Thickness variation estimate from 3D point cloud

Is there any way to estimate thickness variation in my ~ 2"X 4" steel plate approximately 4 mm thick?. I have the 3D point cloud scanned and imported into MATLAB. Thanks.

[darova]

Can you attach the data?

[John D'Errico]

There is no question in this about MATLAB. And you don't even say what information you really have. Do you have the thickness measured at various points? Or do you have a measurement of z at many points on the top of the plate, and then again on the bottom of the plate, perhaps at different points? Is the plate rectangular? Do you have the same locations measured top and bottom?

Next, you give no information about what you even mean by thickness variation. There are multiple ways I could imagine to quantify the variation of a measurement. But until you decide exactly what you mean to do in terms of mathematics, you cannot write code to perform that computation.

[DAMILOLA BALOGUN]

Thanks for you response.

I have measured the Zs of many points on both sides of the plate. I figured out that i could simply subtract the Zs of corresponding points(top and bottom) and then obtain the thickness variations. But it may be difficult to align perfectly when we flip the sample the other way round.

A second option which I thought I could use MATALB is to scan top and bottom separately, plot the two surface maps and then merge them in some way, Im not sure, just thinking out loud.

By thickness variation, I mean that I need to determine the variation in zs for such a sample as shown in the picture below.

[darova]

A second option which I thought I could use MATALB is to scan top and bottom separately, plot the two surface maps and then merge them in some way

I like that idea. YOu can use griddata to create a surface from point cloud

[DAMILOLA BALOGUN]

Thank you, how then is thickness varition estimated from such plots?

Thanks

[darova]

Create two surfaces and substract each from another

Answer 1

[Sri Harsha Kondapalli]

If I understand the question correctly you were asking about the thickness variation across whole slab. so I would suggest you to have a reference point by drilling a hole through the sample and then gather your point cloud data for both front and back of your sample.

Later you can align the drill hole from the point cloud data and obtain the thickness variation information.

Also consier approximate mean thickenss as the seperation while combining the front/back point cloud data.

[DAMILOLA BALOGUN]

Thanks. Sounds very reasonable , and worth a try.