1
$\begingroup$

I don't really know much about 3D scans and so on, so I'm just curious about it. I need to scan objects with different forms and I want to know which level of details I can reach during the scan. For example, if the object has a small (0.1 mm) black point on it, will this be detected? Or if the object has a really small scratch (barely eye-visible), will this be reproduced as a 3D model? So how many details deep can I go and at which price? (More or less)

Or do you know if there is any other technology that can do such things? For example, I can see those small scratches or defects on a good camera, the problem is that camera does not scan the entire 3D object, but I would need to shoot each angle of it manually. What can I do to scan the entire object and get deep details of it?

Improve this question
$\endgroup$
2
  • $\begingroup$ A lot of 3d scanning is done with cameras...so what's the cost of a good high res camera? Or, alternate viewpoint, can you afford to store and process the resulting extremely large high res model? And now that you have it, you'll have to reduce the resolution to print it. $\endgroup$
    – user10489
    Commented Feb 5, 2023 at 21:28
  • $\begingroup$ Ok thank you so much, so basically cameras are good enough I would say $\endgroup$
    – g3Dscans
    Commented Feb 6, 2023 at 12:02

1 Answer 1

Reset to default
1
$\begingroup$

Structured Light Scanning uses a turntable and a dedicated light source that creates patterns on the scanned object. The patterns are interpreted by the software to generate a 3D model. Photogrammetry uses a series of photographs, sometimes on a turntable, sometimes not. The software that does not support turntable use tends to use the apparent background movement to assist in creating reference points to generate the model.

The resolution aspect of your question is dependent on the resolution of the camera and the ability to detect and determine any variations in height/depth/width, etc. If there is insufficient camera resolution (Macro lens), the feature you desire to appear will not appear.

Photogrammetry and SLS are susceptible to specularity and "invisible blacks." The former is the result of a reflection of light, causing a bright spot that interferes with the calculations and dimensional determination. Transparent objects also have a similar problem. Blacks on an object are easy to detect by the human eye, often because they contain specularity as well. The dark sections are erroneously detected by the software as depth or detected as gaps/chasms/open sections. Spray-on powders provide a more uniform surface to remove these problems, but may not be suitable to your purposes.

Multiple cameras is one method used to deal with the necessary angles of captures. President Obama was the subject of a fairly complex photogrammetry construction as well as the subject of a series of hand-held structured light scanning image.

Image below from linked site:

obama photogrammetry

Improve this answer
$\endgroup$
3
  • $\begingroup$ Wow, thank you so much. So structured light scanning would be one way to go. But I think I also need to turn ON/OFF some lights to detect some small details in the glass (some details are only visible at certain angle with certain light. Is this still possible? $\endgroup$
    – g3Dscans
    Commented Feb 6, 2023 at 12:07
  • 1
    $\begingroup$ You should expect best results with consistent lights. SLS creates its own light patterns and only ambient light should be used with this process. Direct lighting will wash out the projected patterns. SLS will pick up those details you reference if the operator changes angles appropriately during the scanning. $\endgroup$
    – fred_dot_u
    Commented Feb 6, 2023 at 14:29
  • $\begingroup$ That, and light patterns swap between four different bar patterns usually: two horizontal, two vertical ones. $\endgroup$
    – Trish
    Commented Feb 9, 2023 at 14:05

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .