ICAT was developed in an effort to create a general purpose environment for cluster analysis.  Within the realm of image analysis this is known as the identification of connected components.  The thesis which resulted from this work can be obtained from the links below:
 

• MS Thesis -- The Design and Application of ICAT: Interactive Cluster Analysis Tool
  Dax M. Westerman, Department of Computer Science, University of Tennessee, May 1998.
• Manual.txt -- User Manual for C++ side of ICAT
• Cluster Form Analysis Techniques for Diabetic Retinopathy, M. W. Berry and D. M. Westerman, Mathematical Models in Medical and Health Sciences, M.A. Horn, G. Simonett, and G. Webb (Eds.), Vanderbilt University Press, (1998).

Berry, M., E. Comiskey, and K. Minser, Parallel Analysis of
Clusters in Landscape Ecology, IEEE Computational Science
and Engineering 1:2 (1994), 24-28.
 
J. M. Constantin, M. W. Berry, and B. T. Vander Zanden,
Parallelization of the Hoshen-Kopelman Algorithm Using a
Finite State Machine, Int. J. of Supercomputer Applications
and High Performance Computing 11:1 (1997), 31-45.
 
J. Hoshen, M. W. Berry, and K. S. Minser, Percolation and
cluster structure parameters.  II.  The Enhanced Hoshen-Kopelman
algorithm, Physical Review E 56:2 (1997), 1455-1460.