Abstract
This paper presents the evolution of the Associative Mesh, a massively parallel SIMD architecture based on reconfigurability and asynchronism. To favor its System on Chip implementation, we introduce a reorganization of the structure based on processors virtualization and evaluate its consequences on hardware cost and algorithmic performances. Using an evaluation environment based on a programming library and a parameterized description of the architecture, we show that a virtualized Associative Mesh achieves real-time execution for a number of complex image processing algorithms, including split and merge segmentation, watershed segmentation and motion detection.
Similar content being viewed by others
References
Batcher, K.E.: Design of a massively parallel processor. IEEE Trans. Comput. C-29(9), 836–840 (1980)
Beucher, S.: Watershed, hierarchical segmentation and waterfall algorithm. Mathematical Morphology and its Applications to Image Processing, pp. 69–76. Serra & Soille, Springer, Heidelberg (1994)
Blelloch, G.: Vector Models for Data-Parallel Computing. MIT, Cambridge (1990)
Caplier, A.: Modèle markovien de détection de mouvement dans les séquences d’images: approche spatio-temporelle et mise en oeuvre temps réel. Ph.D. Thesis, Institut National Polytechnique de Grenoble (1995)
Denoulet, J., Mérigot, A.: System on Chip evolution of a SIMD architecture for image processing. IEEE Conference on Computer Architecture for Machine Perception, New Orleans (2003)
Denoulet, J., Dulac, D.: System on chip approach for a SIMD architecture dedicated to 2D and 3D image processing. IS&T & SPIE’s Electronic Imaging, Real Time Imaging Conference, Santa Clara (2003)
Denoulet, J., Mostafaoui, G., Lacassagne, L., Mérigot, A.: Implementing Motion Markov Detection on General Purpose Processor and Associative Mesh. In: IEEE Conference on Computer Architecture for Machine Perception, Palermo, pp. 288–293 (2005)
Denoulet, J.: Architectures massivement parallèles de systèmes sur circuits (SoC) pour le traitement de flux vidéos. Ph.D. Thesis, University of Paris XI (2004)
Ducourthial, B.: Les Réseaux Associatifs, un modèle de programmation à parallélisme de données, pour algorithmes et données irrégulières, à primitives de calcul asynchrones. Ph.D. Thesis, University of Paris XI (1999)
Ducourthial, B., Mérigot, A., Sicard, N.: Anet: a programming environment for parallel image analysis. In: IEEE Conference on Computer Architecture for Machine Perception, Padova, pp. 280–289 (2000)
Dulac, D.: Contribution au parallélisme massif en analyse d’image: une architecture SIMD fondée sur la reconfigurabilité et l’asynchronisme. Ph.D. thesis, University of Paris XI (1996)
Duff, M.L.B.: Clip4, a large scale integrated circuit array parallel processor. In: International Conference on Pattern Recognition, Coronado, pp. 728–733 (1976)
Dulac, D., Guezguez, S., Bertrand, G.: Parallel segmentation based on topology with the associative nets model. In: IEEE Conference on Computer Architecture for Machine Perception, Padova, pp. 95–104 (2000)
Galilee, B., Mamalet, F., Renaudin, M., Coulon, P.Y.: Watershed parallel algorithm for asynchronous processors array. In: IEEE International Conference on Multimedia and Expo, Lausanne, pp. 793–796 (2002)
Horowitz, S., Pavlidis, T.: Picture segmentation by a tree traversal algorithm. J. ACM 23(2), 368–388 (1976)
Lalande, P.: Détection du mouvement dans les séquences d’images selon une approche markovienne, application à la robotique sous-marine. Ph.D. Thesis, University of Rennes I (1990)
Li, H., Stout, Q.: Reconfigurable Massively Parallel Computers. Prentice-Hall, Englewood Cliffs (1991)
Lohier, F., Lacassagne, L., Garda, P.: Generic programming method for real time implementation of MRF based motion detection algorithm on a multiprocessors DSP with multidimentionnal DMA. GRETSI, Vannes (1999)
Mérigot, A.: Asociative nets model: a graph based parallel computing model. IEEE Trans. Comput. 46(5), 558–571 (1997)
Mérigot, A., Zavidovique, B.: Image analysis on massively parallel computers: an architecture point of view. Int. J. Pattern Recognit. Image Anal. 6(3) (2002)
Meyer, F.: Topographic distance and watershed lines. Signal Processing 38(1), 113–125 (1994)
Mostafaoui, G., Kunlin, T., Lacassagne, L.: Détection de mouvements par relaxation Markovienne et seuillage par hystérésis dans des séquences couleurs. In: 2nd International Symposium on Image/Video Communications over fixed and mobile networks, Brest 7–9 July (2004)
Owens, R.M., Irwin, M.J., Nagendra, C., Bajwa, R.S.: Computer vision on MGAP. In: IEEE Conference on Computer Architecture for Machine Perception, New Orleans, pp. 337–341 (1993)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Denoulet, J., Mérigot, A. An architecture based on reconfigurability and asynchronism for real-time image processing. J Real-Time Image Proc 3, 119–130 (2008). https://doi.org/10.1007/s11554-007-0063-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11554-007-0063-8