Real time image processing : algorithm parallelization on multicore multithread architecture. (Imagerie temps réel : parallélisation d'algorithmes sur plate-forme multi-processeurs)
نویسنده
چکیده
....................................................................................................................................................... 3 RÉSUMÉ ............................................................................................................................................................. 4 ACKNOWLEDGEMENTS / REMERCIEMENTS ............................................................................................ 5 PUBLICATIONS ................................................................................................................................................ 6 CONTENTS ........................................................................................................................................................ 8 LIST OF FIGURES ........................................................................................................................................... 10 LIST OF TABLES ............................................................................................................................................ 11 LIST OF DEFINITIONS ................................................................................................................................... 12 LIST OF ALGORITHMS ................................................................................................................................. 13 INTRODUCTION ............................................................................................................................................. 14 1.1 Context and motivations .................................................................................................................... 14 1.2 Contributions ..................................................................................................................................... 15 1.3 Report organization ........................................................................................................................... 17 PARALLELIZATION STRATEGY ................................................................................................................. 18 2.1 Lack of common parallelization strategy for topological operators .................................................. 19 2.2 Fundamental basis for parallelization ................................................................................................ 19 2.3 Classification of SD&M strategy ...................................................................................................... 22 2.4 SD&M strategy conception ............................................................................................................... 23 2.4.1 The splitting phase ..................................................................................................................... 24 2.4.2 Distribution phase ...................................................................................................................... 27 2.4.3 The Merging phase .................................................................................................................... 34 2.5 Conclusion ......................................................................................................................................... 36 TOPOLOGICAL WATERSHED...................................................................................................................... 37 3.1 Watershed transformations ................................................................................................................ 37 3.1.1 Watershed based on flooding .................................................................................................... 39 3.1.2 Watershed based on path-cost minimization ............................................................................. 42 3.1.3 Topological watershed ............................................................................................................... 47 3.1.4 Watershed transform based on local condition .......................................................................... 50 3.1.5 Watershed transform based on minimum spanning forest......................................................... 52 3.2 Classification of watershed algorithms .............................................................................................. 54 3.3 Construction of parallel topological watershed ................................................................................. 55 3.3.1 Basic notions and definitions ..................................................................................................... 56 3.3.2 Parallel stream computing ......................................................................................................... 57
منابع مشابه
Fine-grained parallelization of similarity search between protein sequences
This report presents the implementation of a protein sequence comparison algorithm specifically designed for speeding up time consuming part on parallel hardware such as SSE instructions, multicore architectures or graphic boards. Three programs have been developed: PLAST-P, TPLAST-N and PLAST-X. They provide equivalent results compared to the NCBI BLAST family programs (BLAST-P, TBLAST-N and B...
متن کاملFrom Dataflow Specification to Multiprocessor Partitioned Time-triggered Real-time Implementation
We consider deterministic functional specifications provided by means of synchronous data-flow models with multiple modes and multiple relative periods. These specifications are extended to include a real-time characterization defining task periods, release dates, and deadlines. Task deadlines can be longer than the period to allow a faithful representation of complex end-to-end flow requiremen...
متن کاملRésolution d ’ EDP par un schéma en temps
Résumé. On propose dans cette Note un schéma permettant de profiter d’une architecture parallèle pour la discrétisation en temps d’une équation d’évolution aux dérivées partielles. Cette méthode, basée sur un schéma d’Euler, combine des résolutions grossières et des résolutions fines et indépendantes en temps en s’inspirant de ce qui est classique en espace. La parallélisation qui en résulte se...
متن کاملDesign of a Multi-Strategy Parallelization for an Entire Application of Document Categorization on Low-Cost Multiprocessor PCs Parallélisations de catégorisation de document sur multi- processeur
This paper introduces a research about parallelization of an entire application of Document-Categorization. The objective of this parallel computing research is to obtain a parallelization that can be successfully used on low cost and largely diffused shared memory multiprocessor PCs (not only on powerful and expensive supercomputers), and without any change in the input, output and user interf...
متن کاملWCET analysis of multi-level set-associative instruction caches
With the advent of increasingly complex hardware in real-time embedded systems (processors with performance enhancing features such as pipelines, cache hierarchy, multiple cores), many processors now have a set-associative L2 cache. Thus, there is a need for considering cache hierarchies when validating the temporal behavior of real-time systems, in particular when estimating tasks’ worst-case ...
متن کامل