PPKE ITK PhD and MPhil Thesis Classes
PPKE ITK PhD and MPhil Thesis Classes
PPKE ITK PhD and MPhil Thesis Classes
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
1.2 Cellular Neural/Nonlinear Network - Universal Machine 7<br />
b<br />
0.5<br />
a<br />
0.25<br />
-0.125<br />
-0.18<br />
-0.5<br />
0.18<br />
v uij<br />
-v ukl<br />
-2 -1<br />
v yij<br />
-v ykl<br />
(a)<br />
(b)<br />
Figure 1.3: Zero- (a) <strong>and</strong> first-order (b) nonlinearity<br />
In case of the zero-order nonlinear templates, the nonlinear functions of the<br />
template contains horizontal segments only as shown in Figure 1.3(a). This kind<br />
of nonlinearity can be used, e.g., for grayscale contour detection [21].<br />
In case of the first-order nonlinear templates, the nonlinearity of the template<br />
contains straight line segments as shown in Figure 1.3(b). This type of nonlinearity<br />
is used, e.g., in the global maximum finder template [21]. Naturally, some<br />
nonlinear templates exist in which the template elements are defined by two or<br />
more nonlinearities, e.g., the grayscale diagonal line detector [21].<br />
1.2 Cellular Neural/Nonlinear Network - Universal<br />
Machine<br />
If we consider the CNN template as an instruction, we can make different algorithms,<br />
functions from these templates. In order to run these algorithms efficiently,<br />
the original CNN cell has to be extended (see Figure 1.4) [22].<br />
extended architecture is the Cellular Neural/Nonlinear Network - Universal Machine<br />
(CNN-UM). According to the Turing-Church thesis in case of the algorithms,<br />
which are defined on integers or on a finite set of symbols, the Turing<br />
Machine, the grammar <strong>and</strong> the µ - recursive functions are equivalent. The CNN-<br />
UM is universal in Turing sense because every µ - recursive function can be<br />
computed on this architecture.<br />
The