Vol.1 ● No.

3 ● 2012 Scientific Research Journal of India 70

Recovery of Decayed Species through Image Processing

K.Priyadharsan*, S.Saranya**

Abstract:
The problem of rebuilding a structure from decayed remains has been, until now,
especially relevant in the ambit of forensic sciences, where it is obviously oriented
toward the identification of unrecognizable corpses; but its potential interest to
archaeologists and anthropologists is not negligible. This paper is about recovering
the decayed species’ structure, through Spiral Computed Tomography data and
virtual modeling techniques (in this case with VTK software), 3-D models of the
possible physiognomy of ancient mummies. The species representation is based on
3D models and soft tissues are reconstructed.Isosurfaces generation is based on
Marching cubes algorithm. The resulting voxel models are converted into 3d
wrapped models that are coded using VTK software. The presented results iiustrate
that based on the proposed methods a complete recovery of decayed structure can be
built with less cost.

Keywords: VTK, CT,3-D

INTRODUCTION

Reconstruction is an important key feature amount of information not only about the
of image processing applications. It uses CT mummy and its skeleton, but also about the
scanning’s numbers allowed a very fine artifacts buried with the mummy and its
discrimination between materials with coffin2. Compared to traditional x-ray
different densities providing an enormous techniques, multiple axial images displayed

http://www.srji.co.cc
 Vol.1 ● No.3 ● 2012 Scientific Research Journal of India 71

in a clearer way the different details of car The anthropological study of the
tonnage, wrappings, amulets an
and internal mummified cranial remains allowed us to
3
organs of a mummy and allowed easy identify a male subject with an age at death
measurements of exact distances between of around 40 years. The skull is
objects inside or outside the mummy. In the dolichocranic, of medium height and with
last years, spiral CT has considerably rounded occiput, narrow face, high
enhanced clinical imaging. The use of this cheekbones,
bones, gracile even if well developed
new technique has fatherly widened the in its height, jaw; the orbits are narrow, the
range and quality of possible
possi investigations nose is well-shaped,
shaped, and of Europoid look.
on mummies.
So far, related work only considered initial 2. Spiral CT Scanning
representation of the fossil using CT
scanning. Soft tissue reconstruction and
texture mapping has to be studied in detail.
In my paper, surface is constructed using
Marching cubess algorithm and some
changes are made to the existing aalgorithm
to get better results. 3D models are wrapped
and coded using VTK software
This process is organized as follows. In the
next section, I describe the process of Fig. 2 CT scanning of the head

anthropological and egyptological analysis

The cranial cavity was filled with hot
of the head. Section 2 presents spiral CT of
melted resin, later solidified, introduced
the head. Section 3 presents. Reconstruction
with the mummy resting on its back, as the
of a 3-D
D model of the skull generated from
model reconstructed from the CT images
CT data processing. Section 4 presents
clearly displays.
application of textures fitting the somatic
features.
3. Reconstruction of a 3D Model of the
Skull
1. Preliminary Anthropological
thropological Results
CT slices must be stacked up and
interpolated in order to build a volume.

