Page 24 - Classification Parameter Estimation & State Estimation An Engg Approach Using MATLAB

2

Detection and Classification

Pattern classification is the act of assigning a class label to an object, a
physical process or an event. The assignment is always based on meas-
urements that are obtained from that object (or process, or event). The
measurements are made available by a sensory system. See Figure 2.1.
Table 2.1 provides some examples of application fields in which classi-
fication is the essential task.
The definition of the set of relevant classes in a given application is in
some cases given by the nature of the application, but in other cases the
definition is not trivial. In the application ‘character reading for license
plate recognition’, the choice of the classes does not need much discus-
sion. However, in the application ‘sorting tomatoes into ‘‘class A’’, ‘‘class
B’’, and ‘‘class C’’’ the definition of the classes is open for discussion. In
such cases, the classes are defined by a generally agreed convention that

sensory pattern
object, system measurements classification class assigned to
physical process object, process or
or event event
measurement system

Figure 2.1 Pattern classification

Classification, Parameter Estimation and State Estimation: An Engineering Approach using MATLAB
F. van der Heijden, R.P.W. Duin, D. de Ridder and D.M.J. Tax
Ó 2004 John Wiley & Sons, Ltd ISBN: 0-470-09013-8