3D Diffraction Rangefinder

Document Sample
3D Diffraction Rangefinder Powered By Docstoc
					On the Design, Construction and
   Operation of a Diffraction
                  MS Thesis Presentation
                      Gino Lopes
 A Thesis submitted to the Graduate Faculty of Fairfield University
    in partial fulfillment of the requirements for the degree of
  a Master of Science in the Electrical and Computer Engineering.
•   Problem
•   Approach
•   Motivation
•   Rangefinding
•   Design and Testing
•   Performance and Comparison
•   Conclusion
•   Future Work
• Design a diffraction rangefinder, subject to the
  following constraints:
  – Fit on a desktop,
  – Digitize and display objects,
  – Be affordable,
  – Be easy to use,
  – Not suffer from occlusion issues, characteristic of
    triangulation rangefinders,
  – Characterize the performance of the rangefinder
• Design a Prototype for testing.
  – Hardware
     •   Diffraction grating.
     •   Network Camera instead of USB camera.
     •   Laser line generator.
     •   Motion control hardware.
  – Software
     • JAVA was used for everything.
  – Layout of 3D Scanner
     • Dependent on hardware parameters.
• Diffraction rangefinders represent a new class
  of rangefinder for digitizing object.
• Verify predicted performance.
• Types of Rangefinders:
  – Shape to shading:
     • Process of computing the shape of a three-dimensional
       surface by looking at the brightness of one image of the
     • Shape to shading is difficult to implement.
      Rangefinding Continued
– Triangulation:
   • Finds the range-to-target by using two different views
     (angles) of the target, or by making use of off-access
   • Transmitter and receiver are separated.
   • Subject to shadows.
      Rangefinding Continued
– Light Detection and Ranging (LIDAR) system:
   • Uses laser pulse time of flight.
   • Receiver and transmitter can be co-axial and shadows
     and occlusion limitation are minimized.
   • For surface scanning the laser source or target would
     need to be moved in both the x-axis and y-axis.
      – To collect enough data points to reproduce the surface detail.
      Rangefinding Continued
– Diffraction Rangefinders:
   • Measures the distance to a target by reading the
     curvature of the wave front.
   • Work with (active illumination) using a laser.
   • Less susceptibility to occlusion.
   • Receiver and transmitter can be co-axial.
   • Limitation in range of measurement due to size of the
• Scanner Design:
  – Illumination Source:
     • Off the shelf red laser line generator
  – Vision System:
     • Network Camera
     • 1000 line/mm Diffraction Grating
  – Motion System:
     • Motor and controller.
     • Rotary Turntable.
2D View of Scanner Layout
2D View Of Scanner Layout Cont.
3D Scanner Prototype
• Testing of scanner performance.
  – Calibration wedge used as a resolution target.
     • Target with known dimensions.
     • Verification of operation
Scanner Test Configuration
           Testing Continued
• Calibration wedge was positioned at 49mm,
  92mm, and 135mm from grating.
Wedge at 49mm
After Processing at 49mm
Wedge at 92mm
After Processing at 92mm
Wedge at 135mm
After Processing at 135mm
           Scanner Comparison

• Scanner characteristics was compared against
  two other scanners on the market.

  – One from VXTechnologies.

  – One from Cyberware.
         Scanner Comparison Continued
                      3D Scanner               StarCam                   Cyberware

Field of View   12" X 7" (310mmX178mm)   21" X 16" (533mmX406mm)   14" X 17" (350mmX440mm)

Resolution          0.017" (0.44mm)          0.019" (0.48mm)           0.015" (0.38mm)

Width                    11.5"               16.375” (416mm)           188.2 cm (74.1")

Height                   14"                 11.000” (280mm)           205.3 cm (80.8")

Length                   30"                  9.250” (235mm)              Not Given
3D Image of Chess Piece
3D Image of Chess Piece Cont.
3D Image of Chess Piece Cont.

Average resolution of the 3D Scanner was
     between 0.43mm and 0.44mm.
(Comparable to other rangefinders on the market)
               Future Work
• Replacing the turntable with an improved
• Replacing the Lego motor and RXTX controller
  with a stepper motor and controller.
• Increasing the laser fan angle from 60.
        Future Work Continued
• Replacing the camera with one that allows for
  turning automatic gain off.
  – Reduce noise and blooming.
• Improve image acquisition and processing
  software user interface.
• Verify repeatability of scanner.
                                           Data Analysis
• Using Grating equation to calculate dispersion
  angle of 1000 line per mm grating.

                 Number of slits per mm (q):                          1000

                       One mm in meters:                              0.001

  Center to center distance between slits (p) in meters (1mm/q):    0.000001

         Wavelength of light source (lambda) in meters:            0.000000629

                      Diffraction Order (n):                           1

       Dispersion Angle (sin(a)=(n*lambda)p) in degrees:               39
                            Data Analysis Cont.
• Using trigonometry to calculate mm per pixels
  from acquired data.

          Calculated dispersion Angle of grating:                39

          Distance from grating to target (D) in mm:            135

                           Tan(b):                              0.806

  Distance between zero-order and first-order fringes in mm:   108.865
                         Experimental Data
• Average Number of Pixels:

    Distance to Target      Between Zero Order and First Order on Right   Between Zero Order and First Order on Left
                                               Side                                         Side

          49mm                               301.029                                       323.206

          92mm                               260.559                                       266.059

         135mm                               201.735                                       207.088
   Experimental Data Continued
• Pixels per mm.

                         Between Zero Order and First Order on Right   Between Zero Order and First Order on Left
    Distance to Target
                                      Side (pixels/mm)                              Side (pixels/mm)

          49mm                              2.765                                        2.969

          92mm                              2.393                                        2.444

         135mm                              1.853                                        1.902
    Experimental Data Continued
• Average distance resolvable.

                         Between Zero Order and First Order on Right   Between Zero Order and First Order on Left
    Distance to Target
                                         Side (mm)                                    Side (mm)

          49mm                              0.36                                          0.34

          92mm                              0.42                                          0.41

         135mm                              0.54                                          0.53

         Average                            0.44                                          0.43

    Standard Deviation                      0.091                                        0.096

Shared By: