Page 31 -

1.5 PK Project 17

3. Pre-processing. In some cases the features values are not directly fed into the
classifier or descriptor. For instance in neural net applications it is usual to
standardize the features in some way (e.g. imposing a LO, 11 range).
4. The class$cation, regression or descrktion unit is the kernel unit of the PR
system.
5. Posf-processing. Sometimes the output obtained from the PR kernel unit cannot
be directly used. It may need, for instance, some decoding operation. This, along
with other operations that will be needed eventually, is called post-processing.

- J
Classification I
Pattern * Feature Pre-processing + Regression / 4 Post-processing
Acquisition Extraction
P
Figure 1.12. PR system with its main functional units. Some systems do not have
pre-processing and/or post-processing units.

Preliminary analysis: Initial evaluation of
Choice of features feature adequacy
Choice of approach

Design of the Feature
classifier / descriptor

Training and test of the
classifier / descriptor

END

Figure 1.13. PR project phases. Note the feature assessment at two distinct phases.

Although these tasks are mainly organised sequentially, as shown in Figure
1.12, some feedback loops may be present, at least during the design phase, since