RMS

[aloe]

how to validate the RMS error? and is there a maximum value for valid error ?

[daniel]

Could you be more specific? I'm a bit confused about what you mean by 'validate' the RMS error...

[aloe]

how to ensure my results are valid? is there any inaccuracy error in the software itself?
thanks

[daniel]

Well, the internal computation are done with 32 bits floating point values, which generally induce a numerical error around 10^-7 (i.e. probably very far from the actual RMS you get ;).

The RMS (Root Mean Square) is computed in CloudCompare on the set of distances between a point cloud and another one. It's simply a statistical parameter that gives an information about the distances distribution, just as the mean and standard deviation. Depending on what you are actually measuring it can represent the quality of a registration process, the smoothness of a surface, the accuracy of a sensor, etc...

[aloe]

you mean that this error will lead to inaccuracy of my results and the RMS computed is not valid for comparison and research purposes ?

[daniel]

No the numerical error is ridiculous compared to any sensor's error.

And whether the RMS is suitable for a research work or not depends on what you want to do!

[aloe]

I'm comparing between the scans by same scanner between a tooth and the impression of the same tooth by overlapping them and check RMS for evaluating accuracy of impression......