http://www.srji.co.cc
 Vol.1 ● No.3 ● 2012 Scientific Research Journal of India
Once created a volume, it is possible, by
means of suitable algorithms, to generate
surfaces whose points have the same
function value. They are called is surfaces A
popular algorithm for determining
surfaces is the so called marching cubes.
The principle underlying the application
applica of
this algorithm to the kind of problem here
described is that similar materials have the
same radio-opacity
opacity and are, consequently,
represented in a CT scan by the same
densitometry level. In CT slices, the
intensity associated to each pixel in the
grey-scale
scale is proportional to tissues density:
black corresponds to air, white to bones. It
is therefore possible processing the CT
scans sequence so as to obtain a 3-D
3 grid,
where to each "knot" (control point) is
associated the densitometry value measured
by the CT scans. The result is a 33-D 256
grey levels image.
Fig. 3 a) hard tissues b) external surface
4. Reconstruction off Soft Tissues
72
This stage of our work is still in a
preliminary phase. Among the possible
methodologies to deal with this complex
problem, I focused two different promising
is ways:
A. Protocols
rotocols developed to the reconstruction
of soft tissues on skull
B. Use of warping techniques.
t
A. The thickness of the soft tissues is
reconstructed on the bones through the use
of pegs at marked points. All the pegs are
joined by strips of plotline of fixed
thickness and the empty spaces among them
are then slowly filled with mould able
material: in this way, it is possible to
reconstruct nearly all the face that belonged
to the living subject; on this, nose cartilage,
eye globes and lips are added.
B. A different method consists in the
distortion (warping) of the 33-D model of a
reference
ce scanned head, until its hard tissues
match those of the mummy. The subsequent
stage is the construction of the hybrid model
composed by the hard tissues of the
mummy plus the soft ones of the reference
head.
5. Textures Fitting the
t Somatic Features
http://www.srji.co.cc
 Vol.1 ● No.3 ● 2012 Scientific Research Journal of India
While hard and soft tissues give
morphological information, textures provide
colours and aesthetical features. They are
"pasted" over the 3D models by means of
mapping procedures. Moreover, being a
frontal view, it does not give sufficient
information for the mapping of the entire
model. We get a low resolution image (fig4).
Fig 4 low resolution image
The texture was mapped onto the 3D model
to perfectly match the frontal view of the
mummy but it loses its grain as soon as we
depart from the frontal view. Much better
results could be obtained with different high
resolution views of a new subject.
Fig 5texture, processed and colored, is
mapped onto the 3-D model.
73
Fig.6Lateralview
Development of the project: soft tissue
reconstruction using VTK
After a first part of work, open problem is to
reconstruct the lacking elements of a 3D
digital model generated from CT scans
applied to a mummified cranial remains.
The aim is to obtain a perfect match among
hard tissues so that soft tissue of reference
model can be used to represent those of the
mummy with a good approximation.
Moreover a tool is developed in order to
apply to the model cylindrical textures
obtained multiple views of a well suited
individual Software implementation has
been designed using VTK.
Cylindrical textures obtained multiple views
of a well suited individual Software
implementation has been designed using
VTK.
CT scans data representing our model and
mummy should have the same placing,
orientation, dimensions and resolution. This
is generally not true especially when dealing
with data coming from different machine so
http://www.srji.co.cc
 Vol.1 ● No.3 ● 2012 Scientific Research Journal of India 74

the first step is to perform a manual

registration (Figure 3), that is a rigid
transformation, among volumes in order to
work in the same system of coordinates.
Software like AIR are also available for
automatic registration but sometimes,
especially when volumes are quite different,
they do not produce satisfactory results.
As further requirement grayscales of hard
tissues must be similar, in spite of different
methodologies of acquisition though
Fig.6 Manchester points placed over the
mummy’s tissues has been deteriorated. It is
mummy
possible to correct these differences shifting
Calculating vector displacement among
and scaling intensities using histogram
couples of corresponding points we obtain a
information.
scattered field to drive a first warp phase. A
For volume resembling, smoothing (to
feature tracking consists in determining a
remove aliasing phenomena) and surface
correspondence between sets of
generation Vtk internal facilities are used.
characteristic points pertaining to the
At this point we precede with the setup of
volumes in order to obtain a scatter motion
the Manchester pegs onto the surface of the
field with more details. It is the most
hard tissues of the mummy while for the
important step.
reference model it can be predetermined.
Initially this set of points is chosen as a
The aim of this phase is to fix some
subset of points that are vertices of hard
constraints for the resulting physiognomy
tissues surface of the mummy; some of
and to provide a first guess for the following
these points, corresponding position in the
step that is the features tracking.
reference volume. If the result is good, the
Pegs are mapped onto a spherical surface of
resulting motion field is defined among
parametric ratio, so that the user can place
subsets of bone surfaces, with particular
quickly the whole set and the adjust single
characteristics, are identified as features.
pegs.
If, consecutively a test, a feature is retained
reliable, we search the from the reference

http://www.srji.co.cc
 Vol.1 ● No.3 ● 2012 Scientific Research Journal of India 75

model to the mummy volume..….

