From: "Vladimir Galogaza" 
To: "ATM shore" 
Reply-To: "Vladimir Galogaza" 




>I've always had the intuitive suspicion that zone radii errors were
>Robo-Foucault's Achilles' heel.
> J. Burrows.

Achilles' heel was fatal fault (for Achilles). Accuracy of the zone
determination is not fatal (for Robotest , in its JL incarnation). It is
influencing its accuracy. To what extent, we shall see soon.

It seems that I  forgot what has caused RoboFoucault investigation. As far
as I remember it was observation ( by Robo author) that results are
sensitive ( depending) on the source ( or image ) intensity. But later this
aspect was not mentioned any more nor resolved. Instead efforts are
concentrated on the comparison of the Robo results with
 " absolute truth" that is to find out absolute accuracy ( and
repeatability) of
RoboFoucault test.
Initial  intensity related problem is forgotten or under a carpet. Why is that so?

I see no proof (supported by  extensive M. Peck results) that Robo zone
determination random error is in any way larger than  errors done by
average ATMer applying Foucault manual and "ocular" methodology.
Robo systematic error ( in central zones, in edge detection,...) is now
revealed enabling estimate of Robo inherent capability and may lead to
modification of Robo to implement lateral wire test as suggested by James
Burrows.

Vladimir.

--- BBBS/NT v4.01 Flag-4