Rachel Dines interactive design print
contact
blog

CSC400 Explorations in 3D Scanning

a special studies by Rachel Dines, with advising by Joe O'Rourke at Smith College

I have done some research on 3D scanners, and determined that there are two main kinds: Contact and Non-Contact Scanners.

  • contact scanners scan by physically touching the object
  • one example of this is a CMM (coordinate measuring machine), which uses a mechanical probe to discern the coordinates of the object in space
  • contact scanners are used mainly in manufacturing and in animation (for digitizing clay models)
  • there are TWO MAJOR DISADVANTAGES:
    • Since the object that is being scanned actually must be touched by the scanner, it is not suitable for ancient or delicate objects that could be damaged and/or changed
    • Contact scanners are very slow when compared to non-contact scanners, as they are mounted on a physical arm that can only move so fast. The fastest CMMs can only operate on a few hundred hertz, which is very slow when compared to an optical system like a laser scanner, which can operate from 10 to 500 kHz.
Non-Contact Scanners
  • active scanners send out beams of radiations and then record their relection off the object. There are many different kinds of active 3D scanners
  • there are several DISADVANTAGES of active scanners:
    • dark colors, especially red and black, do not show up well on these types of scanners, because they absorb light.
    • transparent surfaces do not show up on the scans because the beams of radiation pass right through them
  • TIME OF FLIGHT SCANNERS
    • these scanners send out beams of light and time how long it takes for the light to return after it has bounced off the object
    • since the speed of light is known, the scanner can then calculate the distance the light has traveled to the object and make a map of it in space
    • Time of Flight scanners can operate over very large distances, so they are better for scanning buildings and other large objects
    • ONE DISADVANTAGE of the Time of Flight scanners, is that they are not the most precise scanner because it makes calculations with the very high speed of light, which leaves room for a large margin of error
  • TRIANGULATION SCANNERS (this is the type of scanner we have)
    • these scanners shine lasers at an object and use a camera to calculate its location.
    • the camera, laser origin and the point make a triangle, and simple trig is then used to calculate the unknown point
    • ONE DISADVANTAGE of Triangulation Scanners, is that they cannot scan very large objects, however, they do have a very high degree of accuracy and detail
  • CONOSCOPIC HOLOGRAPHY SCANNERS
    • this scanner projects a laser beam onto the object and the reflection along the same path is put into a conoscopic crystal and the resultant defraction pattern can be analyzed to determine a point's location
  • STRUCTURED LIGHT SCANNERS
    • this scanner projects a pattern of light onto an object and then a camera analyzes the distortion of the pattern to determine the distance to each point
    • ONE MAJOR ADVANTAGE of using a Structured Light Scanner is its speed--it scans multiple points at a time, unlike the other systems that can only handle one point at a time.
  • MODULATED LIGHT SCANNERS
    • these scanners shine constantly changing light onto the object, while a camera detects the light and how much it shifts, and uses this information to calculate how far the light has traveled.
  • SLICE SCANNERS
    • these scanners create "slices" of the object by using Ultrasound, MRI, or Confocal Microscopes. the slices can be used to compute volumetric data, which can be converted into polygon surface models
  • passive scanners do not send out any kind of radiation themselves, but rather, use the ambient radiation around them
  • ONE MAJOR ADVANTAGE of passive scanners is that they can be very inexpensive
  • there are several different kinds of passive scanners:
  • STEREOSCOPIC SCANNERS
    • these scanners consist of two video cameras a certain distance apart both pointing at an object
    • by analyzing the differences in the scenes seen by the two cameras, a 3D model can be made of an object
  • SILLHOUETTE SCANNERS
    • these scanners employ a squence of photographs taken around an object against a contrasting background. the pictures are used to find the hull of the object
    • the MAJOR DISADVANTAGE of this method is that concavities cannot be detected

Most of my information comes from Wikipedia (for better, or for worse), and from VR News Technology Review .

 


| home | interactive design | print | blog | contact |