Once generated a scattered motion field, it
must be diffused within the whole reference
volume. Diffused motion field can be used
to warp
arp every structure pertaining to Fig .9 Model skin (blue) and mummy skull
reference model coherently with mummy (white)
model therefore we reconstruct mummy soft
tissues warping those of reference model

Fig 10 Face generated

We consider the relation between hard

tissues surface of the reference model and
Fig 7 wrap driven by manchester points hard tissue surface of the mummy as a
continuous deformation in the time.

If is the intensity of a point

of coordinates (x,y,z) at time t in the
mummy volume
and

is the motion
m field,

where ,

e are components in x, y e z
directions of velocity vector, we suppose
that the intensity function is the same at the

time in the
Fig.8 Model skull (blue) after this stage
overlapped with mummy skull (white) point of the

http://www.srji.co.cc
 Vol.1 ● No.3 ● 2012 Scientific Research Journal of India 76

reference model, we can write the 3 as

where , (5)
e and . Known as motion field constraint equation,
(1) where Ex, Ey, Ez ed Et are partial
derivatives.
We say that x is a reliable feature if
If the intensity function change smoothly
sm
(6)
with x, y, z e t, we can manipulate the
equation (1) with Taylor’s series to obtain
(2)

Where:
I( , t) is the matrix of intensity function

where e contains terms in x, y, z e t E in the point =(x,y,z) in the region W(x)
higher than first order. at the time t;
 is the gradient operator;
Eliminating , rationing by t,  min ()) represents the smaller eigenvalue
and calculating limit for , we of matrix  ;
obtain
are predetermined thresholds.
(3)
We consider a window (q) centered in q
of dimensions.
We represent (6) in discrete fashion
that is the totally derivative of in the
(7)
time.

(4)
Using abbreviated notation: The solution of (4) respect to V is given
by In this moment this stage is still in
developing so we have no picture, anyway
the idea is simple: for each of the
Manchester points we
w find its corresponding
http://www.srji.co.cc
 Vol.1 ● No.3 ● 2012 Scientific Research Journal of India 77

on the skin surface, in this way we can amount of data, was to process and visualize
measure the actual soft tissue thickness. By in real time and in 3D the data volume.
consulting the thickness table we find the Through this paper I hope that this method
corresponding desired thickness measure. will be a useful one to the society.
Saying that the actual thickness must
become the desired thickness we generate REFERENCES
another scattered field. 1. S.B. Kang, R. Szeliski, and P.
Anandan, “The Geometry-Image
Texture Application Representation Trade off for
Rendering”, Proc. ICIP, Vancouver,
Canada, September 2000.

2. P. Eisert, E. Steinbach, and B. Girod,

“Multi-hypothesis, Volumetric
Reconstruction of 3-D Objects. Proc.
ICASSP, pp. 3509-3512, Phoenix,
Mar. 1999.

3. W. E. Lorensen, and H. E. Cline,

“Marching Cubes: A high resolution
CONCLUSION: 3D surface reconstruction algorithm,”
In this paper, for obtaining better Proc. SIGGRAPH, vol. 21, no. 4, pp
performances through the virtual 3D 163-169, 1987.
visualization of the reconstruction i have
used the powerful workstation Onyx2 4. P. Debevec, C. Taylor, and J. Malik,
equipped with an architecture of type “Modeling and rendering
multiprocessor, with 4 processors R10K, 1 architecture from photographs: A
Gbyte of RAM, computing power of 1.5 hybrid geometry image based
Gflop, 1 graphic pipeline, that it can process approach,” Proc. SIGGRAPH, pp.
11 millions of polygons per second. In fact 11-20, 1996.
the main problem, processing a large

http://www.srji.co.cc
 Vol.1 ● No.3 ● 2012 Scientific Research Journal of India 78

CORRESPONDENCE
*DR-DO Project Assistant, Centre for R&D, PRIST University, Thanjavur, India. Email:
kvpriyadharshan@gmail.com. **Lecturer, Department of Comp Science & Engg, Bharadhidasan University,
Trichy, India. Email: saranyapristina@gmail.com

http://www.srji.co.cc