Title
Watershed parallel algorithm for asynchronous processors array
Abstract
A joint algorithm-architecture analysis leads to a new version of picture segmentation system adapted to multimedia mobile terminal constraints. The asynchronous processors network, with a granularity level of one processor per pixel, based on data flow model, takes less than 10 μs to segment a SQCIF $88*72 pixels - image (about 2000 times faster than the classical sequential watershed algorithms). The main originality of the proposed algorithm is only one global synchronization point is needed in order to complete the segmentation transformation, instead of the three (or more) classical points: minima detection, labelization and flooding. Our system tends to cope with multimedia mobile phones constraints, i.e. real time computing circuit, low power. We have simulated and validated this system thanks to "SystemC" library; VHDL synchronous prototyping shows up results accordingly.
Year
DOI
Venue
2002
10.1109/ICME.2002.1035901
Multimedia and Expo, 2002. ICME '02. Proceedings. 2002 IEEE International Conference  
Keywords
Field
DocType
hardware description languages,image segmentation,mobile handsets,multimedia communication,parallel algorithms,parallel architectures,SystemC library,VHDL synchronous prototyping,asynchronous processors array,asynchronous processors network,flooding,global synchronization,granularity level,joint algorithm-architecture analysis,labelling,minima detection,multimedia mobile phones,multimedia mobile terminal,parallel hill-climbing algorithm,picture segmentation system,pixels,real time computing circuit,segmentation transformation,watershed parallel algorithm
Computer science,Image segmentation,Real-time computing,Artificial intelligence,Hardware description language,Computer vision,Asynchronous communication,Synchronization,Parallel algorithm,Parallel computing,SystemC,Pixel,VHDL
Conference
Volume
Citations 
PageRank 
1
3
0.55
References 
Authors
4
4
Name
Order
Citations
PageRank
Bruno Galilée130.55
Franck Mamalet230216.35
Marc Renaudin349849.15
Pierre-Yves Coulon4826.24