Choice of technology continued. Photogrammetry provides a practical way to make 3D scans of landscapes and other large objects where it is impractical to project structured light onto the surface.
- Like Mount Chiginaugua.
Alternatively, structured light can be used to assist in scanning objects that do not have enough texture of their own for photogrammetry.
Also, another technique with similar characteristics is depth-of-field 3D scanning, using the contrast of a large depth-of-field lens to compute depth. However, this is another technique that requires surface texture in order to work well, so the same rule applies with structured light.
Another idea for 3D scanning. Rather than sweeping a laser across an object, you entirely use contrast thresholding and sweeping the camera focus with a narrow depth-of-field lens to determine depth. Advantage: faster photography is possible with a wider aperture, and there is better noise stabilization in the recorded images. Disadvantage: it might not be possible to increment the focus of your camera lens in as small increments as are the case with a low-cost, continuous sweep DC motor.
Again, note that you may need to project a light pattern onto your object in order to get sufficient surface texture.
Okay, wait. For 3D scanning via laser triangulation, in order to obtain good results fast, you’re going to want to be able to take single photographs with the entire object in complete focus. That means that you need a small aperture size. Hey, and guess what? Smartphone cameras already come in a small aperture size. Does that mean that the extra sensor area of a DSLR camera is not needed. I’m wondering what the difference is between sensor size is with identical aperture size. Let’s see if I can find good information about this on the Internet. Indeed I can, just not so much on Wikipedia. For starters, that is.
20161215/DuckDuckGo small aperture dslr versus smartphone camera
Why aperture matters on your smartphone camera
20161215/http://newatlas.com/smartphone-camera-aperture/42524/
Camera sensor size: Why does it matter and exactly how big are they?
20161215/http://newatlas.com/camera-sensor-size-guide/26684/
Okay, so here’s the thing. Apparently, larger sensor size can get a lower noise floor even with the same aperture size. BUT, more importantly, from reading the first article, it appears that DSLR camera sensors simply have more advanced electronics in the fundamental characteristics. In particular, they can take photos at much faster ISO film speeds than are the case with compact smartphone image sensors, which, in comparison, are distinctively limited in their available parametrization.
SOoo, the point I wanted to make was that in theory, as low-cost small-aperture smartphone camera sensors improve, it should be entirely possible to take excellent quality 3D scans using those low-cost sensors. In the mean time, however, if you want additional sensor quality characteristics, you’ll more likely find those in a DSLR camera.
Okay, I’ve got to make this point absolutely clear with a good explanation. Why do you want to orient your laser line vertically? Well, here’s the point. If you place your laser module and your camera with horizontal adjacency, and the laser line shines completely level at one point, you won’t be able to extract any depth information from such an image, and that would effectively correspond to wasted time. Also, the frames just around that particular frame will have less precision in their depth information. But, when the alignment of the camera, laser module, and laser line are perpendicular to each other, there is never such a frame where that is an issue. So, that’s the reason why you need that camera alignment.