Summary: Parallel Image Component Labeling
with Watershed Transformation
Alina N. Moga and Moncef Gabbouj
Signal Processing Laboratory
Tampere University of Technology
P.O. Box 553, FIN­33101 Tampere, Finland
Published in IEEE Transactions on Pattern Analysis and Machine Intelligence, May 1997, Vol. 19, No. 5

Abstract
The parallel watershed transformation used in grayscale image seg­
mentation is reconsidered in this paper on the basis of the com­
ponent labeling problem. The main idea is to break the sequen­
tiality of the watershed transformation and to correctly delimit the
extent of all connected components locally, on each processor, si­
multaneously. The internal fragmentation of the catchment basins,
due to domain decomposition, into smaller subcomponents is ulteri­
orly solved by employing a global connected components operator.
Therefore, in a pyramidal structure of master---slave processors,
internal contours of adjacent subcomponents within the same com­
ponent are hierarchically removed. Global final connected areas are

Source: Albert-Ludwigs-Universität Freiburg, Institut für Informatik,, Lehrstuhls für Mustererkennung und Bildverarbeitung

Collections: Computer Technologies and Information Sciences