We present a multiagent computer segmentation system, named KISS. This
system has been implemented under our multiagent problem solver (MAPS
), a generic programming environment dedicated to multiagent architect
ure design. MAPS entails a basic distinction between object-oriented a
nd action-oriented agents, which allows flexible alternation between f
igurative and operative focusing tasks. KISS demonstrates the applicab
ility of a multiagent approach to computer vision by offering a clear
distribution of knowledge among several agents, dedicated successively
to low-, intermediate-, and high-level analysis steps. Segmentation i
s approached through a cooperative analysis, involving both region and
contour-based detection. Interpretation of patterns is made under thr
ee successive steps, using geometrical, relational, and semantic label
ing, respectively. Such interpretation makes it possible to guide the
selection of handling procedures to improve the initial segmentation.
The potential power of such an approach is exemplified by its applicat
ion to muscle fiber analysis.