Bill,
How does microscopy compare with scanning in your experience? I have accumulated hundreds of pictures using microscope-mounted cameras but have never matched either the resolution or reproducibility that I get using a simple scanner. Optical pictures are so sensitive to surface condition and lighting that I sometimes dispair of getting reproducible results.
There would be great merit in developing an automated vision system having the properties that you mention - if it was possible! Image manipulation programs are readily available that calculate mean diameter and number of isolated craters in a given area. They start to fail as soon as craters overlap - are they treating a pair of overlapping indents as one or two objects? Treated as a pair, the 'diameter' and number results are obviously false. It is also virtually impossible to identify tiny areas of unpeened surface.
Scanning has the practical advantages of (1) being available in every shot peening plant (whereas it is exceptional to find microscopes and cameras) and (2) not requiring specific skills or training (virtually everyone knows how to use a scanner).
David
