“Synthetic aperture” you ask? Yeah, that’s what people are trying to do with modern smartphone cameras to reduce the size, but they are using different terms like “light field.” Synthetic aperture is not a new idea, and the terminology has been used with radar for many years.
20180304/https://en.wikipedia.org/wiki/Synthetic-aperture_radar
20180304/https://en.wikipedia.org/wiki/Aperture_synthesis
20180304/https://en.wikipedia.org/wiki/Synthetic_aperture_sonar
20180304/https://en.wikipedia.org/wiki/Synthetic_Aperture_Ultrasound
20180304/https://en.wikipedia.org/wiki/Synthetically_thinned_aperture_radar
20180304/https://en.wikipedia.org/wiki/Beamforming
The best is by far the swept line methods in terms of noise. You can do swept line with either a laser module or a video projector. The video projector method is faster, but the laser method is cheaper.
There is also another video projector method where you shine a series of varying bar width patterns, one after another, which gives each “pixel” a binary code. Two cameras measure the same object, and that creates images where you can cast rays with common binary codes. The nearest points on those rays will give you a 3D point. This method is very sensitive to noise because one uncertain time value can completely throw off your binary code. Suffice it to say, it is not recommended from a quality standpoint, not to mention that the cost of such a method is higher too.
-
Again, I reiterate. All processes are the same in this regard. You start by collecting some sort of raw sensor data, then you decode the sensor data to get the 3D data format.
-
Additionally, most structured light 3D scanners suffer from another artifact reducing scan quality: secondary bounce light. The algorithms are designed to assume that only direct lighting from the structured light source illuminates the material. Secondary lighting from interreflections on the material effectively degrade the accuracy of the structured light methods.
-
These light artifacts mostly degrade the quality of scanning highly rough surfaces, but they do not significantly affect the quality of smooth surfaces, relative to the size of the laser.
-
Laser interferometry, which measures by detecting phase shift patterns, may be different in this regard, but it is prohibitively expensive and typically only used by scientific researchers, not by engineers or artists. Now, as you understand, that’s a euphemism for saying that the technology is not very widely used, period. In other words, a “commercial failure” from a commercial standpoint.
-
So, you were wondering about the accuracy of optical angle measurement compared to a manual or electronic device? What kind of accuracy do you really get from an HD video camera? Assuming the video camera field of view is 90 degrees, the angular width of a pixel is 0.09 degrees. Because in practice we cannot measure angles accurate to a single pixel, we round up to two pixels, effectively doubling the angular width to 0.18 degrees. Finally, we can be pessimistic and round up further to 0.5 degrees, because we might need space to identify a barcode pattern. So there you have it. You’ve been told that the turntable mechanism in FabScan is accurate to about half a degree, so optical measurement gets you that same angular accuracy, but at a lower cost because you do not need to buy your turntable devices. Yep, optical measurement is definitely the way to go, don’t take second place with a manual paper/cardboard turntable with reading angles off of a compass.