An experimental system for optical detection of layout errors of printed circuit boards using learned CNN templates

Cellular neural networks (CNNs) are considered as cellular analog programmable multidimensional processing arrays with distributed logic and memory. The interconnecting weights between the neighboring processing elements are defined by the temperature values. A systematic way to find robust templates is presented. Using the new learning algorithm some templates were found for a CNN based layout design rule checking algorithm. The algorithm has been tested in an experimental system with real life examples. A typical design rule checking of a 432-pixel×164-pixel area takes 8 s of computation time on the CNN hardware accelerator board