Superresolution Imagery Based SVM Classification of Radar Targets

Konferenz: EUSAR 2006 - 6th European Conference on Synthetic Aperture Radar
16.05.2006 - 18.05.2006 in Dresden, Germany

Tagungsband: EUSAR 2006

Seiten: 4Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Radoi, Emanuel; Quinquis, André (ENSIETA, France)
Totir, Felix (METRA Research Agency, Romania)
Anton, Lucian (Military Technical Academy, Romania)

Automatic target recognition using SVM (Support Vector Machine) in the context of an anechoic chamber experiment is presented in the paper. The targets are first imaged using MUSIC-2D (Multiple Signal Classification) algorithm and their shapes are then extracted using ADC (Active Deformable Contours). The feature vector includes the Fourier descriptors calculated from the target shapes. The classification is finally performed by a RBF kernel based SVM classifier. It is compared to a standard KNN (K Nearest Neighbors) classifier in terms of classification accuracy and robustness.