The parallelpiped classifier (often termed multi-level slicing) divides each axis of multi-spectral feature space, as shown in an example in Figure 11.4.1. The decision region for each class is defined on the basis of a lowest and highest value on each axis. The accuracy of classification depends on the selection of the lowest and highest values in consideration of the population statistics of each class. In this respect, it is most important that the distribution of population of each class is well understood.
The parallelpiped classifier is very simple and easy to understand schematically. In addition the computing time will be a minimum, when compared with other classifiers.
However the accuracy will be low especially when the distribution in feature space has covariance or dependency with oblique axes. Orthogonalization should be undertaken using principal component analysis, for example, before adopting the parallelpiped classifier.
Figure 11.4.2 shows an example of classification with the use of the parallelpiped classifier.
Copyright © 1996 Japan Association of Remote Sensing All rights reserved
