Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Inspection of High Magnification Fracture Surfaces using 3D from Stereo Images of Large Chamber SEM
 

Summary: Inspection of High Magnification Fracture Surfaces using 3D from Stereo
Images of Large Chamber SEM
S. Huq*, B. Abidi*, D. Page*, and M. Abidi*
J. Frafjord** and S. Dekanich**
*Department of ECE, University of Tennessee, 1508 Middle Drive, Knoxville, TN 37996
**Y12 National Security Complex, Bear Creek Rd, Oak Ridge, TN 37831
In this paper, we show 3D reconstructions of metal surface fractures obtained from affine stereo
images. The images are captured with Large Chamber Scanning Electron Microscope (LC-SEM),
matched with energy minimizing algorithm, and then reconstructed into 3D. A virtual 3D obtained
from the stereo images shows the validity of our reconstruction. Furthermore, spectral information
at two energy levels, one for AlK and the other for NbL, is extracted using Energy Dispersive
Spectrometer (EDS) and overlaid on the 3D surface. Thus, in addition to performing various
measurements on the 3D surface one is able to visually inspect the distributions of constituent
materials and correlate them with the surface structure caused by fracture or other impacts.
The LC-SEM image capture is performed in a controlled fashion. The tilt angle between view
points and the physical dimensions of the pixels of the images are known. The high magnification
of the images allows the use of affine transformation. The principal points of both images are
chosen to appear approximately at the centers of the images. Imaging in such a restricted and
known set-up enables us to reconstruct the scene in the Euclidian space without calibration of the
microscope (Fig. 1(a)) [1]. First, we rectify the images for matching. Since they are captured at

  

Source: Abidi, Mongi A. - Department of Electrical and Computer Engineering, University of Tennessee

 

Collections: Computer Technologies and Information Sciences