer) are available from a given site, unlabeled in the sense that it is not known a priori whether a given signature is associated with a target or clutter. For N signatures, the data may be expressed as {Xi, Yi}i=1,N, where xi is the measured data or buried object i, and yi is the associated unknown binary label (target/nontarget). Let the N xi define the set X. The algorithm works in four steps : 1) the Fisher information matrix is used to select a set of basis functions for the kernel-based algorithm, this step defining a set of n signatures Bn X that are most informative in characterizing the signature distribution of the site; 2) the Fisher information matrix is used again to define a small subset Xs X, composed of those xi for which knowledge of the associated labels yi would be most informative in defining the weights, for the basis functions in Bn ; 3) the buried objects associated with the signatures in Xs., are excavated, yielding the associated labels yi, represented by the set Ys.; and 4) using Bn, Xs, and Ys, a kernel-based classifier is designed for use in classifying all remaining buried objects. This framework is discussed in detail, with example results presented for an actual buried-UXO site. Numéro de notice : A2004-462 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2004.836270 En ligne : https://doi.org/10.1109/TGRS.2004.836270 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=26982 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
065-04111	RAB	Revue	Centre de documentation	En réserve L003	Disponible