Detection and classification in a high clutter environment
Abstract
A new system for detecting and classifying signal waveforms in the presence of non-stationary, high intensity clutter is presented. The system is developed within the context of detecting point targets from a sequence of image frames, but is not limited to problems of this kind. System performance is demonstrated using measured data. The importance of clutter suppression as a preprocessing step to reduce false alarms and improve signal to clutter ratio is emphasized. A new clutter suppression technique is introduced that makes only limited assumptions concerning the statistical distributions of the clutter and targets. Finally, a new family of fast and accurate sub-space tracking algorithms are also introduced which help to make the clutter suppression and classification systems realizable in real time. The applicability of these sub-space tracking algorithms to a wider class of problems than those considered here is noted. ^
Subject Area
Statistics|Engineering, Electronics and Electrical
Recommended Citation
Edward Christopher Real,
"Detection and classification in a high clutter environment"
(1997).
Dissertations and Master's Theses (Campus Access).
Paper AAI9812224.
http://digitalcommons.uri.edu/dissertations/AAI9812224