Andrew Zabolotny wrote:
>Do you really mean *pixels* or the units read from touchscreen? As you can
>see, the range of the units is approximatively three-four times bigger than
>screen resolution, so if you manage to get a 3 *units* error, it would mean
>you have almost pixel-accurate linearization.
>
>
The errors are measured between the measured touchscreen values
transformed into screen space and the coordinates of the reference
points. Unfortunately the errors are therefore in actual scren pixles :-(
As both Gwyneth and Benny have made additional datasets from their
IPaqs, i now have an even better data set and will try to refine the
model. I hope i will have time for this tomorrow. I think it is a good
idea to stick with the polynomial model and just try to calculate a
better set of constants.
/Lau
Received on Wed May 19 2004 - 10:50:40 EDT
This archive was generated by hypermail 2.2.0 : Mon Jul 25 2005 - 17:19:27 EDT