Aircraft Classification from Estimated Models of Radar Scattering
Abstract:
In this paper a novel method for estimating point scatter models from simulated HRR range profiles is introduced. This method uses the positions of amplitude peaks in a set of profiles to find a maximum likelihood solution for the parameters of a point scatter model of radar scattering. Occlusion and interference effects are also included in the model parameters. An Expectation-Maximisation algorithm is used to find the maximum likelihood solution of the model parameters, with the set of assignments between peaks and scatterers playing the role of hidden variables. The expectation step is implemented as a Markov chain Monte Carlo sampling method. The point scatter models constructed using this method are capable of accurately predicting the distribution of peak locations in profiles at any aircraft pose. Classification of an independent test set containing simulated profiles yield
Citations
| 5 | Mututal modeling of teammate behavior – Kok, Vlassis - 2002 |
| 4 | The interface specification and implementation internals of a program. module for geometric algebra – Zaharia, Dorst, et al. - 2002 |
| 3 | Computer graphics from a geometric algebra perspective – Zaharia - 2002 |
| 3 | The Generative Self-Organizing Map: A Probabilistic Generalization of Kohonen’s SOM – Verbeek, Vlassis, et al. - 2002 |
