An efficient watershed segmentation algorithm suitable for parallel implementation

Author

Moga, Alina ; Cramariuc, Bogdan ; Gabbouj, Moncef

Author_Institution

Signal Process. Lab., Tampere Univ. of Technol., Finland

Volume

2

fYear

1995

fDate

23-26 Oct 1995

Firstpage

101

Abstract

An important aspect of designing a parallel algorithm is exploitation of the data locality for minimization of the communication overhead. We propose a reformulation of a global image operation called the watershed transformation. The method is one of the various approaches for image segmentation and works by labeling connected components. Both serial and parallel programming models are presented and evaluated when running on SUN and DEC Alpha AXP workstations, and a Cray T3D, respectively

Keywords

image segmentation; parallel algorithms; parallel programming; Cray T3D; DEC Alpha AXP workstations; SUN workstations; communication overhead minimization; data locality; global image operation; image segmentation; parallel algorithm design; parallel implementation; parallel programming models; serial programming models; watershed segmentation algorithm; watershed transformation; Biomedical signal processing; Image segmentation; Labeling; Laboratories; Parallel programming; Rain; Signal processing algorithms; Surface morphology; Surface topography; Water resources;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing, 1995. Proceedings., International Conference on

Conference_Location

Washington, DC

Print_ISBN

0-8186-7310-9

Type

conf

DOI

10.1109/ICIP.1995.537424

Filename

537424