The Input Layer


You do not have to read or understand this section unless you want to!

Also, the Input Layer is only applicable to users of the Scan2CAD Pro version, so you don’t have to worry about this if you use the Lite version.


When it is being trained to recognize a font a Scan2CAD neural network is made up of three parts called “layers” – the Input Layer, the Hidden Layer and the Output Layer. See Advanced neural network information for a diagram.

The Input Layer is a grid of 12 x 16 (192) pixels that allows the example characters in the training set to be presented to the neural network in a consistent manner for learning.


When you add an example character to the training set Scan2CAD “standardizes” it by scaling it to fit within the Input Layer.

Each pixel in the Input Layer is termed a “node” and can either be black or white depending on how each example character appears on the grid.



Have questions on this topic? Talk to us