1 System 
especially 
‘he vision 
urce with 
ses. This 
ze triplets 
alibration 
package, 
ee image 
DSM. To 
s used to 
and real 
could be 
n of the 
it of the 
percent 
s and 12 
to be at 
address 
healing 
f wound 
practical 
'€ on an 
d heels. 
are age, 
erlands, 
sappear 
erficial 
iscle or 
Boersma, Saskia M. 
  
  
  
  
  
Stage 1 Stage 2 Stage 3 Stage 4 
  
  
Figure 1: Different stages of pressure sores (Van Dijk, 2000) 
1.3 Objective 
This research focuses on the possibilities to measure the metrical features of pressure sores using a three-camera vision 
system. The central question in this research is 
Which accuracy can be achieved in the measurement of the metrical features of pressure sores with a three-camera 
vision system? 
The metrical features that have to be determined in order to compare different curing methods in an objective way are 
the volume and eventually related quantities such as the contour and surface of the wounds. 
1.4 Related research 
Traditional methods to measure pressure sores are based on determination of the volume by instilling fluid materials 
into the ulcer. Such methods have many disadvantages such as pain, irritation, allergic reaction, and inexactness. Single 
photographs only provide a limited amount of information concerning the volume and therefore do not reveal the 
beginning of a healing process in the bottom of an ulcer (Eriksson, 1979). Single photographs or video images are also 
used to measure wounds (Ring, 1985). These techniques are unable to supply 3D information and are limited to 
relatively small surfaces. Ng et al (1994) used laser to measure wounds, based on measurement of the distance of the 
laser array. Plassman et al (1995) developed a method to measure the volume of wounds with structured light. Stereo- 
photogrammetry may also provide 3D information. Bulstrode et a/ (1986) and Eriksson et a/ (1979 and 1984) use 
analogue stereophotogrammetry to determine the volume of wounds. The disadvantage of this method is the large 
instruments. Furthermore, photographs need to be developed before further processing and thus real-time results are not 
possible. De Jong (1996) proved that digital stereophotogrammetry is a good method to acquire 3D information on 
wounds. She used only one camera and thus the two images of the stereopair were not made simultaneously. The 
approach resulted in a limited precision. 
2 SYSTEM CONFIGURATION 
The configuration of the vision system was designed taking into account the type of camera, the objects to be measured, 
the illumination, the targets, the flexibility, and the costs of the system. 
2.1 Premises 
For this research we used three Sony XC-55 progressive scan video cameras. The cameras are connected to a frame 
grabber in a personal computer. This gives the possibility to view the three images during image acquisition. The focal 
length of the lens of the cameras is approximately 6 mm. The image size is 640 x 480 pixels, with a pixel size of 7.4 x 
7.4 um. 
The objects to be measured are pressure sores at different parts of the body. The images should cover the wounds 
completely, together with a part of the surrounding healthy skin, because the volume of the wound will be determined 
using interpolation from this surrounding healthy skin. 
The quality of the illumination is important for the accuracy of the measured points. The illumination should be uniform 
and illuminate the object as equal as possible. Since wounds have a moist surface, specular reflections should be 
avoided as much as possible. It is not necessary to use a very strong illumination, because the cameras are very sensitive 
to light. However, there should be sufficient light to allow a good depth of field. In case of lack of texture of the human 
skin, additional texture projection has to be used. Lack of texture causes difficulties in the matching process during 
DSM generation. 
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B5. Amsterdam 2000. 85