International Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 5. Hakodate 1998
A DIGITAL IMAGING SYSTEM FOR THE PRECISE 3D MEASUREMENT OF SURFACE DISPLACEMENT IN
GEOTECHNICAL CENTRIFUGE MODELS
S. Robson,
Department of Geomatic Engineering
University College London, WC1E 6BT
Telephone: +44 171 504 2740
Email: S.Robson @ ge.ucl.ac.uk
M.A.R. Cooper and R.N. Taylor
Department of Civil Engineering, City University
London, EC1V OHB
Telephone: + 44 171 477 8967
Email: R.N.Taylor@city.ac.uk
Commission V, Working Group IC WG V/III
KEY WORDS: Image Sequences, Geotechnical Engineering, Deformation Monitoring.
ABSTRACT
The application of digital imaging to the two dimensional measurement of deformations in soil models undergoing
experimentation in a geotechnical centrifuge is increasing. Typically a single camera is used to image, through a window,
targets located in the side of a soil sample. Digital image measurement and analysis techniques of varying sophistication
and geometric fidelity are then used to compute displacement information from the sequence of images. A number of
discrete displacement transducers are also used to provide information concerning changes in shape of the soil surface
during the experiment. This paper describes a new complementary system incorporating multiple CCD cameras that can
be used to measure many hundreds of 3D locations on the upper surface of the soil. The paper focuses on the imaging
system, calibration procedures and 3D target co-ordination and registration algorithms necessary to compute reliable
surface information in the harsh centrifuge environment.
1. INTRODUCTION
In order to understand the detailed behaviour of
geotechnical events and processes it is important to be
able to observe how soils respond to load. Single element
testing apparatus can be used to investigate the stress-
strain behaviour of soil when subjected to particular stress
paths. However, the response of geotechnical structures
is the integrated effect of a large number of soil elements
each following their own particular stress path. It is
therefore of major importance to be able to measure
displacements and hence strains during real geotechnical
events. Instrumentation of prototype structures can yield
valuable results, but much more can be learned from
comprehensive test series on small-scale geotechnical
models.
The behaviour of geotechnical structures can be studied
using physical models, the main requirement being to
create in the model stress profiles corresponding to those
in the prototype. This can be achieved by accelerating
small-scale (1:n) models to n times earth’s gravity using a
geotechnical centrifuge. Thus a 10 m layer of soil can be
represented by a 10 cm deep model of the same soil
accelerated to 100g because the reality and the model will
then experience the same self weight stresses at
homologous points.
Centrifuge testing allows the study of geotechnical
processes in scaled models with properly established
scaling laws relating the model to the corresponding
prototype. Particularly valuable are measured movements
in vertical sections of plane models that can be observed
through a perspex window in the sidewall of a model
container. These subsurface deformations can be
compared directly with those from finite element
predictions and can be used to test and improve
constitutive models of soil behaviour.
783
In order to monitor such movements, the technique
commonly adopted is to place markers or targets in the
soil face that is in contact with the window. A single CCD
camera vision metrology system allows these targets to
be viewed during centrifuge flight (Figure 1). Thus, by
measuring the position of these targets in the resultant
sequence of calibrated digital images, displacements in
the model can be determined. Such measurements of 2D
soil movements are accepted as an appropriate technique
(Allersma, 1991; Ethrog, 1994). A model width is typically
of the order of 500 mm which in an experiment at 100 g
represents a prototype distance of 50 m. The most useful
measurements of displacement will need to have an
accuracy of 0.01 - 0.1 mm. i.e. 1 - 10 mm. prototype
scale.
Whilst measurements in image space are straightforward,
utilising established circular target recognition techniques
such as dynamic thresholding and subsequent centroiding
(Shortis et al 1995), their transformation into object space
defined by the plane of the soil is undertaken using a
variety of techniques. Methods range from precise opto-
mechanical alignment of the camera and soil plane,
through deterministic mathematical transformations, to
complete photogrammetric solutions employing camera
calibration, dynamic computation of camera location, and
a refractive model to account for the optical effects of the
window between the camera and soil. Measurement
precisions, in the soil plane, of between 0.05 and 0.08 mm
are typically achieved at City University (Taylor et al,
1998) using a mathematical model based on established
photogrammetric procedures.
It has become apparent during such experiments that
some means of determining to what extent the
measurements made in the soil plane at the window
surface are representative of the movements throughout
the depth of the soil. One means of at least partial
verification is to make measurements of the soil surface
