作者: G. Jones , B. Bhanu
DOI: 10.1109/7.913694
关键词: Low probability of intercept radar 、 Computer vision 、 Radar lock-on 、 Radar engineering details 、 Scattering 、 Side looking airborne radar 、 Radar imaging 、 Continuous-wave radar 、 Bistatic radar 、 Computer science 、 Synthetic aperture radar 、 Artificial intelligence 、 Inverse synthetic aperture radar 、 Automatic target recognition
摘要: Recognizing occluded vehicle targets in synthetic aperture radar (SAR) images is addressed. Recognition algorithms, based on local features, are presented that successfully recognize highly objects both XPATCH SAR signatures and real of actual vehicles from the MSTAR data. Extensive experimental results for a basic recognition algorithm, using scattering center relative locations as features with data an improved scatterer magnitudes The show effect occlusion performance terms probability correct identification, receiver operating characteristic curves, confusion matrices.