On the possibilities of using a class of CNN's for texture classification
Ion Vornicu, Liviu Goraş · 2011
The paper deals with the possibility of using the spatio-temporal dynamics of a class of CNN's for texture classification. The principle of the proposed architecture is discussed and the design of a bank of spatial filters for this kind of application is presented. Transistor level simulation and considerations regarding the architecture reconfiguration will be given as well.