Please use this identifier to cite or link to this item:
|Document Type: ||Thesis or dissertation|
|Title: ||Analysis of performance of automatic target recognition systems|
|Authors: ||Marino, G|
|Supervisors: ||Hughes, Dr E J|
|Issue Date: ||22-Aug-2012|
|Abstract: ||An Automatic Target Recognition (ATR) system is a sensor which is usually
able to recognize targets or objects based on gathered data. The application
of automatic target recognition technology is a critical element of robotic warfare.
ATR systems are used in unmanned aerial vehicles and cruise missiles.
There are many systems which are able to collect data (e.g. radar sensor,
electro-optic sensor, infra-red devices) which are commonly used to collect
information and detect, recognise and classify potential targets. Despite significant
effort during the last decades, some problems in ATR systems have
not been solved yet.
This Ph.D. tried to understand the variation of the information content into
an ATR system and how to measure as well as how to preserve information
when it passes through the processing chain because they have not been
investigated properly yet. Moreover the investigation focused also on the
definition of class-separability in ATR system and on the definition of the
degree of separability. As a consequence, experiments have been performed
for understanding how to assess the degree of class-separability and how the
choice of the parameters of an ATR system can affect the final classifier performance
(i.e. selecting the most reliable as well as the most information
As results of the investigations of this thesis, some important results have
been obtained: Definition of the class-separability and of the degree of classseparability
(i.e. the requirements that a metric for class-separability has
to satisfy); definition of a new metric for assessing the degree of classseparability;
definition of the most important parameters which affect the
classifier performance or reduce/increase the degree of class-separability (i.e.
Signal to Clutter Ratio, Clutter models, effects of despeckling processing).
Particularly the definition of metrics for assessing the presence of artefacts
introduced by denoising algorithms, the ability of denoising algorithms in
preserving geometrical features of potential targets, the suitability of current
mathematical models at each stage of processing chain (especially for clutter
models in radar systems) and the measurement of variation of information
content through the processing chain are some of them most important issues
which have been investigated.|
|Appears in Collections:||PhD, EngD, MPhil and MSc by research theses - Cranfield Defence and Security, Shrivenham|
Items in CERES are protected by copyright, with all rights reserved, unless otherwise indicated.