Increasing the discrimination of synthetic aperture radar recognition models
Bir Bhanu · Optical Engineering · 2002
The focus of this work is optimizing recognition models for synthetic aperture radar (SAR) signatures of vehicles to improve the performance of a recognition algorithm under the extended operating conditions of target articulation, occlusion, and configuration variants. The recognition models are based on quasi-invariant local features, scat- tering center locations, and magnitudes. The approach determines the similarities and differences among the various vehicle models. Methods to penalize similar features or reward dissimilar features are used to increase the distinguishability of the recognition model instances. Exten- sive experimental recognition results are presented in terms of confusion matrices and receiver operating characteristic (ROC) curves to show the improvements in recognition performance for real SAR signatures of ve- hicle targets with articulation, configuration variants, and occlusion. © 2002 Society of Photo-Optical Instrumentation Engineers. (DOI: 10.1117/1.1517286) Subject terms: articulated object recognition; automatic target recognition; object similarity; recognizing configuration variants; recognizing occluded objects.