Anti-alias an intensity plot

[elset191]

I have data that I plot on both a surface plot and an intensity graph.  My goal is to get the intensity graph to anti alias so it looks nice, like the surface plot.  Anyone know of any clever, or not so clever ways to achieve this?

[altenbach]

Create a denser grid, then use a 2D interpolation to fill the gaps. LAbVIEW has a few 2D interpolations built in, or you could use my old bilinear interpolation posted here.

Of course if you have a mathematical description of your data, you could fit it and use the parameters to generate a surface at any desired resolution. One example would be a 2D gaussian as described here.

[altenbach]

(duplicate post)

[elset191]

I'm trying to decide if interpolating my data is acceptable or not, given the analysis I perform on it downstream.  Plust a lot of our scans have a lot of points (several times the image in my first post), so interpolation isn't as necessary in these cases. It may be that it doesn't matter at all, but I don't know

And, I certainly do not have any mathematical formula for the data, since the it is being acquired from our tool, and represents any arbitrary part we may be inspecting at the moment.

What I really want is an option to interpolate/anti-alias the graph just as an aesthetic thing, while not manipulating the data any.  I understand I can interpolate and graph that, but use the original under the hood, but in some cases my computer already has trouble handling just the graphing that I'm doing

[Darin.K]

You can always use a 3D surface plot and simply view the projection on the XY axis.   A little property jiggering and you have a better looking graph than the raw intensity graph.

[elset191]

The problem there is that the graph is interactive.  The user can click and drag to 'measure' the size of a feature on the graph, and I overlay images onto the intensity graph.