Surface and color problems after merging

Hi Katharina,
i did a lot of scanning tests the last days and noticed that merging problems were much less heavy when fusing the single scans with lower accuracy before. In advanced mode the best 0.08 accuracy often caused seam lines or merging failed completely, whereas switching to 0.12 or 0.15 yielded much better results. And processing was faster;-)
How come that? Do too much points overstress the merging algorithm? Or: the more points, the more usless and confusing noise?
Hoe does reducing fusing accuracy affect details? Do I have to try or are there some “rules”? In general, only accoury of the decoration is important to my, and this only insofar, as I want to see all the decoration (and its remains), even if it is badly preseverved, not wether it is 0.05 more to the right or not…

Cheers, Daniel

Hi Daniel , not really , MINI produces very much data for thr size of your object , and any difference in distances even slightly will show up while merging objects at that size.

The more points usually is a junk data , overlapped points and sometimes noises , mostly half of it is not usable , the scanner actually need minimal 12 frames for one 360 rotation to construct an object nit 360 frames or 2000K , so all the extras do not helps .

of course since the Algorithms removing the extra junk and overlapped points while fusing , leaving the most important frames that are needed , fusing is resolution ( Not volume accuracy ) and most of the objects people scanning don’t need 0.02mm accuracy , 90% of them don’t
it is like you trying to take a picture of your object with a microscope when you need just a simple phone .
The Advanced fusion is very powerful yet producing simple results and my favorite Algorithms since Handy Scan , it was later replaced with Standard Algorithms what was not always my to go option , so it is great to see it coming back again .

It don’t change the object volume only the top surface , Imagine a you scanned a smooth sphere , it will not change if you used 0.02mm or 0.1mm , however if you scanned a sphere with engraved details , the details may get slightly smoother if you fuse at large pitch point above 0.1mm but only if the object have truly details at 0.02mm … so to have the perfect settings you need to evaluate the object details , to determine if you really need 0.02mm or 0.2mm , are you going to print it at higher resolution , use higher pitch point , if you for plan to print it at 0.6mm, fuse it at 0.6mm

Please remember that you cant see 2 objects next to each other with the wide of 0.02mm with your naked eye , 0.04mm yes like for example 2 human hair at the distance of 15cm , so MINI is slightly better than human eye at 15 cm .

For that reason using maximum resolution not always creates the desired results , and as in your case where the object is long , a slightly difference in distance will change the resolution that is captured that why you have difference while merging , but when you reduce the resolution to 0.1mm then the level of details ( resolution ) is equal for both scans .

MINI was created to scan very small objects , best when used with turntable , at proper distance of 15-20 cm Max . So if your object is more than 15 cm you will get different resolution in details , imagine you projecting a image with a projector on your object … it will be not everywhere perfect picture , it will depends of a angle you are projecting , if the angle is wrong , you get less details while scanning .

But I think with extra practice you going to master it for your needs as you doing good job already … I don’t know if you got the POP3 already or not but it going to be perfect tool for you needs, and you can keep MINI for a very MINI objects :wink:

Thanks a lot!
The projection picture is a very good example - i understand that the angle of the light rays shouldn’t differ to much because of distortion issues, so with longer objekts standing upright on the turntable an the mini from above in an 45 or 60 degreee angle to the turntable the problems begin…
The angle to the object, not to the TT should be the same!
So i should scan longer objects not standing upright with their long sides, but rather with the long sides parallel to the TT plane …but this will cause more different distances on the other hand while the table turns.
The POP 3 is already at the customs in Germany, as I understood the last delivery notification. Looking forward to try it!

1 Like