Faculty of Informatics Vienna University of Technology Institute of Computer Aided Automation PRIP Home PRIP Home
Personal tools
You are here: Home Teaching Informatikpraktika Riverbed Grain Size Distribution Estimation from Photographic Images

Riverbed Grain Size Distribution Estimation from Photographic Images

Praktikumsendbericht

What
  • Presentation
When Oct 23, 2007
from 05:00 pm to 05:30 pm
Where Seminarraum 183-2
Contact Name Allan Hanbury
Contact Email
Add event to calendar vCal
iCal
AutorIn:  
Miha Drenik

Betreuer:
Allan Hanbury

Abstract:
River clogging affects aquifier water level causing deterioration of the
drinking water supply, foundation dehydration or flooding. Observation
of the clogging process helps predict and prevent adverse effects.
Grain size estimation from photographic images is a cost attractive
method for the observation of the clogging process where river bottom
images of known scale are analyzed by an operator marking the extents of
individual stones. As a typical image contains over 100 stones, an
automatic method is desired to decrease the time required for analysis.

This report proposes two such methods. The first method performs a
morphological granulometry on the image under the assumption that space
between grains is darker than the grains themselves. The resulting
histogram is treated as being equivalent to actual grain size
distribution. The second method performs a watershed transformation on
the image after enhancing grain edges with an entropy filter. The
containing ellipses of the watershed basins are treated as stones.
Results of both methods are compared to those obtained manually.
Document Actions