CNN cloning template: shadow detector

A report on an extremely simple cloning template for cellular neural network (CNN) is presented. The template is capable of detecting the shadow cast by a two-dimensional image. The operation of the template is illustrated by an example. It is shown that when the shadow detector is used in addition to the connected-component detector (CCD), the recognition rate of handwritten characters (numerals, alphabets, symbols Japanese characters) increases significantly. This means that the output of the shadow detector contains a valuable feature, not revealed by the output from the CCD alone.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>