TY - GEN
T1 - Sparse representations for automatic target classification in SAR images
AU - Thiagarajan, Jayaraman J.
AU - Ramamurthy, Karthikeyan N.
AU - Knee, Peter
AU - Spanias, Andreas
AU - Berisha, Visar
PY - 2010
Y1 - 2010
N2 - We propose a sparse representation approach for classifying different targets in Synthetic Aperture Radar (SAR) images. Unlike the other feature based approaches, the proposed method does not require explicit pose estimation or any preprocessing. The dictionary used in this setup is the collection of the normalized training vectors itself. Computing a sparse representation for the test data using this dictionary corresponds to finding a locally linear approximation with respect to the underlying class manifold. SAR images obtained from the Moving and Stationary Target Acquisition and Recognition (MSTAR) public database were used in the classification setup. Results show that the performance of the algorithm is superior to using a support vector machines based approach with similar assumptions. Significant complexity reduction is obtained by reducing the dimensions of the data using random projections for only a small loss in performance.
AB - We propose a sparse representation approach for classifying different targets in Synthetic Aperture Radar (SAR) images. Unlike the other feature based approaches, the proposed method does not require explicit pose estimation or any preprocessing. The dictionary used in this setup is the collection of the normalized training vectors itself. Computing a sparse representation for the test data using this dictionary corresponds to finding a locally linear approximation with respect to the underlying class manifold. SAR images obtained from the Moving and Stationary Target Acquisition and Recognition (MSTAR) public database were used in the classification setup. Results show that the performance of the algorithm is superior to using a support vector machines based approach with similar assumptions. Significant complexity reduction is obtained by reducing the dimensions of the data using random projections for only a small loss in performance.
UR - http://www.scopus.com/inward/record.url?scp=77953847761&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77953847761&partnerID=8YFLogxK
U2 - 10.1109/ISCCSP.2010.5463416
DO - 10.1109/ISCCSP.2010.5463416
M3 - Conference contribution
AN - SCOPUS:77953847761
SN - 9781424462858
T3 - Final Program and Abstract Book - 4th International Symposium on Communications, Control, and Signal Processing, ISCCSP 2010
BT - Final Program and Abstract Book - 4th International Symposium on Communications, Control, and Signal Processing, ISCCSP 2010
T2 - 4th International Symposium on Communications, Control, and Signal Processing, ISCCSP-2010
Y2 - 3 March 2010 through 5 March 2010
ER -