Wide-angle SAR presents an opportunity to explore new methods for CV classification. Polarization features are of particular interest and various examples of wide-aperture images using flight-collected and simulated data are shown to emphasize how odd and even reflectors are revealed in the imagery. Using the well-known frequency parameter from high-frequency scattering models and the Krogager polarization decomposition, wide-angle polarimetric SAR imagery is attributed with a basis of ASC primitive targets. We observe trends when extracting ASCs from 2D images and study the behavior for 10 civilian vehicles using the AFRL Civilian Vehicle Data Dome. Passenger cars categorized as the sedan class tended to exhibit scattering by cylinders, horizontal edges, and horizontal dihedrals at low elevation angle (30°) but primarily showed horizontal cylinders and horizontal dihedrals for the case of high elevation angle (60°). Likewise, the dominant scatterers for the SUV class were horizontal cylinders and horizontal edges for 30° elevation and primarily horizontal cylinders and dihedrals for 60° elevation. These results suggest that there are noticeable differences between the classes when viewed as primitives in the joint frequency/polarization/wide-angle SAR feature space. Thus, rethinking vehicle classification to include efficient primitive type extraction during image formation may lead to new, near real-time vehicle classification algorithms.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Rethinking vehicle classification with wide-angle polarimetric SAR


    Contributors:


    Publication date :

    2014-01-01


    Size :

    3212250 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English




    Passive polarimetric IR target classification

    Sadjadi, F.A. / Chun, C.S.L. | IEEE | 2001


    Vehicle mirror assembly with wide angle element

    FOOTE KEITH D / BENTHEM KEVIN P / LYNAM NIALL R | European Patent Office | 2017

    Free access


    Snow Cover Classification Using Polarimetric Data

    Venkataraman, G. / Singh, G. / European Space Agency | British Library Conference Proceedings | 2009