Most modern pattern recognition filters used in target detection require a clutter-noise estimate to perform efficiently in realistic situations. Markovian and autoregressive models are proposed as an alternative to the white-noise model that has so far been the most widely used. Simulations by use of the Wiener filter and involving real clutter scenes show that both the Markovian and the autoregressive models perform considerably better than the white-noise model. The results also show that both models are general enough to yield similar results with different types of real scenes. (C) 2002 Optical Society of America.
History
Publication status
Published
Journal
Applied Optics
ISSN
0003-6935
Issue
32
Volume
41
Page range
6858-6866
Pages
9.0
Department affiliated with
Engineering and Design Publications
Notes
The paper reports the development of Markovian and autoregressive models to provide more realistic two dimensional clutter models. These have been incorporated into the transfer function of a pattern recognition Wiener filter we have developed. The performance achieved, using both our digital/optical hybrid correlator and all digital hardware implementations is much improved as compared to previous results obtained using models employing only white noise approximations. The clutter resistant filtering methods developed in this paper are to be incorporated as part of a plastic sorting process being implemented by Jayplas Ltd, www.jayplas.co.uk, Contact: J.Young, Tel: 01572-770060.