So welcome to the Monday session, 45 minutes today.

And we will briefly hook up to the situation

where we stopped last time.

And today I will talk a little bit about

fair parameterization and different ways

to solve least square problems.

Good, so what was the point where we stopped last week?

Good, let me just here introduce one additional slide.

We were considering x-ray systems.

And with x-ray systems today we have

two different technologies to convert the x-rays

that were propagated through the human body

and their energy as intensity images.

And the two detector technologies we are considering

are image intensifiers, intensifiers,

II and flat panel detectors.

And each of these detector systems have some drawbacks

and have a certain impact on the image quality.

And we started to consider image intensifiers.

And the idea of image intensifiers is that

you use an electron optics to amplify the signal.

And having moving electrons in the earth magnetic field,

well we know what happens, they are deviated.

And these deviations cause distortions in the image

and these distortions have to be

calibrated and eliminated by a proper software tool.

And the situation is as follows

that we are currently considering.

We have the undistorted image.

We say the distorted and the distorted one.

Is it readable?

It's not that black as I would like to see it.

And what we do is we have here a mapping

that takes the X and Y values

and maps it to X and Y values over there.

And this is the undistortion function.

And this function here is a F1, F2 function

taking X and Y, X and Y.

And this leads to X prime, X prime,

and this leads to X prime and Y prime

in the distorted image.

So we map here these two coordinates

to these two coordinates by using F1 and F2 as a mapping.

Okay, you remember that?

And then we said how can we, first of all, represent them,

the functions F1 and F2,

that's the problem of model selection.

Yeah, first problem, model selection.

That's the representation of F1 and F2.

And for instance, we have seen that we can use polynomials.