Meta-classifiers for exploiting feature dependencies in automatic target recognition

Of active interest in automatic target recognition (ATR) is the problem of combining the complementary merits of multiple classifiers. This is inspired by decades of research in the area which has seen a variety of fairly successful feature extraction techniques as well as decision engines being developed. While heuristically based fusion techniques are omnipresent, this paper explores a principled meta-classification strategy that is based on the exploitation of correlation between multiple feature extractors as well as decision engines. We present two learning algorithms respectively based on support vector machines and AdaBoost, which combine soft-outputs of state of the art individual classifiers to yield an overall improvement in recognition rates. Experimental results obtained from benchmark SAR image databases reveal that the proposed meta-classification strategies are not only asymptotically superior but also have better robustness to choice of training over state-of-the art individual classifiers.

[1]  Atindra Mitra,et al.  Improved HRR-ATR using hybridization of HMM and eigen-template-matched filtering , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[2]  Yoav Freund,et al.  A Short Introduction to Boosting , 1999 .

[3]  Trevor Darrell,et al.  The pyramid match kernel: discriminative classification with sets of image features , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[4]  David G. Lowe,et al.  Object recognition from local scale-invariant features , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[5]  G. A. Baraghimian,et al.  Artificial neural networks for automatic target recognition , 1992 .

[6]  Benyong Liu,et al.  Radar Target Recognition Using A Modified Kernel Direct Discriminant Analysis Algorithm , 2007, 2007 International Conference on Communications, Circuits and Systems.

[7]  Nasser M. Nasrabadi,et al.  Fusion techniques for automatic target recognition , 2003, 32nd Applied Imagery Pattern Recognition Workshop, 2003. Proceedings..

[8]  Michael I. Miller,et al.  Hilbert-Schmidt Lower Bounds for Estimators on Matrix Lie Groups for ATR , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Qun Zhao,et al.  Support vector machines for SAR automatic target recognition , 2001 .

[10]  Scott T. Acton,et al.  Speckle reducing anisotropic diffusion , 2002, IEEE Trans. Image Process..

[11]  D. Fernandes,et al.  Automatic Target Recognition in Synthetic Aperture Radar image using multiresolution analysis and classifiers combination , 2008, 2008 IEEE Radar Conference.

[12]  Clark F. Olson,et al.  Automatic target recognition by matching oriented edge pixels , 1997, IEEE Trans. Image Process..

[13]  Heesung Kwon,et al.  Kernel matched subspace detectors for hyperspectral target detection , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  David Casasent,et al.  A hierarchical classifier using new support vector machines for automatic target recognition , 2005, Neural Networks.

[15]  Jim Schroeder,et al.  Automated Target Recognition Using the Karhunen-Loéve Transform with Invariance , 2002, Digit. Signal Process..

[16]  Arnab K. Shaw,et al.  Improved automatic target recognition using singular value decomposition , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[17]  N M Sandirasegaram.  Spot SAR ATR Using Wavelet Features and Neural Network Classifier , 2005 .

[18]  Alexander Hauptmann,et al.  Meta-Classification of Multimedia Classifiers , 2002, KDMCD.

[19]  Jean-Claude Souyris,et al.  Target recognition in SAR images with Support Vector Machines (SVM) , 2007, 2007 IEEE International Geoscience and Remote Sensing Symposium.