Target identification using dictionary matching of generalized polarization tensors

Habib Ammari (1) , Thomas Boulier (1) , Josselin Garnier (2, 3) , Wenjia Jing (1) , Hyeonbae Kang (4) , Han Wang (1)

1 DMA - Département de Mathématiques et Applications - ENS Paris
2 LJLL - Laboratoire Jacques-Louis Lions
3 LPMA - Laboratoire de Probabilités et Modèles Aléatoires
4 Inha University

The aim of this paper is to provide a fast and efficient procedure for (real-time) target identification in imaging based on matching on a dictionary of precomputed generalized polarization tensors (GPTs). The approach is based on some important properties of the GPTs and new invariants. A new shape representation is given and numerically tested in the presence of measurement noise. The stability and resolution of the proposed identification algorithm is numerically quantified. We compare the proposed GPT-based shape representation with a moment-based one.

Mots clés: imaging in complex media

Domaines: Probabilités [math.PR]; Equations aux dérivées partielles [math.AP]

Habib Ammari, Thomas Boulier, Josselin Garnier, Wenjia Jing, Hyeonbae Kang, et al.. Target identification using dictionary matching of generalized polarization tensors. Foundations of Computational Mathematics, 2014, 14 (1), pp.27-62. ⟨10.1007/s10208-013-9168-6⟩. ⟨hal-01017774⟩