Neural Networks versus Image Pyramids
Horst Bischof, Walter G. Kropatsch · Artificial Neural Nets and Genetic Algorithms · 1993
In this chapter, neural networks ( section 2.1 ) and image pyramids ( section 2.2 ) are introduced in order to define the basic concepts. The goal is to provide a common vocabulary for these two fields. Section 2.3 gives a translation table of the vocabulary used in image pyramids and that used in neural networks. In section 2.4 , an example on satellite image (Land-sat TM) classification that speeds up considerably by the use of image pyramids is presented. This example serves two purposes: it shows how “conventional” neural algorithms can be applied to image processing and how they compare to classical (statistical) methods; and it demonstrates how much can be gained by the use of image pyramids.