Analysis of multiple-view Bayesian classification for SAR ATR

Classification of targets in high-resolution synthetic aperture radar imagery is a challenging problem in practice, due to extended operating conditions such as obscuration, articulation, varied configurations and a host of camouflage, concealment and deception tactics. Due to radar cross-section variability, the ability to discriminate between targets also varies greatly with target aspect. Potential space-borne and air-borne sensor systems may eventually be exploited to provide products to the warfighter at tactically relevant timelines. With such potential systems in place, multiple views of a given target area may be available to support targeting. In this paper, we examine the aspect dependence of SAR target classification and develop a Bayesian classification approach that exploits multiple incoherent views of a target. We further examine several practical issues in the design of such a classifier and consider sensitivities and their implications for sensor planning. Experimental results indicating the benefits of aspect diversity for improving performance under extended operating conditions are shown using publicly released 1-foot SAR data from DARPA's MSTAR program.

[1]  Joseph A. O'Sullivan,et al.  Recognition performance from synthetic aperture radar imagery subject to system resource constraints , 2001 .

[2]  R. Hummel,et al.  Model-based ATR using synthetic aperture radar , 2000, Record of the IEEE 2000 International Radar Conference [Cat. No. 00CH37037].

[3]  Tomoharu Nakahara,et al.  A Multiple-baseline Stereo Method , 2002 .

[4]  Larry L. Horowitz,et al.  Fundamental SAR ATR performance predictions for design trade-offs , 1997, Defense, Security, and Sensing.

[5]  Joseph A. O'Sullivan,et al.  SAR ATR performance using a conditionally Gaussian model , 2001 .

[6]  Rama Chellappa,et al.  Target aspect estimation from single and multi-pass SAR images , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[7]  Larry L. Horowitz,et al.  Fundamental SAR ATR performance predictions for design trade-offs: 1D HRR versus 2D SAR versus 3D SAR , 1999, Defense, Security, and Sensing.

[8]  Leslie M. Novak,et al.  Performance of 10- and 20-target MSE classifiers , 2000, IEEE Trans. Aerosp. Electron. Syst..

[9]  Anil K. Jain,et al.  Statistical Pattern Recognition: A Review , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Larry L. Horowitz,et al.  Benefits of aspect diversity for SAR ATR: fundamental and experimental results , 2000, SPIE Defense + Commercial Sensing.

[11]  Timothy D. Ross,et al.  SAR ATR: so what's the problem? An MSTAR perspective , 1999, Defense, Security, and Sensing.

[12]  Timothy D. Ross,et al.  Evaluation of SAR ATR algorithm performance sensitivity to MSTAR extended operating conditions , 1998, Defense, Security, and Sensing.

[13]  Rama Chellappa,et al.  Context-aided false alarm reduction for SAR automatic target recognition , 1997, Proceedings of International Conference on Image Processing.