5. Experimental results 
In the static case, we have obtained very fast and accurate 
results which are independent of the image complexity. Results 
obtained for mobile case are coarser than for the static case, but 
they allow us a real-time discrimination between camera motion 
and motion of agents in the scene, and to extract kinematic 
characteristics of rigid motions (translations and rotations) of 
the center of segmented regions. Next step, would be to obtain a 
more precise information about the kinematic behaviour of 
boundaries to have an estimation of time-of-impact and prevent 
collisions. 
6. Conclusions and future developments 
Relative to the image analysis, the spent time to process the 
sample extracted from a video sequence is considerably shorter 
than the geometric mise-in-correspondence for conventional 
approaches based on the extraction and grouping of 
minisegments. Furthermore, our real-time analysis is 
independent of the complexity of the scene and the movements 
appearing at the image. The optimal properties of Delaunay 
decompositions are transferred to the color space and provide a 
self-organizing pattern whose parameteres can be selected by 
user. By using the propagation mechanism with so many centers 
(Voronoi sites) as much as winner colors, it is possible to 
generate an easily updatable dynamic segmentation. In the next 
future, we hope to prove that such dynamic segmentation is 
optimal w.r.t anisotropic diffusion. Some another open 
questions concern to the computational management of dynamic 
3D graphs linked to Delaunay simplicial decompositions in 
order to obtain a real-time segmentation for mobile complex 
scenes without extracting directly boundaries as in the classical 
case. 
Our results relative to the tracking are not enough satisfactory, 
and it would be desirable to extract some kinematic 
characteristics about mobile data. The analogy with Kohonen's 
Self-Organized Maps for tessellation and learning is purely 
formal, and we intend to develop it to improve coarse 
localization and tracking mobile articulated objects, s.t. the 
human hand. 
References 
[Ber93] Bertin E., Parazza F., and Chassery J.M., 
"Segmentation and Measurement Based on 3D Voronoi 
Diagram: Application to Confocal Microscopy", 
Computerized Medical Imaging and Graphics, vol.17, 
pp.175-182, 1993. 
[Cel86] Celenk M., Smith S.H., " Gross Segmentation of Color 
Images of Natural Scenes for Computer Vision Systems", 
Applications of Artificial Intelligence III, pp.333-344, 1986. 
[Che98] C.H.Chen, Jian-Der Lee, Jenq-Rong Wand and 
C.W.Mao: “Color Image Segmentation for Bladder Cancer 
Diagnosis”, Mathl. Comput. Modelling 27(2), 1998, 103- 
120. 
[Ede93] H.Edelsbrunner and T.S.Tan: “An upper bound for 
conforming Delaunay triangulations”, Discrete and 
Computational Geometry 10(2), 1992, 197-213. 
[Fer92] F.Ferri and E. Vidal: “Color image segmentation and 
labelling through multidedit condensing”, Pattern 
Recognition Letters, 13, 1992, 561-568. 
[Gau99] J.M.Gauch: “Image Segmentation and Analysis via 
Multiscale Gradient Watershed Hierrchies”, IEEE Trans on 
JP, 8(1), 1999, 69-79. 
[Hea92] J.Healey: “Segmenting Images using normalized 
color”, in IEEE Trans on Systems, Man and Cybernetics 
22(1), 1992, 64-73. 
[Hun95] R.W.G. Hunt: “The reproduction of color”, 
Ed.Fountain Press, Kingston-upon-Thames, UK, 1995. 
[Law77] C. L. Lawson: “Software for C1 surface interpolatio ", 
in Mathematical Software III, J. R. Rice, ed., Academic 
Press, New York, 1977, pp. 161-194. 
[Oht80] Ohta Y.I., Kanade T., and Sakai T., ‘Color Information 
for Region Segmentation", Computer Graphics and Image 
Processing, vol.13, 1980, 222-241. 
[Per90] P.Perona and J.Malik: “Scale space and edge detection 
using anisotropic diffusion”, IEEE Trans on PAMI, 12(7), 
1990, 629-639. 
[Shi99] A.Shiji and N.Hamada: “Color Image Segmentation 
Method using Watershed Algorithm and Contour 
Information", Proc ICIP, 1999, 305-309. 
[Wu93] Z.Wu amd R.Leahy: “An optimal graph theoretic 
approach to data clustering theory and its application to 
image clustering”, IEEE Trans on PAMI 15(11), 1998, 
1101-1113. 
—154- 
KE 
Th 
sur 
La 
dec 
“le: 
Orc 
Sur 
Of 
im] 
or 
arc 
tec 
prc 
to 
by 
Se 
ter 
the 
ren 
rea 
wh 
det 
his