Supervised Self-Organizing Classification of Superresolution ISAR Images: An Anechoic Chamber Experiment

The problem of the automatic classification of superresolution ISAR images is addressed in the paper. We describe an anechoic chamber experiment involving ten-scale-reduced aircraft models. The radar images of these targets are reconstructed using MUSIC-2D (multiple signal classification) method coupled with two additional processing steps: phase unwrapping and symmetry enhancement. A feature vector is then proposed including Fourier descriptors and moment invariants, which are calculated from the target shape and the scattering center distribution extracted from each reconstructed image. The classification is finally performed by a new self-organizing neural network called SART (supervised ART), which is compared to two standard classifiers, MLP (multilayer perceptron) and fuzzy KNN ( nearest neighbors). While the classification accuracy is similar, SART is shown to outperform the two other classifiers in terms of training speed and classification speed, especially for large databases. It is also easier to use since it does not require any input parameter related to its structure.

[1]  Josef Kittler,et al.  Pattern recognition : a statistical approach , 1982 .

[2]  R. H. Phaf,et al.  CALM networks: a modular approach to supervised and unsupervised learning , 1989, International 1989 Joint Conference on Neural Networks.

[3]  M. Teague.  Image analysis via the general theory of moments , 1980 .

[4]  Stephen Grossberg,et al.  ART 2-A: an adaptive resonance algorithm for rapid category learning and recognition , 1991, IJCNN-91-Seattle International Joint Conference on Neural Networks.

[5]  Stephen Grossberg,et al.  ARTMAP: supervised real-time learning and classification of nonstationary data by a self-organizing neural network , 1991, [1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering.

[6]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[7]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1994 .

[8]  D. Wehner.  High Resolution Radar , 1987 .

[9]  Stephen Grossberg,et al.  Fuzzy ART: Fast stable learning and categorization of analog patterns by an adaptive resonance system , 1991, Neural Networks.

[10]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[11]  G. Bartfai,et al.  Hierarchical clustering with ART neural networks , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[12]  Thomas Kailath,et al.  Detection of signals by information theoretic criteria , 1985, IEEE Trans. Acoust. Speech Signal Process..

[13]  Torsten Söderström,et al.  Statistical analysis of MUSIC and subspace rotation estimates of sinusoidal frequencies , 1991, IEEE Trans. Signal Process..

[14]  Stephen Grossberg,et al.  ART 3: Hierarchical search using chemical transmitters in self-organizing pattern recognition architectures , 1990, Neural Networks.

[15]  King-Sun Fu,et al.  Shape Discrimination Using Fourier Descriptors , 1977, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  D. C. Munson,et al.  Wide-angle ISAR passive imaging using smoothed pseudo Wigner-Ville distribution , 2001, Proceedings of the 2001 IEEE Radar Conference (Cat. No.01CH37200).

[17]  Ernest L. Hall,et al.  Three-Dimensional Moment Invariants , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Allen Gersho,et al.  Vector quantization and signal compression , 1991, The Kluwer international series in engineering and computer science.

[19]  André Quinquis,et al.  Superresolution ISAR image classification using Fourier descriptors and SART neural network , 2004 .

[20]  Stephen Grossberg,et al.  Adaptive pattern classification and universal recoding: II. Feedback, expectation, olfaction, illusions , 1976, Biological Cybernetics.

[21]  David G. Stork,et al.  Pattern Classification , 1973 .

[22]  Stephen Grossberg,et al.  ART 2-A: An adaptive resonance algorithm for rapid category learning and recognition , 1991, Neural Networks.

[23]  André Quinquis,et al.  A New Method for Estimating the Number of Harmonic Components in Noise with Application in High Resolution Radar , 2004, EURASIP J. Adv. Signal Process..

[24]  Guszti Bartfai,et al.  An ART-based modular architecture for learning hierarchical clusterings , 1996, Neurocomputing.

[25]  E. Radoi,et al.  Some radar imagery results using superresolution techniques , 2004, IEEE Transactions on Antennas and Propagation.

[26]  Etienne Barnard,et al.  Two-dimensional superresolution radar imaging using the MUSIC algorithm , 1994 .

[27]  Gerald Sommer,et al.  Pattern Recognition by Self-Organizing Neural Networks , 1994 .

[28]  Stephen Grossberg,et al.  A massively parallel architecture for a self-organizing neural pattern recognition machine , 1988, Comput. Vis. Graph. Image Process..

[29]  M. Desvignes,et al.  Déroulement de phase : application à la correction de distorsions géométriques en IRM , 2000 .

[30]  Stephen Grossberg,et al.  Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional maps , 1992, IEEE Trans. Neural Networks.

[31]  S. Grossberg,et al.  ART 2: self-organization of stable category recognition codes for analog input patterns. , 1987, Applied optics.

[32]  Jerry L. Prince,et al.  Snakes, shapes, and gradient vector flow , 1998, IEEE Trans. Image Process..

[33]  Thomas Kailath,et al.  On spatial smoothing for direction-of-arrival estimation of coherent signals , 1985, IEEE Trans. Acoust. Speech Signal Process..

[34]  V. G. Nebabin.  Methods and techniques of radar recognition , 1994 .

[35]  James M. Keller,et al.  A fuzzy K-nearest neighbor algorithm , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[36]  Ming-Kuei Hu,et al.  Visual pattern recognition by moment invariants , 1962, IRE Trans. Inf. Theory.