Re: correction : ts calibration

Hello,

> > For better understanding the value given by the touchscreen. I have got
> the
> > raw values when the pen follows the borders of the screen.
> > It gives this :
> > http://alain2210.free.fr/geombrut.pdf
> Pretty sad. Until now I have assumed all touchscreens have a linear
> response,
> but now I see that at least h22xx doesn't. I don't know yet how a general
> transform equation should look like, but it will have to be more complex
> than
> just multiplying an 1x2 vector by a 2x2 matrix.

So I have tested another transformation to get a "linear" touchscreen. The
result is better
but the constants still not optimal.

xcal = ((x * 89) >> 8) - 58;
ycal = 320 * (((9 * x * x) >> 16) - ((37 * x) >> 8) + 905 - y) / ((((9 + 18) * x
* x) >> 16) - (((37 + 32) * x) >> 8) + (905 - 326));
The idea of the formula above is to know ymax(x) and ymin(x) the above and
below curves and to say that ycal= 320 *(y - ymax(x)/(ymax(x)-ymin(x).
But the difficulty is that is it more non linear near the border....

> How the calibration process should look like? Does it make sense to
> display something like "Please follow with the pen the borders of the screen
> in the following order: top-left - top-right - bottom-right - bottom-left -
> top-left"? Maybe this calibration process will be even easier to accomplish
> than clicking four or five screen points. Perhaps a final step would be to
> click the central screen point in order to detect non-linearity within the
> touchscreen values area. Oh.

It could be that the calibration work after this linearization to ajust it
finer. I do not know.

Sincerly,

Alain
Received on Sat Apr 17 2004 - 14:38:30 EDT