Lucanus cervus - female
(NL - Vliegend Hert) -
Scanned with the MICROscanR - a 3D focus-stacking robot: www.microscanr.com
"MINIMUM EFFORT test - workflow
______________________________________
This was the first scan with the MICROscanR that was executed fully automated. After putting the specimen on a pin and mounting it on the robot, I marked 4 Viewpoints where the depth of the specimen was to be measured. That depth is used for the dynamic assignment of the stack-depth in each viewpoint. After that I had the software select 29 other viewpoints to measure and continue scanning when ready. All other 287 viewpoints had interpolated stack depths. Measurement took 15 minutes and scanning took 2h10m. So 2H25m total robot time.
Then I started the fully scripted post production (set source and destination folders and press START) that creates the dataset for LFS or Postshot in COLMAP format. (total: 1h01m). Then I dragged the dataset into LFS, set the correct masking mode, left everything else default and it trained for 30 minutes. This splat is the result. So this was a minimum effort test.
Capture
_____________________________________
- Camera: SonyA7R5 camera (61 Mp) - 1/25 sec. - f16 - ISO 200 - JPG.
- Lens: Laowa 90 mm 1-2 x Macro at ~ 0.35 x
Scan
_____________________________________
- Scanned with: MICROscanR
- Total viewpoints (VPs): 320 (optimally distributed along a spherical scan shape)
- Average stack-depth per VP: 24 frames, ranging from 19-29 (dynamically assigned per viewpoint).
- Automatic Stack Depth measurement on 33 Viewpoints
- Total pictures: 7834
- Scan time: 2h10m.
- Scan date: 06-10-2026.
Splat
______________________________________
- Total splats: 1.000.000
- image resolution used: default LFS settings
- Software: Lichtfelt Studio
Lighting
______________________________________
- Lights: Object in diffuse light field between 2x side-beam round (28 cm) LED panels, 2 x 24 Watt
+ 1x Microscope LED beamer, fixed on front of lens.
+ 1x 40 Watt led 120x30 cm led panel from above
- Background: black velvet.
On the specimen
_______________________________________
The specimen is one I bought online, and obviously they have used some kind of preservation method on the insect that glues some fine details together, so that is a shame and not something I will do more often. It was however a good test for the automatic depth measurement routine of the robot, because the black parts on a black background could have been a challenge, but the method used is relative to the noise floor so that is why even the darkest details get caught and all depth measurements have proven to be trustworthy.