In this section we describe the method for multi-net response
integration in a general form, even if in this paper we applied it
to solve the face recognition problem. It is necessary to explain a
generalized view of the method and show how it can be adapted
to modular or ensemble multi-net structures, and to applications
that can be solved with neural networks like pattern
recognition or time series prediction. Depending on the type of
application some blocks need to be developed in a different form